Interpreters as Translation Machines: Telephone Interpreting Challenges as Awareness Problems

Abstract

Telehealth has been widely adopted during the COVID-19 pandemic, and this article examines challenges faced by telephone interpreters in working with healthcare providers in the context of the Australian healthcare system. Based on one-on-one interviews with 67 healthcare interpreters in Australia, it explores various elements which affect communication processes in telephone interpreting and interpreters’ views on healthcare providers’ abilities to collaborate with interpreters. Data analysis indicates that telephone interpreting is often affected by a lack of briefing, poor acoustics and the absence of visual cues. While these factors pose significant challenges to telephone interpreters, a provider’s tendency to see interpreters as ‘translation machines’ was perceived as a deeper underlying problem by the interpreters. The mechanistic approaches to interpreting among healthcare providers pose barriers to interpreter–provider collaboration and exacerbate communication problems caused by the external elements in telephone-interpreted encounters. The article calls for urgent need to raise awareness of interpreting among healthcare providers as a key to ensuring desirable health outcomes for patients from minority backgrounds.

Introduction

Interpreting has traditionally taken place face to face, yet the rapid development of communication technology over the past three decades has led to alternative ways of delivering interpreting services to people located in remote areas. Referred to as ‘remote interpreting’, this type of interpreting connects people who are geographically separated from each other, enabling linguistic minority populations to gain access to key public services (Braun, 2015). Telephone interpreting serves as the most common mode of communication for remote interpreting, with healthcare representing a key service domain for telephone interpreting (Wang & Fang, 2019).

Telephone interpreting has taken off in earnest following the outbreak of the COVID-19 pandemic due to the minimisation of face-to-face contact to prevent the spread of the virus (René et al., 2020). In the case of Australia, telehealth usage increased more than five times from less than 6% in 2018 to 37% in 2021 (Zurynski et al., 2022). While there is an increasing body of research on telehealth, relatively little attention has been paid to interpreter-mediated telehealth delivered to patients from culturally and linguistically diverse backgrounds. The present article addresses this gap by examining challenges faced by telephone interpreters in working with healthcare providers in the Australian healthcare context. The study is based on data from in-depth one-on-one interviews with 67 healthcare interpreters (representing 28 languages) in Australia. It focuses on elements which affect communication processes in telephone interpreting and healthcare providers’ abilities to collaborate with interpreters, as reported by the participants. Given that telephone interpreting primarily serves patients from minority backgrounds, addressing this topic is important to ensuring health equity and quality of care for vulnerable populations.

Telephone Interpreting in the Australian Context

Telephone interpreting in Australia began in 1973 as a world-first in response to the growing linguistic diversity brought about by an increasing number of migrants from non-English-speaking backgrounds (Ozolins, 1991). Since a free Emergency Telephone Interpreting Service was first established to assist non-English-speaking migrants to handle emergency situations, telephone interpreting has evolved to cover a wide range of areas relating to daily life, with healthcare as one of the key service domains (Department of Home Affairs, n.d.). Currently, the Translating and Interpreting Service (TIS) operating under the Department of Home Affairs offers free interpreting across the nation, and the availability of a nationwide free interpreter service has helped Australia to be recognised as a global leader in terms of interpreting services (Bischoff, 2020).

There is a growing body of research on telephone interpreting in Australia. Table 1 summarises the advantages and disadvantages of telephone interpreting identified by research in the Australian context.

Table 1.

Advantages and Disadvantages of Telephone Interpreting in Australia.

Advantages	Disadvantages
Affordances for interpreting service providers for service expansion (Wang, 2018a)	Low levels of remuneration for interpreters (Wang, 2018a)
Convenience due to the removal of the need to travel (Lee, 2007; Wang, 2018b)	Occupational factors including poor working environments, health impacts from prolonged sitting, and a lack of professional recognition compared with face-to-face interpreting (Lee, 2007; Ozolins, 2011; Wang, 2018a; Xu et al., 2020)
Improved concentration and impartiality compared with face-to-face interpreting (Ko, 2006; Lee, 2007)	Communication issues relating to poor acoustics, a lack of visual cues, overlapping speech, a lack of briefing and inadequate understanding of telephone interpreting among parties to communications (Ko, 2006; Lee, 2007; Wang, 2018a, 2018b, 2021; Xu et al., 2020)

Among the issues discussed above, communication problems caused by external factors and inadequate understanding of telephone interpreting among parties to communication may directly impact quality of care for patients. There is, however, a conspicuous lack of research which examines how (if at all) telephone interpreters and healthcare providers are able to effectively manage the communication issues that inevitably arise. Examining this issue is important because a shared understanding and willingness to collaborate among primary parties is a key to achieve communication success in telephone interpreting (Wang, 2021). It is all the more significant when considering power differentials embedded in interpreter-assisted healthcare consultations, as is discussed in the next section.

Theoretical Frames

Power is inseparable from community interpreting, in which communication usually occurs between a party familiar with dominant language, cultural and social norms and a person with little knowledge of those critical elements (Mason & Ren, 2012). In interpreter-mediated clinical encounters, power differentials can impact not only patients’ abilities to contribute to communication (Kilian et al., 2021; Meyer et al., 2003) but also those of interpreters (Cho, 2021; Woll et al., 2020). While interpreters are said to be situated in a neutral space to serve as impartial mediators away from power dynamics (see Mason & Ren, 2012, for further discussion), interpreters’ choices and actions tend to be shaped by power asymmetries in healthcare settings, and these asymmetries generally accord greater power to healthcare professionals (Brisset et al., 2013; Garrett, 2009; Hsieh & Kramer, 2012).

In discussing power dynamics in healthcare contexts, it is important to consider the notion of ‘interpreters’ invisibility’, according to which ‘interpreters are not considered to be parties to the conversation, but rather they are seen as language-switching operators in line with the conduit model of communication’ (Angelelli, 2004, p. 7). The conduit model of interpreting posits that interpreters should deliver exactly what has been said between the sender and the receiver without making any changes to the original message (Hsieh, 2009). The neutral conduit role is shaped by the view of communication for which there should be only one true meaning (Angelelli, 2004) and is defined solely on the basis of the ‘core function of “message transmission,” performed by a third party, whose presence, ideally, is as “invisible” as possible’ (Avery, 2001, p. 6).

In recent years, the conduit model has been increasingly challenged by interpreting scholars who pay attention to the visibility of interpreters. A growing body of research emphasises interpreters as key communication participants who actively co-construct messages with other primary parties by utilising not only language but also cultural, interpersonal and socio-political skills to achieve desirable communication outcomes (see, for example, Angelelli, 2004; Chernyshova & Ticca, 2020; Davidson, 2001; Hsieh, 2008, 2016; Latif et al., 2022; Major & Napier, 2019).

Extant research on healthcare providers’ perceptions of interpreting, however, shows a different picture. Healthcare providers tend to see interpreting as a channel for word-for-word conversion (Bischoff et al., 2012; Bot, 2005; Brämberg & Sandman, 2013; Lai & Costello, 2021; Leanza, 2005; Singy & Guex, 2005), and interpreters are generally regarded as messengers working exclusively between languages on their own, rather than working with other parties (Cho, 2022). Considering these mechanistic views of interpreting among healthcare providers, it is important to examine interpreters’ own experiences relating to the institutional understandings of interpreting in the case of telephone interpreting, in which the interpreter is literally ‘invisible’, and is present only as a voice coming out of a machine. In such encounters, what specific challenges do telephone interpreters experience and how would the physical invisibility of telephone interpreters impact their abilities to work with providers?

There is a significant lack of research which investigates the intersections between interpreters’ invisibility, communication challenges and interpreter–provider collaboration in telephone interpreting settings. This research examines this issue from the perspective of telephone interpreters, with a focus on the following questions:

• (i) What are the factors which impact telephone-interpreted encounters?

• (ii) How does the physical invisibility of telephone interpreters intersect with elements which impact communication in telephone interpreting?

• (iii) How are communication challenges in telephone interpreting managed in light of inter-party collaboration?

Method

Participants

This study is based on qualitative interviewing (Byrne, 2004). The data included in this study were gathered from one-on-one interviews with 67 healthcare interpreters (representing 28 languages) from across Australia. Two key criteria were applied to participant recruitment: (i) holders of interpreter certification administered by the National Accreditation Authority for Translators and Interpreters (hereafter ‘NAATI’), a national governing body of translators and interpreters in Australia, and (ii) a minimum one-year experience with healthcare interpreting. The interpreters were recruited through two different sources: an online NAATI directory (which contains the contact details of all translators and interpreters in Australia by language category and location) and the NSW Multicultural Health Service, which administers healthcare interpreting services across New South Wales. The author and a research assistant contacted interpreters via email, and those who met the inclusion criteria and expressed willingness to participate in the research were invited for in-depth interviews.

Ethics approval was obtained from the Human Research Ethics Committee of the university for which the author works. All 67 participants signed written information and consent forms sent out together with recruitment emails. Prior to the interviews, the participants were asked to complete a brief online survey relating to demographic details, professional experiences (e.g. years of interpreting) and interpreting challenges. The demographic data of the participants are listed in Table 2.

Table 2.

Participants’ Demographic Data.

Category	Range	Number	%
Gender	Male	20	29.9
	Female	47	70.1
	Total	67	100.0
Age	50+	35	52.2
	45–49	8	11.9
	40–44	11	16.4
	35–39	8	11.9
	30–34	4	6.0
	25–29	1	1.5
	Total	67	100.0
Education level	PhD	3	4.5
	Master’s	35	52.2
	Bachelor’s	14	20.9
	Undergraduate	12	17.9
	Incompletion	2	3.0
	No response	1	1.5
	Total	67	100.0
Profession	Full-time	35	52.2
	Part-time	26	38.8
	Retirement job	6	9.0
	Total	67	100.0
Years of experience	<1	1	1.5
	1–5	18	26.9
	6–10	11	16.4
	11–15	17	25.4
	>15	20	29.9
	Total	67	100.0

Procedure

All interviews were held during the first half of 2022 via Zoom as part of social distancing measures due to the COVID-19 pandemic. The interviews were recorded on Zoom by the author. Before starting interview recording, the author reminded participants of the information on Zoom recording in the information and consent forms which they had signed before participating in the interviews. The interviews were conducted by the author in English and each interview took approximately 1 hour. The interviews were semi-structured and were informed by the principle of responsive interviewing (Rubin & Rubin, 2011), in which the interviewer begins an interview with a topic in mind but recognises the possibility of modifying questions to match the knowledge and interests of the interviewees. The Zoom-recorded interviews were transcribed by a professional transcription company, with any personal details that might lead to participant identification removed.

The proofed interview transcripts were coded in NVivo by the author, with thematic content analysis proposed by Braun and Clarke (2012) as a key analytical prism. Analysis involved four specific steps. In the first phase, the author began by reading the whole data set a number of times for data familiarisation, which enabled them to identify core themes relating to communication challenges and inter-party collaboration in telephone interpreting. During the process of data familiarisation, the author documented their thoughts and impressions as memory triggers for coding and analysis. This phase reflects Step 1 of Braun and Clarke (2012).

In the second phase of coding, which relates to Step 2 of Braun and Clarke (2012), the author conducted a systematic analysis of the data. Interpretive approaches (Elliott & Timulak, 2005) were used as an appliable tool to organise the data into distinctive meaning units. The author remained flexible with and open to using the data throughout the process of meaningful organisation, and critical auditing as proposed by Elliot and Timulak (2005) was used. As part of the auditing strategies, the research assistant analysed a sample data set relating to communication challenges arising from telephone interpreting. The result of the coding conducted by the research assistant was compared by the author with the author’s own coding to ensure the rigour of the analysis. This process revealed similar coding outcomes from both the research assistant and the author.

The third phase of data analysis concerned theme generation through the abstraction of the main findings and theme reviewing for validity. The author actively read and reviewed the coded data to generate themes and subthemes. Not only the individual distinctiveness of the codes but also the relationships between the codes were considered to the process of theme generation. Throughout this process, the author had constant dialogues with the coded data in order to create a meaningful thematic map that reflects the depth and richness of the data. The generated themes were then reviewed through a recursive process whereby themes were compared with the coded data and the entire data set to ensure validity. This phase refers to Steps 3 and 4 of Braun and Clarke (2012).

In the final phase of data analysis, defining and naming themes (Step 5 of Braun & Clarke, 2012) was carried out. With the key research questions as a guiding principle, each theme was developed not only in its own right but also in relation to the other themes and the research questions in order to tell a coherent story about the data. Figure 1 provides an overview of the thematic results.

Figure 1.

Overview of thematic results relating to challenges to telephone interpreting.

Results

Three themes are identified and presented in this section. Quotes from the interview data are provided where appropriate to ensure credibility. Each participant is referred to using a number, for the purpose of anonymity.

Theme 1: The Gap Between Guidelines and Practices Regarding Inter-Party Collaboration

One of the most salient themes which emerged from the data analysis is the gap between the guidelines and practices in terms of provider–interpreter collaboration. Guidelines on working with healthcare interpreters published by various states of Australia recommend that healthcare providers have a pre-interview briefing with the interpreter to facilitate smooth communication during interpreting (see, for example, NSW Health, 2017; Queensland Health, 2007; Victoria Department of Health and Human Services, 2017; Western Australia Department of Health, 2017). According to the ‘Standard Procedures for Working With Healthcare Interpreters’ published by the NSW Health (2017), for example, interpreters must be briefed on the nature of the consultation, the healthcare provider’s communication objectives, cultural background information (if relevant) and any potential risks. The same guidelines emphasise briefings as particularly important to sensitive cases such as domestic violence, sexual assault and trauma counselling.

Data analysis, however, indicates a significant gap between the guidelines and practices. The participants reported that they are hardly ever provided with briefings and just ‘jump in’ to consultations without adequate advance knowledge. While this lack of briefing applied to both telephone interpreting and face-to-face interpreting, the participants noted that face-to-face interpreting usually enables prior interactions between the interpreter and the patient in a waiting room, and interpreters sometimes obtain useful information that is relevant to the upcoming consultation. For telephone interpreting, however, interpreters are contacted by service providers and whoever is available for a given assignment is generally asked to do interpreting straight away. The predominantly on-demand nature of telephone interpreting is found to be not conducive to interpreters’ chances of obtaining relevant background information:

Most of the time when I – I mean, majority of the time, I don’t receive anything. Just there’s a job booked for you. Go in there. (Participant 56)

The participants reported that interpreting without pre-interview briefing can be particularly challenging in the case of recurring appointments. In such cases, the provider and the patient have shared history, whereas the interpreter does not. Where additional details are needed for the interpreter to understand a communication context, participants noted that being physically invisible limited their capacity to seek the necessary clarifications. Participants also noted that it is much harder to work out when to intervene in telephone interpreting, compared to face-to-face consultations in which interpreters can see and pick up cues for a pause. Attempts to seek clarifications often result in communication inefficiencies and delays, as exemplified below:

If the patient and the doctor is together, they know what they’re talking about. They have the history. You know, they’ve been talking about this before and they can say oh, this, or that and that. Then all together jumbled and I try to understand what they mean when they refer to him or her or it or that or this. I have no idea what they’re talking about and I have to say hang on, stop it there. Can I – do you refer to him or her or that or yesterday or the day before? So, I have to make them clarify and it could be quite tricky sometimes on the phone. (Participant 37)

While some participants reported that they usually try to request information when a briefing is not given, it is worth noting the relationship between a lack of briefing and provider perceptions of interpreting. Generally, the participants perceived that providers tend to believe that interpreters can manage interpreting without any briefing at all. For example:

They say, if you a professional interpreter, you should be able to interpret everything. There’s no need for briefing. (Participant 55)

The expectations that a good interpreter should be able to perform well without any prior knowledge of the situation can be understood in relation to a lack of shared understandings about interpreting among healthcare workers. As much as 60% of the participants discussed a gap in expectations of interpreting between interpreters and providers. Whereas the interpreters generally regarded communication in interpreting as a co-constructed process (as exemplified by the expressed need for pre-interview briefings), participants reported that the healthcare providers’ approach to the interaction often suggested a view of interpreting as an independent and automated process. This often resulted in situations where interpreters felt that they were seen as little more than ‘translation machines’:

They don’t introduce an interpreter. They don’t allow a chance to – they just treat you as a robot or as a machine. They haven’t been trained. You can tell that they don’t know how to work with interpreters. (Participant 2)

A mechanistic view of interpreters may also have implications for cultural elements of patient care. Taking patients from Latino backgrounds as an example, it is culturally important to take time to greet patients and engage in small talk to build a human-level connection to ensure communication success (Magaña, 2020). Culturally competent care is particularly relevant to historically marginalised groups such as First Nations People in Australia (Hansen & Charles, 2022). If interpreters are not acknowledged as legitimate participants in the clinical encounter, this may limit their capacity to engage in fundamental communication practices that are culturally linked to notions of care. Such problems may be further compounded by the western scientific approaches to ‘culture’, as exemplified below:

They want the magic wand. Just tell me how to fix it. Just tell me how to fix it. Give me the right word so I can fix it. Sometimes culture is not like that. (Participant 10)

The mechanistic view of interpreters reportedly held by providers caused dilemmas for some participants as to whether or not they could specifically request briefings. As interpreters are often expected to perform independently without any pre-knowledge or information at all, participants worried that requests for briefings might negatively impact views on their professional competence:

I do ask them sometimes. I say, so what is this about? Can I know a little bit about this? They do, oh, excuse me, I should’ve said. Yes, that’s a good one. Some, yes, they give me briefing, kind of, about what to expect. I don’t know, but maybe I shouldn’t have this thinking. But I think when I ask that, they kind of, not really not appreciate – not appreciate, not consider me competent. (Participant 15)

Examining the gap between the guidelines and practices regarding inter-party collaboration provides an important clue as to how interpreters are perceived by healthcare providers. The usual practice of not offering any pre-session briefing indicates interpreting as a solo activity and an automated process, rather than as an inter-party collaborative process. This reported tendency that sees interpreters as independent translation robots is not limited to briefings but applies to other communication challenges specific to telephone interpreting, as the following emergent themes attest.

Theme 2: Interpreting as an Activity Occurring in Soundproof Space

The mechanistic view of interpreting is salient in the case of poor acoustics which frequently occur during telephone interpreting. Data analysis indicates three specific elements which impact sound quality in telephone interpreting. The first element relates to background noises which can disrupt interpreters’ abilities to listen. Sound quality was reported to be particularly bad in the case of dental treatment, obstetrics and emergency departments. Dental treatment generally involves loud machines for treatment, and interpreting for a patient giving birth is often interrupted by pains and screams. In the case of emergency departments, there are significant background noises such as people talking loudly, hospital staff running around, document shuffling and patients in distress or discomfort. For example:

So, in the emergency departments, sometimes it’s somebody’s mobile on a speaker, and there’s a lot of noise. A lot of noise, the medical instruments being shuffled and handled and all that thing, and microphone sometimes pick certain frequencies and sometimes once you are there, you don’t feel those noises, but over the phone you feel those noises very loudly. (Participant 47)

Second, the common use of speakerphones for telephone interpreting was found to pose challenges to telephone interpreters. The participants reported that speakerphones had become even more common since the onset of the COVID-19 pandemic as part of social distancing strategies. Furthermore, mask-wearing (which remains mandatory in Australian clinics) makes it even more difficult for interpreters to hear:

With COVID, the other thing is that there is the addition of masks – face masks – and that makes it more difficult to hear because both the patient and the professional – the doctor – they both have masks and it muffles the voice. (Participant 7)

Third, data analysis reveals sound challenges associated with interpreting for elderly patients over the phone. Older patients tend to have hearing impairment, and interpreters often have to raise their voices when serving elderly patients. Interpreters’ efforts, however, often fail due to background noises and the use of speakerphones, which together work to significantly impact elderly patients’ abilities to hear properly. Furthermore, older patients with hearing impairment may lack a sense of communication management, which adds more complexities to the already challenging interpreting jobs:

That’s a real challenge, because they – and sometimes they have a hearing problem, and they can’t hear my interpreting and they kept repeating what they want to say. (Participant 31)

The participants reported several strategies to deal with sound issues. These include asking for clarification, requesting to stop using speakerphones, explaining their need to hear well in order to interpret accurately and even warning that they will have to stop interpreting unless noises are addressed. Data analysis indicates that chances of success largely depend on providers’ attitudes to interpreting. According to participants, some providers are willing to accommodate interpreters’ requests, whereas others appear to regard sound problems as irrelevant to interpreting:

When they call you from the hospital, the reception is really bad and you can’t hear properly. Sometimes you repeat, sorry, I don’t understand, I’m catching every second word. But still, doctors, they want to continue, they want the job to be done. (Participant 62)

The reported lack of acknowledgement of the impact of poor sound on interpreting is found to affect quality of care for patients from culturally diverse backgrounds. For instance, cancer is strongly feared by certain ethnic groups, and culturally appropriate management of cancer patients is an important element of patient care (Costas-Muñiz et al., 2020). In many cultures, non-disclosure is valued because finding out about the truth is feared to impact patients’ hope for survival. In Australia, however, patients’ right to know is the norm, and patients are generally informed of bad news straight away (Cho, 2021). While such cultural differences need careful approaches to patients, the reported provider tendency to treat sound problems as irrelevant to interpreting can affect interpreters’ abilities to manage sensitive cases appropriately. For example:

It was cancer. The patient was going to die basically. They told the patient right there and then. They were telling the patient that it was better to go home to die. The interpreter had to interpret that over the speaker. I could hardly hear what was being said. I had to interrupt so many times […] I would have liked to do it in a more compassionate way by being prepared myself for it. But I wasn’t prepared and the interpretation was poor audio quality – not something that I would have liked for myself if I was the patient. (Participant 2)

In cases where sound problems persist, the participants reported that they have no other choice but to keep asking for repetitions yet in an apologetic manner in order to avoid frustration on the part of healthcare providers. This reported provider view on interpreting as an activity uninterruptable by sound issues can also be found at an institutional level. Participants who work for large hospitals pointed out a poor working environment for telephone interpreting work, which usually occurs in a designated space. As a group of interpreters work in the same room speaking loudly on telephones, the level of noise significantly impacts interpreters’ abilities to manage communication:

It is even more stressful because it’s stressful for everybody, because if I have an interpreter screaming next to me in Chinese and I have to scream in Italian and the patient – it's really bad […] You can’t be 10 in a room and work over the phone. (Participant 39)

Data analysis thus illustrates a lack of understanding about interpreting at both individual and institutional levels. In line with the findings in Theme 1, the lack of acknowledgement of external factors that clearly impact human capacity to hear and process information reinforces the interpreters’ perception that they are seen as translation machines. It also constitutes further evidence of providers’ perceptions of interpreting as an automated process (rather than a collaborative partnership) that is reliant solely on interpreters’ own competence, rendering inter-party collaboration irrelevant to interpreting success.

Theme 3: Interpreting as a Solely Listening-Based Activity

Another significant challenge to telephone interpreting reported by participants is the absence of visual cues. Non-verbal elements such as body language influence ways in which communication is delivered in healthcare settings (Wanko et al., 2020). As Participant 44 commented, ‘“The person could be rolling their eyes, but saying the opposite’, but the interpreter unable to see sarcasm may get the message wrong. Data analysis identifies six areas in which the absence of visual cues presents challenges to interpreters.

First, physical moves which are common in healthcare settings are reported to be challenging to interpret, and physiotherapy is one good example. Physiotherapy sessions usually involve demonstrations of moves, yet participants reported that instructions are often not specific enough for telephone interpreters who cannot see the physiotherapist. For example:

Without me seeing the actual instruction or how they move or how they utilise their equipment, et cetera, because basically I can’t see anything. It’s not easy to understand what’s going on. I just have to completely rely on what I can hear, but I don’t sometimes understand what’s been happening. (Participant 6)

Second, interpreters’ inability to see makes it difficult when attempting to pinpoint specific areas of pain. Participants reported that patients often point to areas of pain by simply saying ‘here’. As recalled by Participant 20, patients tend to say ‘I feel pain all over here and here and there’ without identifying specific areas of pain. Third, interpreting challenges occur when healthcare providers use visual aids such as X-rays during the session. Fourth, interpreters often have trouble working out when parties have finished talking, which leads to interruptions of flow, silence, awkwardness and repetitions. For example:

You don’t know whether they have finished. You don’t know when you shall start and a lot of times you are just talking over each other because you think he has finished because there is a gap and then he just starts to talk again and you have already started interpreting. So, a lot of times, yeah, you are like either like overlapping with each other or you have to repeat yourself. (Participant 12)

Fifth, not being able to see comes with the challenge of being unable to understand the immediate physical context in which the encounter is taking place. In group communication situations such as a family conference, telephone interpreters cannot tell who is who and may thus have trouble following the flow of conversation. Participants reported that the relevance of visual cues to interpreting is often not acknowledged by other parties, who tend to view interpreting as a solely listening-based activity:

I was asked to do a – to interpret for palliative care. That’s all the briefing that I got. All right. When the call is connected, I can hear five people talking. I can hear Spanish talking. It’s a male. I can hear female speaking in Spanish. I have no idea where they are, whether they were at home, they are in a hospital. They just don’t tell you. You are a machine. You just interpret what you hear. (Participant 16)

Finally, random turn-taking and resultant speech overlaps in group communication contexts were reported as a significant challenge for telephone interpreters. Participants pointed out that many non-English-speaking patients, particularly elderly patients, are accompanied by carers who can speak English. Depending on the importance of the occasions, patients may come with a group of family members and relatives, each of whom may participate in the conversation at random. This is problematic for the interpreter who has not been properly briefed on the assignment and is unable to see who is speaking at any given time. Furthermore, speech often overlaps, which affects already poor sound quality over the phone:

Especially when the recipient is with a family member, they talk over you and you can’t hear anyone at all. If the doctor wants to say something and someone in the room wants to say something and then the patient’s family member also wants to say something, it’s a very big disaster […] over the phone, it’s everyone’s voice raised at the same time, and it’s very hard to hear everyone. (Participant 28)

In order to cope with the challenges caused by the lack of visual cues, participants reported various professional strategies which include (i) checking with parties to ensure proper understanding; (ii) switching to simultaneous interpreting so that interpreting matches physical descriptions (e.g. ‘here’) in the case of pain descriptions and the use of X-rays; (iii) reminding parties of the interpreter’s inability to see who is speaking; and (iv) asking individuals to take turns in an orderly manner. While the first two strategies can be put into practice at the interpreter’s discretion, the third and fourth strategies require collaboration from other parties and are identified to be problematic due to a general lack of awareness of interpreting. Once again, interpreting is regarded as a solo activity which does not need inter-party collaboration:

I have experienced doctors saying, it’s okay, just translate this like that. (Participant 32)

They see the interpreter as you be quiet and translate what I say. (Participant 39)

The findings are in line with those from Themes 1 and 2: interpreters tend to be seen as machines unaffected by external elements. Importantly, this mechanistic view of interpreting is identified as a barrier to interpreter–provider collaboration.

Discussion

The article has examined the intersections between interpreters’ invisibility, communication challenges and interpreter–provider collaboration in telephone interpreting in Australian healthcare settings. The key findings of this research can be summarised as follows.

First, the apparent view of interpreters as translation machines, as reported and perceived by participants in telephone-interpreted encounters, is identified as the most formidable challenge to telephone interpreting. Importantly, this highlights the fact that telephone interpreting problems are not merely technical issues but can be traced to deeper problems of awareness. While there is a growing body of research on the mechanical view of interpreting among healthcare professionals (e.g. Chernyshova & Ticca, 2020; Miller et al., 2005; Sleptsova et al., 2014), the ways in which the perception of ‘interpreters as translation machines’ impacts inter-party collaboration in the case of telephone interpreting have hardly been examined. The present study provides a rare glimpse into this under-researched problem from interpreters’ perspectives and highlights the potential ramifications for effective patient care. The identified provider belief that a good interpreter should be able to interpret well without a briefing, without being able to hear properly and without any visual cues at all illustrates the extremely mechanistic view of interpreting in which the concept of inter-party collaboration is not relevant at all.

Second, power differentials embedded in healthcare interpreting play a significant part in the strengthening of the provider view on interpreting as translation machines. In hierarchical healthcare systems, power often works in favour of doctors, rendering interpreters’ presence and work invisible (Drennan & Swartz, 2002; Hsieh, 2016). In such an environment, interpreters sometimes remain passive by trying to meet institutional expectations, rather than making themselves visible (Leanza, 2005; Major & Napier, 2019). This tendency is observed among the participants as well. As demonstrated, interpreters were worried about negative provider perceptions of their professional competence when requesting briefings due to concerns that such requests may lead clinicians to question their professional competence. Relatedly, interpreters reported having to be apologetic when requesting repetitions from healthcare providers to avoid frustrating them in the case of poor sound quality. The power hierarchies are found to discourage interpreters from actively seeking providers’ support which, in turn, may serve to reinforce the established mechanistic approaches to interpreting among healthcare providers. Unlike on-site interpreters who are said to have ‘invisible power’ (Davidson, 2001) or ‘interactional power’ (Mason & Ren, 2012) and may exercise their bilingual and bicultural skills to influence the course of communication, the different modality and working conditions of telephone interpreting hardly accord such power to telephone interpreters, revealing perhaps a bigger power imbalance between healthcare providers and interpreters.

Taken together, the findings of the present study suggest that there is a significant gap between the theoretical understanding of interpreting and actual practices. In the global interpreting literature, there has been a notable shift in the view of healthcare interpreters from passive language mediators to key communication participants who actively co-construct messages with primary parties (see the Theoretical Frames section). While this view of interpreters as co-participants discredits the traditional concept of interpreters’ invisibility, the findings demonstrate that this understanding of interpreters has yet to translate into practice, at least in the domain of telephone interpreting in Australian healthcare settings.

The current study, therefore, highlights an urgent need to raise awareness of telephone interpreting among healthcare providers, for the specific purpose of promoting effective interpreter–provider collaboration to achieve desirable health outcomes for linguistic minority populations. Considering the aforementioned issue of power differentials, enhancing provider awareness of telephone interpreting is all the more important to ensuring successful telephone-interpreted encounters. Training of clinicians who work with interpreters is central to addressing health disparities which tend to affect minority populations disproportionately (Hilder et al., 2019). The findings reported here demonstrate clearly that training should not focus only on surface-level factors such as background noise and sound quality but must help clinicians to work through what the interpreting process involves and how the quality of interpreting can be enhanced through a collaborative approach. This ultimately helps to deliver better healthcare.

Conclusion

Telehealth is regarded as the future of medicine, and telephone interpreting represents the most common medium to deliver healthcare service to the marginalised. Considering the rising global demand for telephone interpreting following the outbreak of COVID-19, it is important to improve the quality of telephone interpreting to ensure health equity for disenfranchised groups. Issues with telephone interpreting have been predominantly approached as technical issues, but the present research lends weight to the argument that telephone interpreting problems are fundamentally awareness problems. Improving provider awareness of the human element that underpins professional healthcare interpreting is, therefore, suggested as the most important solution to telephone interpreting challenges, and training of healthcare providers on interpreting is urgently required.

Finally, there is a need for interdisciplinary approaches to telephone interpreting issues. Challenges relating to telephone interpreting have been predominantly investigated by interpreting scholars, and there are very few studies on telephone interpreting from a clinician’s perspective. How, for instance, is the quality of clinical communication between clinicians and patients affected by different models of interpreter involvement during the consultation? Would patient care be impacted more in the case of telephone interpreting, compared with face-to-face interpreting, considering the limited opportunities for telephone interpreters to be culturally engaged in the communication process? Another area for interdisciplinary inquiry would be the investigation of healthcare provider perspectives on telephone interpreting. Such information has the potential to impact the quality and safety of patient care. Interdisciplinary approaches are, therefore, needed in order to further identify areas of improvement to enhance the quality of telephone interpreting. Considering the global impact of COVID-19 and growing use of telephone interpreting, this article has relevance not only to the Australian context but to the world.

ORCID iD

Jinhyun Cho https://orcid.org/0000-0002-4185-0592