Article body

1. Introduction

The growing complexity of the world’s countries, confronted with the transnational migration resulting from globalization, includes the challenge of organizing the interactions between the members of their populations (Brysk 2002). The health crisis resulting from the COVID-19 pandemic calls for a resolute reorganization of different forms of interaction, particularly in health and other public services where it is possible to intervene remotely, to prevent the spread of the virus (Greenhalgh, Koh, et al. 2020; Greenhalgh, Wherton, et al. 2020; Thornton 2020). A population segment that is particularly at risk are ethno-cultural communities (with diverse values, beliefs, behaviours and language needs), for whom healthcare is often less effective at all levels: promotion, prevention, early diagnosis, treatment, and rehabilitation (Epstein 2008).

Public service interpreting (PSI), a field of study and practice that has been expanding in an increasing number of countries since 1980, aims to improve universal accessibility to public services regardless of the language(s) spoken by the user (Pöchhacker 2016). Today, it is generally acknowledged that PSI reduces the language barriers that affect migrant populations (Crezee 2013; Flores 2005; Karliner, Jacobs, et al. 2007).

These barriers continue to be one of the main causes of inequality in healthcare, particularly in Canada (Ahmed, Shommu, et al. 2016). Employees in public services are often unfamiliar with the procedures for obtaining the services of an interpreter, and may even be unaware that such services exist, at least for spoken languages (Brisset, Leanza, et al. 2014). Even when an interpreted consultation actually takes place, there are several challenges involved, related especially to the working alliance between the practitioner, the interpreter and the user (Brisset, Leanza, et al. 2013).

In Quebec, PSI services are essentially provided and managed by banks of interpreters, whether they are part of the Health and Social Services System, such as the Banque d’interprètes de la Capitale Nationale[1] (BICN) and the Banque interrégionale d’interprètes[2] (BII) of Montréal, or run by community organizations, such as the Service d’aide aux Néo-Canadiens[3] (SANC) in Sherbrooke. These banks also have many challenges to deal with in recruiting, training and retaining interpreters (Ozolins 2010) and in their role as intermediaries between interpreters and practitioners (Ozolins 2007).

Institutional administration faces challenges too. The symbolic support of the administration (Novak-Zezula, Schulze, et al. 2005) and the identification of organization facilitators (Karliner and Mutha 2009) are essential when implementing interpreting services. Karliner and Mutha talk about three types of facilitators. The first is the coordinator, who is in charge of ensuring the fluidity of services (e.g., making sure that audio-visual equipment is ready prior to the consultation) and liaising between the administration, practitioners and interpreters, while also assessing the process in order to propose any necessary adjustments. Next are the champions, employees who are targeted for their influential role in the institution and their interest in interpretation services. Their job is to encourage their colleagues to use interpreting services. Finally, there is the supervisor, an administrator who acts as spokesperson for the interpreters within the institution to ensure they are fully integrated.

1.1 Remote interpreting

In its simplest form, remote interpreting (RI) refers to interpreters using telecommunications technology—telephone (telephone-based remote interpreting, T-RI) or videoconference (videoconference-based remote interpreting, V-RI)—to offer their services to users and practitioners located elsewhere (Braun 2015). Improving the availability of trained interpreters, providing services in a wide variety of languages and reducing costs (for travel, for example) are all incentives in favour of RI (Skinner, Napier, et al. 2018). Preventing the spread of COVID-19 and intervening to help a particularly at-risk population are other incentives that add to this list since communication barriers have a demonstrable effect on the effectiveness of health promotion messages (Hommes, Borash, et al. 2018). Thus, RI is seen as a contributing factor to disease prevention, reducing the risk to public health that communication barriers may present for people with infectious diseases (such as tuberculosis, AIDS, etc.) (Burdeus-Domingo 2015).

Advances in telecommunications and data transfer make V-RI a promising option (Braun 2015), especially as a growing number of studies are confirming the importance in PSI of non-verbal communication (for instance, Skinner, Napier, et al. 2018), relational proximity and the working alliance (for instance, Becher and Wieling 2015; Gartley and Due 2017; René de Cotret, Brisset, et al. in press; Resera, Tribe, et al. 2015). In spite of these advantages, however, RI adds “a further layer of complexity” to PSI that needs to be examined in greater detail (Skinner, Napier, et al. 2018: 2).

This includes notably the concept of virtual presence, introduced by Mouzourakis (1996; 2006) in reference to the physical and psychological consequences for all interlocutors, and the feeling of alienation experienced by the interpreter in particular. Interpreters are required to find ways to deal with the perceived distance between themselves and the interpreted event and to compensate for information lost in the virtual exchange. In connection with the feeling of alienation, Braun (2012: 314) talks about a “reduced social presence” that translates into a less natural manner of speaking: a tendency to speak louder, to over-elaborate, and to be less coherent. Moser-Mercer (2003; 2005) documents the more rapid decline in interpreters’ performance due to fatigue caused by the virtual presence: the peak in interpreter fatigue occurs twice as rapidly (15 to 18 minutes) than the conventionally recommended work intervals of 30 minutes. This decline is even more pronounced given that interpreters work “at the limit of cognitive saturation” (Gile 2005: 724), even in a laboratory setting in the same room as the other participants where stressors are controlled. The allocation of a greater portion of cognitive resources to comprehension appears to deprive other processes—language production, in particular—of the necessary resources to maintain an optimal performance level throughout the standard duration of the work interval (Moser-Mercer 2005).

Furthermore, Braun and Taylor (2012) note that V-RI can lead to several errors, to the extent where they recommend that this modality be used with caution. Errors related to coordinating speaking in turns are the most common (false starts, overlapping speech). Errors also tend to peak after 25 minutes, which corroborates the link described above between RI and increased fatigability. Braun and Taylor also observed that videoconference interviews, which are longer on average, hamper the non-verbal communication that plays such a crucial role in the interpreter’s work. For example, technical constraints (regardless of the quality of the cameras, screens and image) make it virtually impossible to catch the eye of a participant to see if they have understood what has been said. This interrupts the flow of the exchange and impedes mutual comprehension.

Finally, access to V-RI would not necessarily be sufficient for practitioners to use this modality. Although the 99 Danish head nurses surveyed by Mottelson, Sodemann, et al. (2018) had unlimited access to V-RI, two-thirds of them said they still occasionally used face-to-face interpretation, sometimes even using untrained interpreters (family members or hospital employees)—an option that the literature has long shown to be a source of serious misunderstandings and errors.[4] This result is particularly surprising in light of the fact that V-RI has been widely available and encouraged in Danish hospitals since 2011. Reasons given by participants were primarily related to the diminished social presence of the interpreter during videoconferencing and the technical challenges involved.

It should be noted that the operational distinctions between V-RI and T-RI are beyond the scope of our research. That said, an overview of these distinctions is provided in the report we produced for the Government of Quebec (Leanza, René de Cotret, et al. 2019). Also, the Guide (see Appendices) includes a few recommendations regarding the technologies underlying these modalities; we suggest the Handbook of Remote Interpreting (Amato, Spinolo, et al. 2018) for more information on this subject.

1.2 Objective and questions

Considering the challenges posed by the implementation and practice of remote public service interpreting, our research proposes to identify ways to provide a framework for and to ensure the successful interpretation of a public service interview by videoconference (primary focus) or telephone (secondary focus). More specifically, our research objective was to produce a list of recommendations for institutions and organizations from Quebec interested in developing V-RI services or improving their supervision and support of existing services. The research questions that arose from this objective were as follows:

1. Do the recommendations in the literature regarding the planning and supervision of RI services reflect the actual experiences of interpreters, practitioners and administrative personnel in the Quebec public sector?

2. If not, what recommendations should be added to the existing ones or how should the existing ones be modified to do so?

2. Method

The study stems from a mandate by the Government of Quebec to investigate the challenges and solutions related to the implementation of RI (V-RI in particular) in the Quebec Health and Social Services System, where videoconferencing for this purpose was marginal at the time, at least prior to the COVID-19 pandemic, as implementation is now much faster[5]. We therefore did not aim to interview experts on this modality but rather the social actors who would have to deal with this modality if it were implemented. We therefore drew on their expertise to adapt the recommendations in the scientific literature to the Quebec context and to supplement them. The methodology of the literature review and its results (a list of 32 existing recommendations, based mainly on Braun 2012) are provided in Leanza, René de Cotret, et al. (2019).

2.1 Data collection

Data was collected during focus groups (n=4) and individual interviews (n=6) conducted between January and March 2019. Focus groups allow researchers to facilitate structured discussions on a specific topic and encourage interactions resulting from differences in participants’ opinions and experiences (Geoffrion 2003). The number of focus groups conducted corresponded to the number needed to reach data saturation (Morgan 1996). Individual interviews were also arranged to accommodate participants who were unable to be part of a group. The focus groups lasted between 75 and 90 minutes, which is in line with Morgan’s (1988) suggestions. The six individual interviews lasted between 45 and 70 minutes, which is also the norm (Royer, Baribeau, et al. 2009).

The sites chosen were three major urban centres (Quebec City, Trois-Rivières and Sherbrooke) and two peripheral cities (Rimouski and Drummondville). The participating interpreters were assigned to groups that were separate from practitioners and administrators in order to allow them to talk freely about their working experiences, something they might have been reluctant to do if administrators (i.e., their employers) had been present.

The focus groups and individual interviews were facilitated by the first author and transcribed verbatim by the second author (who was also present at the focus groups) or by research volunteers. The interviews were semi-structured, with participants being encouraged to comment on the recommendations, drawn from the literature, that were presented to them. A copy of the summary document was given to each participant prior to the interview, and it was also displayed on-screen during the actual interview. Discussions with interpreter and practitioner participants focused mainly on the practice of RI, while administrator participants discussed the planning aspect.

To compensate for their time and travel expenses, the self-employed interpreters were paid $75. The salaried practitioners and administrators received $25.

2.2. Participants

Participants (n=27) were recruited based on a single criterion: to have work experience in the RI of interviews. This experience could be direct, as an interpreter or practitioner involved in an interpreted encounter, or indirect, as an administrator in a welcome centre for migrants or in an organization that uses the services of interpreters. Participants were recruited through the intermediary of resource people from the SANC in Sherbrooke, the BICN in Quebec City, the Service d’accueil des nouveaux arrivants[6] (Trois-Rivières) and the Regroupement interculturel de Drummondville[7].

Of the 27 participants, 18 were working in Sherbrooke, four in Quebec City, two in Rimouski, two in Drummondville and one in Trois-Rivières. Participants’ average age was 44 years (spread=24-67) and 22 of the 27 were women.

The 15 interpreter-participants were self-employed workers with an average of six years’ experience (spread=1-19) in various public service contexts (e.g., welcoming refugees, healthcare, youth protection, judicial, education) in a wide variety of languages (English, Swahili, Dari, Lingala, Arabic, Tshiluba, Persian, Spanish, Pashto and Sango). The nine practitioner-participants—five nurses, two physicians, one social worker and one psychoeducator—all worked in the health and social services sector and had been using RI services for an average of five years (spread=1-12). The three administrator-participants had worked in interpretation services for an average of three years (spread=1-7).

Only five of the participants, all interpreters, reported having direct experience with V-RI, and only once a month or less, on average. The 23 participants who had direct experience with T-RI generally used that method between once and a few times a month, with the exception of one participant who used it every day. The three administrator-participants had no direct experience with RI.

In the Results section, participants’ remarks have been reported verbatim (translated from French into English). In order to identify their main characteristics without compromising their anonymity, they are all identified by a code: In for Interpreter, Pr for Practitioner and Ad for Administrator. The first letter of the participant’s gender (F or M) as well as their age (rounded off to the nearest decade) is given in parentheses after the code. For example, Pr4(F,40) is practitioner #4, a woman in her forties.

2.3. Data analysis

A thematic analysis (Paillé and Mucchielli 2008) of the participants’ discourse was done by both the first and second authors, in collaboration with the third author, using QDA Miner software. Response saturation was obtained with four themes.

3. Results

Our analysis of the participants’ discourse highlighted four themes: 1) telecommunications technology, 2) the diversity of practice contexts, 3) barriers to communication between key actors and 4) supervision and support of the virtual presence.

3.1. Telecommunications technologies

This theme includes issues related to the telecommunications technologies required for RI. Although telecommunications technologies are at the heart of RI, the key actors questioned had very little to say on this subject.

When questioned in greater detail on certain recommendations drawn from the literature, a few participants (e.g., In1(M,40) and In2(F,30)) mentioned that they were sometimes bothered in their practice by an echo effect (i.e., they could hear themselves) and a certain amount of transmission delay. With respect to T-RI, participants agreed that teleconferencing was a relatively simple process and that a land line produced a better signal than a cell phone. No problems were reported in connection with V-RI.

3.2. Diversity of practice contexts

When commenting on recommendations from the literature and generally sharing their experience with RI, participants always talked about the many contexts in which they practised their profession. It appears that these contexts have an influence on the planning as well as the practice of RI, and that they must be examined and discussed to ensure the quality of the services provided.

For example, is the interview taking place in two different places (with the practitioner and the user at the hospital and the interpreter somewhere else, or in a courtroom situation, with the interpreter and the accused in one room and the victim in another, etc.)? Or even three, as it might now be the norm in the pandemic context? In4(M,30), In5(F,20) and In9(F,30) explained that the three people involved are generally in three separate locations during their RI sessions, which complicates the interview even further, particularly in terms of determining whose turn it is to speak and confidentiality issues. Both of these points are discussed later in this article.

The purpose of the interview is also a determining factor. In3(M,40) explained that RI is an effective way to transmit technical messages or ask very specific questions, but that in other contexts (such as mental health, for example), he sometimes had the impression that the interview was cut short:

The last time I opened a file for someone in person, the patient was in such distress and crying so hard that she had to be sent to hospital right away. That made me think of several times when I had done the same thing over the phone and had thought that if we had been face-to-face, the result might have been a different, more immediate intervention.

This participant added that it wasn’t always easy to deal with the emotions that arise during interpreted interviews involving mental health issues, which are much more delicate to manage from a distance. The fact that the interview may be cut short can also affect the interpreter, left on their own after what they have just experienced without having a chance to compose themselves in the way they can when an interview is conducted and interpreted in person.

Apart from situations involving mental health issues, certain other medical contexts are particularly stressful for interpreters. Ad3(F,30) explained that “it’s difficult to find interpreters for paediatric chemotherapy—women interpreters drop like flies.” Pr1(F,50) responded that interpreters should have the right to choose the areas in which they want to work, just as some doctors and nurses refuse to practise in chemotherapy.

How do interpreters perceive their responsibility? When interpreters were questioned on the subject of V-RI, they said the legal/judicial context, particularly the courtroom setting, had a much greater impact on the quality of their interpreting than videoconferencing. In9(F,30) was particularly eloquent on this subject:

In the courtroom, there are a lot of emotions and truths that clash. We’re right there in the middle of it. We can’t neglect a single word; every detail is important. Everyone knows everyone else so well that sometimes, if I miss a word or get it wrong, someone will correct it. Then I get really confused because I can’t listen to the person who’s speaking and the person beside me at the same time… it can get pretty complicated to manage. […] I think that’s what stresses us the most: the weight of justice.

What language is interpreted? In8(F,60) reported that having interpreters available for Anglophone users can serve several purposes. Functionally bilingual practitioners may request an interpreter so that the interview takes less time or even to make up for a lack of time even if this practice is not recognized by the profession’s codes of ethics (e.g., National Standard Guide for Community Interpreting Services[8]). Following a vaccination, for example, an interpreter may be asked to stay with the child for fifteen minutes or so to watch for any allergic reaction.

3.3. Barriers to communication between key actors

One-on-one interviews allowed us to pinpoint some significant communication problems between the key actors in PSI. Barriers to communication refer to all aspects of communication related to the interpreted interview, either directly—from the initial request for services to the debriefing—or indirectly, from overall service planning to interpreter training and supervision—all aspects that involve a transfer of information.

Results for communication barriers are divided into two sub-themes based on whether they are related to procedures involved in planning RI services or to the actual human resources (practitioners and interpreters).

3.3.1. Procedure: Clarify what constitutes an emergency service request

It is recommended that parameters be established for emergency service requests to ensure the quality of services provided and interpreters’ well-being. Ad2(F,30) stressed the importance of prioritizing the urgency of service requests more efficiently in order to better meet users’ actual needs: “The problem doesn’t have to do with processing urgent requests. The problem occurs when people make emergency requests when no actual emergency exists.” Interpreting services can also be requested without notice—from a police station or hospital, for example.

3.3.2. Procedure: Determine what information to include with the request

The information included with the service request varies depending on the source of the request. Direct requests from a hospital department provide more information than those from an appointment centre, explained Ad1(F,40). The reason for the consultation and the name of the department making the request are examples of the type of information that could help banks of interpreters be more efficient, enabling them to contact the best interpreter for a given job based on individual skills and preferences. As the participant cited above explained, “There are interpreters who don’t want to work in certain situations—abortions, for example—but since we aren’t provided with any information on the type of consultation, we don’t have much leeway.”

Similarly, it would also be helpful if interpreters were given some basic information about the service request when the appointment is being set up. If they knew more about the context of the interview, interpreters could better prepare by reviewing any specialized terminology, for example, as explained by In8(F,60), who always makes a point of finding out the context of an interview beforehand.

3.3.3. Procedure: Clarify procedures regarding appointment cancellations or changes

With respect to the logistics of setting up appointments, several interpreters complained that when interpreted interviews are set up by telephone, practitioners are more casual about cancelled or changed appointments. In2(F,30) said that she was often not informed when interviews were moved forward, postponed or cancelled:

Another thing we’ve noticed is that they sometimes don’t respect appointment times for remote interpreting. For example, if we’re supported to be ready to interpret from 1:00 pm until such-and-such a time, they’ll either be late or ahead of schedule. […] They sometimes forget that they’ve reserved an interpreter or, when the user arrives, they decide their French is good enough and forget to notify us. When that happens, you just have to wait. And when you call the bank of interpreters to find out what’s happening, they’ll tell you that the practitioner completely forgot to inform you.

In the case of face-to-face interviews, it is rare that practitioners forget to cancel because, if they do, they are required to pay the interpreter the equivalent of two hours’ work to compensate for travel time and lost income. Quebec interpreter banks have no such rules, however, for telephone interpreting.

3.3.4. Procedure: Allow time for briefing the interpreter

According to a majority of participants, the practice of briefing interpreters is a rare occurrence. In fact, several participating practitioners had never even heard of it. In2(F,30) told us that interviews interpreted over the telephone generally start as soon as she picks up, with no preliminary remarks at all:

The doctor is familiar with the case because he has already spoken several times with the user through an interpreter, but we just got there. It’s like day 1. We have no idea what has happened before or what will happen after. We’re just dropped right into the middle of things. We can’t interpret properly under those conditions.

This impression that interpreters can have of being “thrown into the deep end” is particularly disquieting in the case of RI. Unlike interpreting in person, V-RI or T-RI requires additional support to compensate for the interpreter’s virtual presence. This point is discussed in the following sub-section.

3.3.5. Procedure: Establish a debriefing procedure

According to participants, it is rare that the practitioner and the interpreter will take a few minutes after a remote interview to talk about how it went. Aware of the importance of debriefing, Pr6(F,30) explained why she found it harder to debrief at a distance than face-to-face:

It’s not always easy to say to an interpreter, “I don’t like it when you do such-and-such a thing.” Especially if you want to maintain a good working relationship with that person. […] I think it would be a good idea [if debriefings were required] because, in general, it’s a step that’s neglected.

Pr6(F,30) also mentioned that having someone act as an intermediary between practitioners and interpreters greatly facilitates the debriefing process:

It was made clear that we could send an email to [the administrator] if there were technical problems or if we had told the interpreter about an aspect of their behaviour that we didn’t like and the situation hadn’t been corrected, or simply if we weren’t comfortable talking directly to the interpreter about a more sensitive issue. That’s something I’ve done in the past. It’s a lot easier to do through an intermediary.

This practitioner also pointed out that this type of procedure could also protect practitioners’ working alliance with their interpreters.

Another practitioner, Pr8(F,30), explained that it was simply impossible for her to do a debriefing in the context of her practice because she met users in their homes where she then conducted a conference call with the interpreter. She suggested an alternative, saying that it could be the responsibility of the administrator to initiate the dialogue separately with practitioners and interpreters. This would encourage practitioners to express any problems experienced during the interview, even if they did not initially consider them to be particularly important.

3.3.6. Human resources: Clarify everyone’s availability

Clarifying the availability of interpreters, practitioners and administrators would facilitate the appointment-making procedure and help service operations run more smoothly.

The majority of interpreters are self-employed, working part-time on call, juggling other activities at the same time. As Ad2(F,30) explained, they are usually available in the late afternoon or evening to deal with last-minute requests for interpretation services. Practitioners’ schedules often require that the interpreted interview take place on the same day the request was made, however, and their standard hours are from 8:00 am to 4:00 pm.

Practitioners often go through a bank of interpreters to send an urgent message to a user. The remotely interpreted interview therefore involves two other people: the interpreter and the user. As Ad2(F,30) explains, however, it can be difficult to meet such requests—the user might not be home when the interpreter calls or may have questions that the interpreter cannot answer without the practitioner’s input.

The service request may also be made directly to the interpreter, which makes it difficult for the interpreter to refuse an urgent request, even if their immediate situation is not ideal for RI—for example, if they are driving or in a location where confidentiality may be compromised.

From the point of view of the banks of interpreters, their business hours may conflict with the time of requests. As Ad1(F,40) explained, if a request is made at 3:00 pm, it may not be picked up as staff usually stop work at that time. Pr1(F,50) said that the bank of interpreters she works with closes at 4:00 pm, which makes it difficult to arrange follow-up appointments with users who go to school and are generally only available between 3:30 and 5:30 pm.

3.3.7. Human resources: Provide practitioners with information and training

Pr3(F,50) mentioned that it would be a good idea for hospitals to better publicize the availability of telephone interpretation services, as most practitioners are still unaware of this option. She also suggested that banks of interpreters offer training in accessing the service and that a concise information document be widely distributed.

For example, when she started using RI services, Pr2(F,30) didn’t know that it was up to her to initiate the conference call:

I soon understood that the interpreter and the parent would be on the line at the same time, and that it was my responsibility to make the call. So I had to learn how to do that very quickly.

In9(F,30) also mentioned that her establishment would sometimes omit to unlock the conference call function, which could result in the interview being cancelled.

Finally, Pr6(F,30), the only practitioner who mentioned having received training in all types of interpreted interviews, emphasized how useful that training had been.

3.3.8. Human resources: Preserve expertise

Pr8(F,30) told us that she would soon be leaving her current position and was worried that the expertise she had developed in the area of interpreted interviews would be lost when she left the unit. Although she felt it was important that the other members of the team be able to benefit from her experience, no one in a management position had approached her on this subject.

3.4. Supporting the virtual presence

Virtually all the interpreters and practitioners were adamant about the immediate consequences of remote methods on the interpreted interview and the importance of having techniques to mitigate the effect of the virtual presence. These techniques are divided into two sub-themes based on whether they are related to the interview in general or more specifically to speaking in turn.

3.4.1. The interview: Set a standard length and a maximum length

The race against the clock that characterizes so much of our daily lives—the medical profession being no exception—conflicts with the short pauses recommended in the literature on PSI and supported by all the participants in our study, practitioners and interpreters alike.

Pr6(F,30), who has access to face-to-face interpreting services, explains that she does not recommend interpretation by telephone for interviews that will be longer than 30 minutes.

When describing an interview he interpreted over the telephone that lasted several hours, In1(M,40) told us how exhausted he was and how he had to find ways to avoid falling asleep during the interview:

[They] told me, “I don’t know how you’re going to have time to breathe today!” There were 12 people who were all sick. I started at 8:00 in the morning and went until 2:00 in the afternoon. The patients were talking to different doctors, but I was the only interpreter. […] I was absolutely exhausted. I was so tired that I had to jump up and down in the room where I had shut myself in with the telephone just to keep myself awake.

The challenge of limiting remote interviews is made even more difficult by the fact that they generally require more time, as In10(F,30) explained:

Sometimes we need more time because the practitioners aren’t able to reassure the client over the telephone, whereas when they are present in person, they introduce themselves and reassure the user. […] Sometimes you ask [the practitioner] questions and you can tell that the answer is completely beside the point; it’s like they’re not there or don’t understand. It’s difficult for me to take time with them.

3.4.2. The interview: Consider an initial contact face-to-face

It is more difficult to build a working alliance in the context of remote exchanges, which tend to generate relationships that are more utilitarian, brief and impersonal. When Pr6(F,30) mentioned these problems, she also suggested that the first contact with the user be in person before moving on to RI.

Even though V-RI allows participants to read facial expressions and provides a more complete social presence than T-RI, it is still harder to establish a good working alliance on-screen than face-to-face. In fact, many practitioners are reluctant to use V-RI for sensitive psychosocial interventions. However, the COVID-19 pandemic is forcing a significant number of practitioners to use this modality on a regular basis and thus to become accustomed to it.

Pr1(F,50) raised the issue of equality underlying the choice of interpreting method. She stated that it is important to offer allophone users the same service as other users: “In my opinion, if we wouldn’t do it in French, we shouldn’t do it in translation. First appointments are not done over the telephone.”

3.4.3. The interview: Prepare documentation as needed

Several practitioners and interpreters spoke about the communication challenges involved in talking about medication during medical consultations. When dealing with people who do not speak the language of use, it seems to be easier to talk about medications, side effects and dosages in person, as interpreters can ask users to bring their medications with them to the consultation. Barriers created by language and the poor literacy skills of certain clients make it more difficult to obtain information on medication from a distance.

Pr4(F,40) explained that she tries to remedy this situation by obtaining a list of the patient’s medications from the pharmacy ahead of time in order to simplify discussions during the interpreted interview and reduce the risk of medical error.

3.4.4. The interview: Make sure the user is alone in a quiet space

Several participants raised the importance of the user being alone in a quiet space with no distractions during the interpreted interview. It appears that this is not always the case: the user may be surrounded by family members or in some other very noisy environment. Both of those factors can affect the interpreter’s concentration and even affect the quality of the interpretation.

If the user is in an inappropriate space, family members can get involved in the conversation with the interpreter and interfere with the intervention, potentially affecting the quality of the consultation, as In6(F,50) explains:

I was doing a video conference with a doctor, a nurse [and a user] to discuss surgery scheduled for the next day. Whenever the doctor asked a question, it was the patient’s husband who answered. I asked to speak to the woman directly, but the phone was on speaker. While I was speaking, I could hear someone whispering to the patient. The next day, when I came in for the operation, the lady hadn’t received the same information I had given her; her husband had given her the wrong instructions by talking in her place. As a result, she was not prepared for the surgery, even though I had provided the correct instructions.

3.4.5. The interview: Allow time at the beginning

RI tends to be more expeditious, with less attention paid to introducing the interpreter (i.e., explaining the process and confidentiality). Practitioners are inclined to cut to the chase:

Over the telephone, it’s like doctors are in a hurry, even though the user is right beside them. They want you to interpret what they’re saying, but they don’t give you the time to express yourself. All they say is, “I’m with someone whose name is Mr. X. I am Dr. Y.” And then you introduce yourself: “I’m the interpreter. My name is Mr. Z.” And you start interpreting. But when everyone is in the same room, you have more time… maybe two minutes or five minutes: you introduce yourself, they introduce themselves, and you have a chance to talk about prescriptions, confidentiality and all the rest.

In1(M,40)

Skipping over these introductions does little to foster users’ trust. They might be more reluctant to divulge certain important information for fear, for example, that the interpreter will pass the information on to other members of their community.

3.4.6. Managing turn-taking: Explain the importance of speaking in turn to the user

Comments made by In1(M,40) illustrate the kind of confusion that can arise when everyone talks at the same time during interviews that are being remotely interpreted:

[This goes on] until I say, “No. You need to do this in an organized way. First one talks, then the other. If you talk at the same time, I can’t interpret what you’re saying.” But when you’re in the same room, it’s easy to say to everyone present, “Ok, everyone has to take turns speaking.” When you’re on the phone, you can’t see anyone; you don’t know what’s going on at the other end. You have to try to integrate yourself into the… situation they’re creating. That’s a lot harder.

This feeling of being overwhelmed and not being able to find their own place is particularly acute in a legal/courtroom context, when everyone’s words overlap or when there is a delay caused by the camera (when the hearing is being held in two separate rooms or when the interpreting is gesturing to take a turn but the camera is focused on another interlocutor).

3.4.7. Managing turn-taking: Explain to participants that interjections may be more awkward

Practitioners can also find it difficult to interrupt one of the other participants in the interview in order to intervene. Such interjections may be perceived as abrupt, rude or aggressive, as Pr6(F,30) explained:

When everyone is there in person, it is easier to stop the interpreter and say, “Just a second… there’s something I want to reformulate.” Since the client sees what we’re doing, it maybe seems less aggressive than on the phone. […] But since there’s no visual or non-verbal feedback during a conference call, I don’t take a chance. I let the interpreter translate what I said, even if it takes some time before I can provide feedback, and then I re-explain or say what I wanted to.

3.4.8. Managing turn-taking: Check that predetermined signals are working

Sometimes the people involved in a remote interview forget the predetermined signals, or there is confusion about how to know when someone has finished speaking (unlike during face-to-face conversations). Pr6(F,30) explained how the use of communication signals differs in both contexts:

In person, we have visual contact, so feedback is easier. We don’t have to stop the interpreter, say something, and start again. We can often see at a glance that the speaker hasn’t finished. Being able to see each other makes everything so much easier, more natural, more fluid than over the telephone. So it’s important that we use artificial methods to indicate that we have finished speaking during a telephone interview.

3.4.9. Managing turn-taking: “Chunking”

Although it is generally recommended that practitioners break up their remarks into segments (called “chunking”), the experience of Pr6(F,30) has proved otherwise. She has noticed that during remote interviews, if she slows down her rate of speech to give the interpreter time to assimilate the information she is communicating, the user will generally not wait for her to continue. Instead, they will take advantage of a strategic pause to ask questions or make remarks. She has therefore, somewhat paradoxically, gotten into the habit of segmenting her speech less in order to make things easier for the interpreter and keep the conversation on track:

I realized that segmenting my remarks over the telephone made things a bit more difficult, in that it encouraged users to provide lots of feedback, like “Yes, but I have a question…,” or “Yes, but what about…” That happens in person as well, of course, but it’s easier to handle. You can use non-verbal cues to say, “Yes, I know, but I haven’t finished.” They can see by your body language that you have more to say, so they don’t ask as many questions.

4. Conclusion

Results lead us to conclude that existing recommendations in the literature regarding the planning and supervision of remote public service interpreting only partially reflect the experience of Quebec public sector interpreters, practitioners, and administrators. In response to the second research question, we produced new guidelines. The Guide for the Planning and Practice of Remote Public Service Interpreting (see Appendix) is the core result of our field study, for instance the review and revision of the recommendations in the literature, including the addition of seven new recommendations. The Guide includes a total of 35 recommendations: 10 on planning and managing RI services (R-1.1 through R-1.10), and 25 on the practice of interpretation specifically (R-2.1 through R-2.25).

The Guide is an original contribution for two main reasons. First, it constitutes a review by various key actors in the field of several existing recommendations in the literature on RI. Secondly, the formulation of the revised recommendations and the addition of clarifications and new recommendations enables the reader of the Guide to grasp the importance of effective communication between all the social actors involved in PSI, not only between the practitioner and the interpreter, but also, for example, with interpreter banks, different instances of the health system and the university.

In other words, and to give a formal definition of remote public service interpreting, results obtained show that RI does not refer solely to telecommunications technology, but also to the knowledge and skill set needed to supervise and support the use of that technology in specific practice contexts while minimizing the effect of the virtual presence. The challenges related to communication between the key actors in this practice are generally the same for all types of PSI, whether the interpreted interview is carried out face-to-face or remotely.

Since they have to deal with them every day, the key actors in the field are well aware of the challenges presented by the various forms of RI. Table 1 presents our three additions to the recommendations in the literature on this subject.

Table 1

Diversity of practice contexts: List of changes to recommendations in the literature (also available in the Guide)

Diversity of practice contexts: List of changes to recommendations in the literature (also available in the Guide)

-> See the list of tables

Because of the different practice contexts, the specific requirements of interpreting services (R-1.1 through R-1.7) need to be carefully considered when developing an implementation and management protocol (R-1.8) and during its gradual rollout (R-1.9). This study also shows that a single protocol for implementing and managing RI services could not be applied across the entire territory without jeopardizing service quality.

This result aligns with one of the principal results of a systematic literature review by Bradford, Caffery, et al. (2016: 8) on telehealth services in Australia: “[O]ur findings highlight the importance of adaptability and efficiency, which have not been reported previously; the need to adapt and modify the service model in response to need was a frequently reported factor for the success of services.”

Participants also all agreed that there were significant barriers to communication between key actors in the field. Table 2 presents the changes made accordingly.

Table 2

Barriers to communication: List of changes to recommendations in the literature (also available in the Guide)

Barriers to communication: List of changes to recommendations in the literature (also available in the Guide)

-> See the list of tables

The various changes related to management led us to propose that a better framework be implemented for transmitting information about the interpreted interview. A new recommendation to that effect was therefore added (R-1.10). This recommendation is aimed specifically at reducing barriers to communication by supervising the distribution of information needed to ensure that services run smoothly. All the recommendations related to the planning and management of RI services are also indirectly related to those barriers. For example:

  • Preserving expertise in the regions (R-1.7) can be seen as a means to improving communication between a practitioner with specific skills and other practitioners who could benefit from that knowledge. (For a thorough examination of this subject, see Vatz-Laaroussi 2009.)

  • Training for both practitioners and interpreters (R-1.6) is another example. Even though the importance of existing training programs has been proven (e.g., Leanza, Angele, et al. 2020), training distribution has been sub-optimal in the province. That said, the Psychology and Cultures Laboratory completed a pilot project for online training for healthcare practitioners in Quebec City last year (the preliminary results are encouraging: Burdeus-Domingo, Gagnon, et al. submitted).

  • Supervision of debriefing by management could improve the working alliance between practitioners and interpreters as well as the quality of services rendered (R-1.8.16).

Finally, the Guide contains new recommendations and clarifications (Table 3) to compensate for the reduced social presence of the people involved in a remotely interpreted interview. Virtual presence—and the resulting attenuation of social presence—is indeed a complex phenomenon worthy of further study in the coming years.

Table 3

Supervision and support of the virtual presence: List of changes to recommendations in the literature (also available in the Guide)

Supervision and support of the virtual presence: List of changes to recommendations in the literature (also available in the Guide)

-> See the list of tables

It is important to note that the recommendations in the Guide complement those that already exist on the subject of implementing face-to-face interpretation services (e.g., René de Cotret and Leanza 2019) and that the Guide is in no way intended to replace interpreters’ professional code of ethics, such as the National Standard Guide for Community Interpreting Services8 in Canada.

As illustrated by our results, the challenges related to RI are not only technical but also organizational, such as recognizing the expertise of the different actors involved and disseminating that expertise through clearly identified communication channels. As explained in the Introduction, symbolic support from the administration of the institutions in which interpretation services are or will be used is essential. The measures outlined by Karliner and Mutha (2009) or Novak-Zezula, Schulze, et al. (2005) (see Introduction) are examples of strategies that can be implemented to meet the recommendations contained in our Guide.

Finally, globalization confronts decision-makers with challenges that require a collaborative approach and flowing communication. The arrival of 25,000 Syrian refugees in 2015 in Canada was cited as an example on a number of occasions by study participants to reinforce the idea that pressure on public services is leading decision-makers to find solutions to unprecedented problems. The COVID-19 pandemic is certainly another one of these situations and its grave seriousness is forcing us to rethink the way we interact and collaborate. The Guide is in line with this social reorganisation.