The Dynamics of Miscommunication in

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/270791438
What if the diagnosis was tongue-tied? Linguistic ethnography in a

multilingual hospital Emergency Department
Conference Paper · September 2014
CITATIONS READS
0 149
1 author:
Antoon Cox
Vrije Universiteit Brussel (VUB), King's College London (KCL)
21 PUBLICATIONS 128 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Emergency Department Communication View project
All content following this page was uploaded by Antoon Cox on 18 April 2020.
The user has requested enhancement of the downloaded file.

Thesis submitted in fulfillment of the requirements
for the degree of Doctor of Applied Language Studies
THE DYNAMICS
OF (MIS)COMMUNICATION
IN LANGUAGE DISCORDANT
MULTI-PARTY CONSULTATIONS
IN THE EMERGENCY DEPARTMENT
ANTOON COX
2017
Supervisors:
Lead supervisor
Prof. Dr. Philippe Humblé
Vrije Universiteit Brussel, Faculty of Arts and Philosophy
Co-supervisor
Prof. Dr. Shuangyu Li
King's College London, School of Medical Education
Please cite this dissertation as: Cox, A. (2017) The dynamics of

(mis)communication in language discordant multi-party consultations in the
Emergency Department. Zelzate: University Press. ISBN 9789461975515. 292 pp.
THE DYNAMICS
OF (MIS)COMMUNICATION
IN LANGUAGE DISCORDANT
MULTI-PARTY CONSULTATIONS
IN THE EMERGENCY DEPARTMENT
ANTOON COX
2017
Supervisors:
Prof. Dr. Philippe Humblé (Vrije Universiteit Brussel, lead supervisor)
Prof. Dr. Shuangyu Li (King’s College London, co-supervisor)
Jury:
Prof. Dr. Katrijn Maryns (UGent)
Prof. Dr. Raquel Lázaro Gutiérrez (Universidad de Alcalá)
Prof. Dr. Luc Huygens (Vrije Universiteit Brussel)
Prof. Dr. Roel Coesemans (Vrije Universiteit Brussel)
Dr. Koen Kerremans (Vrije Universiteit Brussel)
Chair:
Prof. Dr. Rik Vosters (Vrije Universiteit Brussel)
"Screw it, let's do it"
(Richard Branson)
To my beloved wife Anneleen

i
Table of Contents
Table of Contents ........................................................................................................................ i

List of Tables ............................................................................................................................. v
List of Figures ........................................................................................................................... vi
List of Excerpts ........................................................................................................................vii
Acknowledgements ................................................................................................................ viii
1 Introduction ........................................................................................................................ 1
1.1 Motivation for this PhD ............................................................................................. 1
1.2 The Emergency Department as a research setting ..................................................... 4
1.3 Aim and research questions ....................................................................................... 5
1.4 Structure of this PhD.................................................................................................. 6
2 Miscommunication in medical settings ............................................................................. 8
2.1 A brief historic overview of research on doctor-patient communication .................. 8
2.2 Language and social interaction research on doctor-patient communication .......... 10
2.3 LSI research on language-discordant medical interactions ..................................... 14
2.4 Research on language barriers in the ED ................................................................. 15
2.5 Addressing language discordance in a medical setting............................................ 17
3 Research methodology ..................................................................................................... 21
3.1 Worldview and strategies of inquiry ........................................................................ 21
3.2 Linguistic ethnography ............................................................................................ 22
3.3 Linking discourse to medical practice: an interdisciplinary approach..................... 23
3.4 A case study approach ............................................................................................. 25
4 Conceptual framework ..................................................................................................... 27
4.1 Defining miscommunication .................................................................................... 27
4.2 Developing a conceptual framework ....................................................................... 28
4.3 Levels of miscommunication ................................................................................... 29
4.4 Sources of miscommunication ................................................................................. 32
4.4.1 Language barrier effects .................................................................................... 32
4.4.2 Violation of interactional rules .......................................................................... 33
4.4.3 Interpreting errors .............................................................................................. 34
4.4.4 Lack of background information or pretextuality .............................................. 35
4.4.5 Problems of role dynamics................................................................................. 36
4.4.6 Lack of engagement in the conversation ........................................................... 37
ii
4.4.7 Misalignment of frames ..................................................................................... 38

4.4.8 Misalignment of goals........................................................................................ 39
4.5 Prevention, accommodation, and repair................................................................... 40
4.5.1 Generic communication strategies in medical interviews ................................. 40
4.5.2 Communication strategies in a second language ............................................... 42
4.5.3 Linguistic accommodation ................................................................................. 44
4.5.4 Non-verbal communication ............................................................................... 45
4.5.5 Meta-communication ......................................................................................... 46
5 Description of research procedure ................................................................................... 47
5.1 Data collection ......................................................................................................... 47
5.1.1 Context of the data collection ............................................................................ 47
5.1.2 Entering the field................................................................................................ 49
5.1.3 Obtaining informed consent ............................................................................... 50
5.1.4 Enabling research by reassuring patients ........................................................... 53
5.1.5 Collecting “broad” contextual elements ............................................................ 54
5.1.6 Participants as co-researchers ............................................................................ 57
5.2 Nature of the data ..................................................................................................... 58
5.2.1 The different components of the ED consultation ............................................. 58
5.2.2 The non-linear structure of the ED consultation ................................................ 62
5.2.3 The medical consultation as a multiparty interaction ........................................ 65
5.3 Transcription and translation ................................................................................... 65
5.4 Analysis of case studies ........................................................................................... 66
5.4.1 Structure of the analytical chapters .................................................................... 66
5.4.2 Brief presentation of the selected case studies ................................................... 67
5.4.3 Motivation of the selection of the cases ............................................................. 68
5.5 Transparency and credibility.................................................................................... 68
5.5.1 Transparency ...................................................................................................... 68
5.5.2 Credibility .......................................................................................................... 69
6 Case study: Midnight mastitis.......................................................................................... 70
6.1 Context of the interaction......................................................................................... 70
6.2 Analysis.................................................................................................................... 71
6.2.1 Presenting complaint .......................................................................................... 71
6.2.2 Timing of the pain .............................................................................................. 73
iii
6.2.3 Setting of the pain .............................................................................................. 78

6.2.4 Location of the pain ........................................................................................... 84
6.2.5 Diagnosis, treatment, and closing ...................................................................... 87
6.3 Discussion ................................................................................................................ 90
6.3.1 Levels of miscommunication ............................................................................. 90
6.3.2 Sources of miscommunication ........................................................................... 93
6.3.3 Prevention, accommodation, and repair............................................................. 96
6.4 Conclusion ............................................................................................................... 99
7 Case study: The patient with a kidney stone .................................................................. 101
7.1 Context of the interaction....................................................................................... 101
7.2 Analysis.................................................................................................................. 102
7.2.1 Presenting complaint ........................................................................................ 103
7.2.2 Onset of the pain: Is it sudden or gradual? ...................................................... 108
7.2.3 Remitting or exacerbating factors .................................................................... 110
7.2.4 Treatment negotiation and closing ................................................................... 112
7.3 Discussion .............................................................................................................. 118
7.3.1 Levels of miscommunication ........................................................................... 118
7.3.2 Sources of miscommunication ......................................................................... 120
7.3.3 Prevention, accommodation and repair............................................................ 124
7.4 Conclusion ............................................................................................................. 126
8 Case study: Tuberculosis and technology ...................................................................... 129
8.1 Context of the interaction....................................................................................... 129
8.2 Analysis.................................................................................................................. 131
8.2.1 Associated manifestations: sputum (without UDR) ........................................ 132
8.2.2 Physical examination: sputum sample (without UDR) .................................... 136
8.2.3 Associated manifestations: sputum (with UDR).............................................. 140
8.2.4 Associated manifestations: breathlessness (with UDR) .................................. 143
8.2.5 Referral (with UDR) ........................................................................................ 146
8.3 Discussion .............................................................................................................. 150
8.3.1 Levels of miscommunication ........................................................................... 150
8.3.2 Sources of miscommunication ......................................................................... 151
8.3.3 Prevention, accommodation and repair............................................................ 153
8.4 Conclusion ............................................................................................................. 156
iv
9 Conclusion ..................................................................................................................... 159

9.1 Reflection on findings ............................................................................................ 159
9.1.1 The nature of a language barrier and the “communicative swing” .................. 159
9.1.2 Lack of knowledge on the communicative context ......................................... 162
9.1.3 Exploiting the communicative repertoire......................................................... 163
9.1.4 Redundancy...................................................................................................... 165
9.1.5 Broader causes of miscommunication ............................................................. 166
9.1.6 Patient companions as ad hoc interpreters ....................................................... 168
9.1.7 Interactional conflicts of interest – opening Pandora’s box? ........................... 170
9.1.8 Actual understanding and perceived understanding ........................................ 170
9.2 Implications for research........................................................................................ 171
9.3 Implications for practice ........................................................................................ 173
9.3.1 Addressing language barriers in the ED .......................................................... 173
9.3.2 Working without professional interpreters ...................................................... 174
9.3.3 Implications for healthcare staff and interpreter training ................................ 175
9.3.4 Implications for organisational ED management ............................................ 176
9.4 Limitations of the research..................................................................................... 177
10 References ...................................................................................................................... 178
Annex 1. Theoretical foundations of Language and Social Interactions research .......... 198
Annex 2. Application form for ethics committee ............................................................ 210
Annex 3. Initial informed consent form .......................................................................... 217
Annex 4. Simplified informed consent form ................................................................... 221
Annex 5. Transcription symbols used in transcript analysis ........................................... 222
Annex 6. Full transcript case study Midnight mastitis .................................................... 223
Annex 7. Full transcript case study The patient with a kidney stone .............................. 243
Annex 8. Full transcript case study Tuberculosis and technology .................................. 256
v
List of Tables
Table 4.1: Integrative model of levels of analysis of miscommunication ............................... 31

Table 4.2: Common interpreting errors ................................................................................... 34
Table 4.3: Possible conflicts between patient and physician agendas in the ED ..................... 40
Table 4.4: Generic communication strategies in medical interviews ...................................... 41
Table 4.5: Inventory of communication strategies in a second language ................................ 42
Table 5.1: Involvement of communities of practice and of speech in data preparation .......... 66
Table 6.1: Languages used by the interactants in case study 1 ................................................ 70
Table 7.1: Languages used by the interactants in case study 2 .............................................. 101
Table 8.1: Languages used by the interactants in case study 3 .............................................. 129
Table 8.2: Symptoms of tuberculosis..................................................................................... 130
Table 8.3: MRC breathlessness scale used to assess the severity of breathlessness.............. 144
vi
List of Figures
Figure 3.1: The interconnection of worldviews, design, and research methods ...................... 22
Figure 5.1: Using the smartpen for note-taking ....................................................................... 54
Figure 5.2: Components of the medical consultation .............................................................. 58
Figure 5.3: The structure of a medical consultation ................................................................ 64
Figure 6.1: Components of the medical consultation .............................................................. 71
Figure 7.1: Typical body position for patients with a kidney stone....................................... 102
Figure 7.2: Components of the medical consultation ............................................................ 103
Figure 8.1: Screenshot of Universal Doctor module on respiratory system .......................... 130
Figure 8.2: Components of the medical consultation ............................................................ 132
Figure 8.3: Screenshot of Universal Doctor module on attributes of symptom “cough” ...... 142
Figure 8.4: Screenshot of Universal Doctor module on shortness of breath ......................... 145
vii
List of Excerpts
Excerpt 6.1: Presenting complaint ........................................................................................... 71

Excerpt 6.2: Timing of the pain (Part 1) .................................................................................. 74
Excerpt 6.3: Timing of the pain (Part 2) .................................................................................. 77
Excerpt 6.4: Setting of the pain (Part 1) .................................................................................. 79
Excerpt 6.5: Setting of the pain (Part 2) .................................................................................. 80
Excerpt 6.6: Location of the pain (Part 1)................................................................................ 84
Excerpt 6.7: Location of the pain (Part 2)................................................................................ 87
Excerpt 6.8: Diagnosis, treatment and closing......................................................................... 88
Excerpt 7.1: Presenting complaint ......................................................................................... 103
Excerpt 7.2: Quality of the pain: is it continuous or intermittent?......................................... 105
Excerpt 7.3: Onset of the pain: Is it sudden or gradual? ........................................................ 108
Excerpt 7.4: Remitting or exacerbating factors ..................................................................... 111
Excerpt 7.5: Treatment negotiation and closing .................................................................... 113
Excerpt 8.1: Associated manifestations: sputum (without UDR) .......................................... 133
Excerpt 8.2: Physical examination: sputum sample (without UDR) ..................................... 137
Excerpt 8.3: Associated manifestations: sputum (with UDR) ............................................... 140
Excerpt 8.4: Associated manifestations: breathlessness (with UDR) .................................... 144
Excerpt 8.5: Referral (with UDR) .......................................................................................... 147
viii
Acknowledgements
After more than five years of research, time has come to express my gratitude to everyone who
joined me on this life-changing journey and supported me in many different ways.
In 1998, I first visited the Department of Applied Linguistics at ErasmushogeschoolBrussel

during an Open Day. I was hesitating at that point between signing up for Japanese studies or
for interpreter studies with a combination of Spanish and English. The convivial and friendly
atmosphere of the department immediately convinced me to choose for the latter. Through the
invaluable teaching and support of professors such as Martine Baetens, Martine Goedefroy,
Anne Hoorelbeke, Femke Simonis, Jose Correa, Rita and Martina Temmerman and Jef
Vandebrouck, I obtained my Master’s degree in Applied Linguistics in 2004. I am also grateful
to Martine Mallefroy and Annie Leus, who have supported me during my early student years
and later as a researcher.
After some years of wandering as a journalist, film maker, teacher, and teacher trainer, I
returned to the Department of Applied Linguistics in 2010. Rita Temmerman, my former
Master's thesis supervisor, had hired me to join her in BRIDGE-IT, a European project on
intercultural communication. Via visits to Krakow, Perugia and Malta, she introduced me to
the world of academic research and to inspiring colleagues such as Hilde Van Schaeren,
Massimo Cimichella and Paul Muraille. Later on that year, I was granted the opportunity to
teach Spanish under Philippe Humblé’s wings.
In the summer of 2011, while I was holidaying in British Columbia, Rita skyped me to inquire
whether I would be willing to participate in BABELIRIS, a research project on intercultural
communication in the public hospital sector in Brussels. Colleague Ward Van de Velde and his
supervisor Marc van Campenhoudt from ULB would cover written communication, while we
at VUB would focus on oral communication. Martine Mallefroy had hinted at the idea that I
might be interested, and I owe her thanks for this.
I took the plunge. That was when the idea of pursuing a doctoral degree first came to my mind.
Later, Philippe Humblé agreed to be my supervisor. I will always be grateful to Rita, Philippe,
Willeke Deridder and my colleagues at the Department of Applied Linguistics for the precious
opportunities I was given, and for creating an enabling research environment.
As part of the research project, I first started to study how communication processes were
managed in some of the Brussels public hospitals in the presence of language barriers. The
intercultural mediators (Anne Étienne, Sylwia Szczepanska, Mihaela Vasile, Janna
Boulanovitch, Abdel Ouadi and Malika Bouzinab) took me along on their daily routine of
assisting patients and sorting out communication problems. Their professionalism and
hospitality provided an invaluable starting point for my PhD research.
Katrien Detroyer’s support was pivotal in the subsequent phase of the project. As a coordinator
of the IRIS hospitals language plan, she introduced me to Pierre Mols, head of the Emergency
ix
Department of CHU Saint Pierre, who allowed me to collect data in his department. In addition
to opening doors for me, Katrien provided invaluable advice, and practical and moral support
throughout the process of data collection. She also offered to facilitate the dissemination of my
research findings. In the Emergency Department, I could always count on my compagnon de
route Nicolas Dauby, with whom I teamed up to overcome practical stumble blocks
encountered during the data collection. Special thanks also go to Dominique Cerf, who
facilitated a three-month survey on language practices in the Emergency Department, and
Collette Wonner, Lucas Steverlynck, Anne Malcorps, Peter Rummens, Olivier Ferrali, Anne
Carlier and all the Emergency Department staff for their collaboration and help.
I’m also very conscious of the fact that this research would have not been possible if it were
not for all the patients, patient companions, nurses and doctors who allowed me to observe and
audio-record their interactions during consultations in the often-stressed context of the
Emergency Department.
The transcription and the translation of the recorded consultations were made possible by
Gulzarina Khan, Muhammad Sughis, Kamran Khan, Julio Jimenez, Sophie Segers, Mohamed
El-Madkouri Maataoui, Malika Bouzinab, Sana Makrache, Nele Van Holderbeke and
Agnieszka Wozniak.
Thanks to the support of Katrien Detroyer, Fleur Deboutte, Luda Maltseva, Pierre Mols,
Nicolas Dauby and Abdul Ouadi, I was able to set up an e-learning module “Pitfalls in
Intercultural Communication”, while Ward Van de Velde’s developed a specialised medical
translation memory. In recognition of our joint achievements, the BABELIRIS project received
in 2015 a Language Industry Award in the category of Best Language Project.
Special thanks also go out to my office mate Dany Bauwens for his continuous support, his
great sense of humour, and for the teaching opportunities he offered me at VUB and the ones
he facilitated for me at the Belgian Chamber of Translators and Interpreters and the Flemish
Centre for Community Interpreting. In a similar vein, I owe thanks to Pascal Rillof, Joost
Vaesen and Rudi Janssens for offering me various platforms to disseminate my research. I’m
also grateful to Luc Huygens for inviting me to be a member of the Working Group on
interculturality and diversity of the Brussels University Association (UAB), and to the other
members of the Working Group. I admire his commitment to fostering diversity; to bringing
people together and inspiring them to collaborate towards a shared purpose.
I am also indebted to Shuangyu Li, who agreed to be my co-supervisor, and convinced his
institution, King’s College London, to enrol me free of cost as a research student for more than
three years, granting me access to a fascinating and vibrant research environment in the area of
medical communication and linguistic ethnography. Thanks go out to Ben Rampton, who
invited me to participate in the seminars and data sessions of the Centre for Language,
Discourse & Communication. Judith and Stewart Sanson have been supporting me from the
very beginning in Brussels, where they tutored me for English, and many years later as hosts
x
in their cosy Wimbledon house. Thank you, Judith and Stewart, for the many great evenings,
English breakfasts, and other inspiring moments we have shared!
The International Association for Communication in Health Care has been decisive for my
research. A special mention goes to Olivier Nardi and the regroupement francophone of this
association. Olivier introduced me to Ellen Rosenberg and Yvan Leanza, with whom we
successfully applied for a grant that enabled us to pursue enriching collaborative research, and
to Marie-Thérèse Lussier and Claude Richard whose work was a great source of inspiration for
my PhD. I further wish to thank my fellow members of the EACH Special Interest Group
“Language and cultural discordance in healthcare communication”, notably Shuangyu Li,
Angela Rowlands, Jennifer Gerwing, Demi Krystallidou and Peter Pype, for the good times
and the fascinating discussions we have shared so far, and for the ones that are still to come.
I want to thank my supervisors Philippe Humblé and Shuangyu Li and the members of my jury,
Katrijn Maryns, Raquel Lázaro Gutiérrez, Luc Huygens, Roel Coesemans, Koen Kerremans,
and Rik Vosters for inspiring me, supporting me, and believing in me. Katrijn, I am grateful
for all the time you took to read my PhD and the detailed and invaluable comments you have
provided. Koen, you stood up for me when it mattered. Rik, you were vital in getting my viva
on rails. Thanks so much!
My PhD was financially supported by the Brussels Institute for Research and Innovation
(INNOVIRIS), VUB and the Research Foundation Flanders (FWO). Obtaining this funding
was made possible thanks to the IRIS public hospital network, Rita Temmerman, Marc Van
Campenhoudt, Ward Van de Velde, Nick Ressman, Lotte Clijsters, Katrien Detroyer, Freddy
Iemants, Étienne Wéry, Philippe Humblé, Luc Huygens, Ellen Rosenberg, Yvan Leanza and
Willeke Deridder. Many thanks to all of you!
Finally, I want to thank my parents Christine and Yvan, for having always believed in me. The
same counts for my sister Clara and my brothers Jan and Dries. I am also grateful to my parents-
in-law Luc and Elly, for having conceived Anneleen and their continued support.
Many thanks go out to my friends, who have always been there for me and cheered me on along
the way; and to Radio Scorpio in Leuven, for providing us with an environment for creative
decompression at our weekly show “Radio Spoetnik”.
Starting this PhD would not have been possible without the Eureka school, where my dear
friends and I had a great time. The invaluable support of Anny Cooreman, Hedwig van den
Bosch, Marleen Bringmans, Nadira Lazreg, Ron and Pam Warren, Mattias Hoebeke, Xavier
Nicaise and many others helped me to obtain a secondary school degree.
Finally, there is my lovely wife Anneleen. In addition to providing me with moral support, she
also proofread my whole PhD. She helped me to structure my thoughts and stood by me in
difficult times. I would marry her over and over again!
Introduction 1
1 Introduction
1.1 Motivation for this PhD
“Emergency medicine is largely a communicative activity, and medical

mishaps that occur in this context are too often the result of vulnerable
communication processes.” (Eisenberg et al. 2005, 390)
International migration is at an all-time high. The number of international migrants residing in

Belgium soared from a bit over 850 thousand in 2000 to almost 1.4 million in 2015 (World
Bank 2016). Over the same period, the number of global migrants grew from 172 million to
243 million. A major implication of these expanding migration patterns is the increasing
diversity observed in hospitals in metropolitan areas around the world. Patients of foreign
descent are often overrepresented in hospital-based care, which means that hospital staff around
the world are increasingly confronted with the challenge of communicating with patients with
whom they do not share a common language. While this generates particular challenges for
quality of and access to care in hospitals (Scheeres et al. 2008), most of the existing research
on clinical communication has focused on primary care rather than secondary, hospital-based
care.
A proliferating literature shows that the lack of a common language between a patient and a
doctor is a major cause of health disparities in primary care (Divi et al. 2007; Karliner et al.
2012; Schillinger and Chen 2004). Language barriers may for instance lead to erroneous
diagnosis, incorrect medication intake, or poor follow-up (e.g. missed appointments). Watt
(2008) argues that 80 percent of a medical diagnosis depends on oral communication.
Compared to primary care, oral communication in the Emergency Department (ED) plays an
equally, if not more, important role. Past research has shown that medical errors in the ED often
result from poor communication (Eisenberg et al. 2005) and that language barriers are a major
obstacle to proper history-taking in the ED (Burley 2011).
Specific conditions for communication are very different in the ED than in primary care due to
time pressure, potential distraction resulting from long and tiring shifts, the sense of urgency,
and lack of prior information on patients (Chisholm et al. 2001; Knopp et al. 1996). The five
key components of optimal doctor-patient communication (establishing rapport with the
patient, gathering and giving information, providing comfort and collaboration) often need to
be performed simultaneously in the ED, and communication is often interrupted to answer
phone calls and attend to other problems as the physician is treating more than one patient at
the same time (Chisholm et al. 2001; Knopp et al. 1996). These same conditions also make the
consultation of professional medical interpreters more complex in the ED than in primary care.
The existing literature on language barriers in the ED has focused on measuring its impact on
different types of health outcomes and has shown that language barriers lead to lower patient
comprehension of discharge instructions (Crane 1997; Carrasquillo et al. 1999a), reduce the
Introduction 2
number of follow-up appointments (Sarver and Baker 2000), and increase diagnostic
insecurity. The latter is evidenced by a higher rate of test orderings (e.g. blood sample tests, X-
rays), a higher number of hospital admissions, and an increased length of stay, resulting in
additional costs (Hampers et al. 1999; Garra et al. 2010; Goldman, Amin, and Macpherson
2006; Kalkan et al. 2013). Language barriers have also been shown to reduce patient
satisfaction with medical care (Gany et al. 2007; Carrasquillo et al. 1999a).
In general, the literature recommends the use of professional interpreters to overcome

challenges associated with language barriers in a healthcare setting (Ramirez, Engel, and Tang
2008). However, the utilisation of professional interpreters in the ED is often perceived as
challenging, due to organisational, time and financial constraints (Ramirez, Engel, and Tang
2008; O’Leary, Federico, and Hampers 2003), and the lack of prior information on the patient’s
language skills (Karliner, Pérez-Stable, and Gildengorin 2004). Moreover, medical staff may
not be sufficiently informed about the added value of professional interpreters as they are likely
to overestimate the patient’s own language skills (Elderkin-Thompson, Silver, and Waitzkin
2001). Patients may overestimate their own language skills as well (Zun, Sadoun, and Downey
2006). Finally, in some cases hospital staff is simply insufficiently aware of the available
professional interpreting services (Bonacruz Kazzi and Cooper 2003).
As a result, clinicians often rely on patient companions (persons accompanying patients who
are more proficient in the hospital’s language) who act as ad hoc interpreters or chance
interpreters ; or they use no interpreters at all (Meeuwesen and Ani 2011; Meyer, Pawlack, and
Ortun 2010; Li, Pearson, and Escott 2010; Cox and Lázaro Gutiérrez 2016). The use of non-
trained ad hoc interpreters is considered bad practice in the literature (Meyer, Pawlack, and
Ortun 2010). They are considered to make more errors than professional interpreters (Karliner
et al. 2007), often omitting crucial information at the “detriment of medical practice”
(Cambridge 1999, 201), but this frequently occurs unnoticed (Meyer, Pawlack, and Ortun 2010,
298). They are also more often involved in role conflicts (Boivin, Leanza, and Rosenberg 2013;
Rosenberg, Seller, and Leanza 2008; Hsieh 2016).
On the other hand, the use of ad hoc interpreters may also have some advantages. Family
interpreters can be a valuable resource of specific information on the patient, and they can
create trust between the clinician and the patient (Greenhalgh, Robb, and Scambler 2006;
Meyer, Pawlack, and Ortun 2010; Leanza et al. 2014; Rosenberg, Leanza, and Seller 2007).
Based on a review of the literature, Brisset et al. (2013, 138) advocate for more research on the
role of ad hoc interpreters; to further explore their advantages and limitations; and to pave the
way for doctors to be trained to work more effectively with interpreters.
The existing research on medical interpreters largely focuses on primary care (and not the ED),
health outcomes (not communication processes), professional interpreters (rather than ad hoc
interpreters) and bilingual contexts (notably English/Spanish in the US). Nevertheless, context
matters and depending on the context, different solutions may need to be applied. For example,
in a context where a single foreign language predominates (as is often the case in the US),
ensuring 24h presence of a professional interpreter (or availability of remote interpreters) may
Introduction 3
be the optimal solution. However, it is much more complex to ensure professional interpreting
services around the clock in ED’s where patients come from a “superdiverse” (Blommaert and
Rampton 2012) and rapidly changing pool of different linguistic backgrounds.
As Schwei et al. (2015) argue, it has by now been widely established that language barriers
have a negative impact on health services; but there is a need for further research on what can
be done to overcome these problems, and in particular in highly diverse multilingual contexts.
From a broader perspective, there is also a dearth of literature on the impact of patient
companions on healthcare communication in contexts where these do not necessarily have the
intention to act as interpreters. So far, research on linguistically diverse medical consultations
that are mediated by third parties has mainly focused on these third parties as interpreters, with
less regard to the other roles they may take up. Besides their potential role of interpreter, there
is a range of different roles that can be taken up by patient companions, such as an advocate,
spokesperson or provider of moral support and comfort, each with its own implications. To
better understand these implications, more research is needed on the dynamics of
communication processes in linguistically diverse multiparty interactions.
Extracting information via patient companions, who can be strongly emotionally involved with
the patient; and/or not very familiar with the hospital’s official language either, and getting
messages across through them, especially under time constraints and lack of historical
information on patients, may require special skills from hospital staff. This calls for the
development of specific training material in the field of medical education that allows hospital
staff to become familiar with the importance of language in clinical practice; to recognise the
potential hazards of working with ad hoc interpreters, to learn how to work with patient
companions more effectively; and to recognise the dynamics of miscommunication when it
arises (Brindley et al. 2014; Karliner et al. 2007; Karliner, Pérez-Stable, and Gildengorin 2004;
Diamond and Jacobs 2010; Leanza, Boivin, and Rosenberg 2010). To develop such training
material, more research is needed on this topic. A related question which so far remains
underresearched is to what extent multilingual phrasebooks or IT tools can be used to address
language barriers in a highly diverse context.
From a methodological perspective, there is still scope for strengthening research tools to
address these issues. Very few researchers have studied in detail communication processes in
a multilingual context based on audio-transcripts of patient-clinician encounters in the ED. 1
Nevertheless, such micro-level research could contribute a lot to the existing gaps in our
understanding of how misunderstandings arise and how they can be addressed. Studies that
exist often focus on a relatively narrow aspect of the communication process. In this context,
Brisset et al. (2013, 138) conclude from a review of the existing literature on healthcare
interpreting that research “should […] take the complex nature of these consultations into
1
An exception is Flores et al. (2012) who compared the contents and correctness of ad hoc versus professional
interpreting based on discourse analysis. An earlier study by Farmer et al. (2006) described patient-caregiver
interaction in fairly great detail as well. However, the data collection for this study was limited to written accounts
and was as such not able to take note of physical space and gestures.
Introduction 4
greater account. The themes that emerged […] are clearly interrelated: they should not be
treated separately. Correlating interpreters’ roles and communication characteristics […]
would give us a better understanding of the complex reality of interpretation and how relational
issues affect its impact on patient care and health.” A holistic approach to these aspects calls
for interdisciplinary research, involving collaborations between applied linguists, medical
practitioners, foreign language specialists with a medical background, and others. 2
1.2 The Emergency Department as a research setting

Emergency departments (EDs) are a unique subculture within medicine (Person, Spiva, and
Hart 2012, 1), characterised by uncertainty, open-endedness and multiplicity, and time pressure
(Eisenberg et al. 2005; Knopp et al. 1996; Chisholm et al. 2001; Engel et al. 2010; Slade,
Manidis, et al. 2015). Doctors operate in an uncertain environment: they have no prior
information on which patients they will see, what their social or medical background is, and
what languages they know. Open-endedness implies that ED clinicians cannot control their
workload, as they do not know upfront how many patients will come, and when they will come.
Multiplicity refers to the fact that clinicians are typically seeing and monitoring several patients
at a time, and ED consultations are often distracted by internal phone calls (Chisholm et al.
2001, 148).
As ED services are typically characterised by overcrowding, and long and tiring clinician work
shifts, various components of the usual patient-doctor communication process (establishing
rapport with the patient, gathering and giving information, providing comfort and
collaboration) are often performed simultaneously (Knopp et al. 1996, 1066–67). 3 Patients tend
to be stressed and in pain, and the venue of communication is often noisy and lacking privacy
(Knopp et al. 1996, 1066). Due to these complex working conditions, some have argued that
the achievement of proper doctor-patient relationships in the ED is elusive, encouraging
doctors to prioritise the medical aspects of the encounter, with less attention to relationship
building with patients. Patients, on the other hand, are typically anxious and expect physicians
to be empathic. This often leads to a discrepancy at the level of expectations between doctors
and patients (Lin, Hsu, and Chong 2008; Lachance 2016).
Patients in the ED often find themselves in a special psychological state. More than in other
types of care patients in the ED are likely to feel very insecure and anxious (Wagley and
Newton 2010). Their genuine pain experience may be exacerbated by a hyper-focus on their
symptoms (Lachance 2016). A further complication is that the more anxious patients are, the
less likely they are to express this anxiety verbally (Lachance 2016; Back and Baile 2003).
Their stress experience can be reinforced by the communication problems in language-
discordant ED consultations.
2
Such as for example sociologists, psychologists, …
3
This is true for consultations in general, but even more so in the ED due to the specific conditions in which the
department operates.
Introduction 5
The ED does not only provide a very relevant context to examine communication under
conditions of distress and urgency; it also acts as a meeting point of medicine and society. The
ED can be considered as a super diverse ‘crucible of social experimentation’ (Levitt Center
2016). It serves a dual role, both as provider of emergency care for the critically injured as well
as a provider of basic care for the socially disadvantaged. This latter group includes those who
have limited access to other sources of medical care (such as primary care) because they are
new in the community. People in such situations often call on the most visible and immediate
means of entry into medical treatment, either by calling the emergency services or by going
directly to the ED. Therefore, the ED, perhaps more than any other niche of medical service
providers, is faced with an enormous diversity of languages in their daily practice.
1.3 Aim and research questions

The aim of this PhD is to describe the dynamics of (mis)communication, focusing in particular
on its nature and origins and the strategies used to prevent and overcome miscommunication
in linguistically discordant multiparty ED consultations. This implies the following specific
research questions: Which instances of miscommunication can be identified in a doctor-patient
consultation? What are the characteristics of these instances? How did they arise; can their
origins be traced? Is the miscommunication detected by the participants in the interaction?
Which efforts are undertaken by the participants to prevent or repair miscommunication, or,
conversely, to achieve understanding? Are these successful? Does the doctor eventually
succeed in achieving his communicative medical goals?
The study’s broader objective is to contribute to addressing problems of unequal access to

health services arising from communication problems, notably in the case of language barriers.
The underlying assumption is that “a problem well put, is half solved” (Dewey 1939, 108):
before being able to devise appropriate solutions, one needs to understand first how instances
of miscommunication arise, and what their implications are for the clinical outcome of a
consultation.
The research setting of this PhD is novel in that few existing studies have addressed
communication processes across a language barrier in the Emergency Department, as it is a
space that is not easily accessible (among other reasons due to ethical regulations, see Section
5.1). Nevertheless, it is also a context where problems of service quality caused by language
barriers are likely to be serious, given the time pressure and the lack of prior information on
patients, often exacerbated by tiredness, patient anxiety, and stress.
More specifically, my analysis considers multiparty interactions. So far, multiparty medical

consultations remain understudied in the medical communication literature (Laidsaar-Powell
et al. 2013; Clayman and Morris 2013), in spite of their relatively high incidence (Wolff and
Roter 2011). Within the domain of interpreting studies, analyses of triadic medical interactions
are more common, but these focus more often on interventions by professional interpreters,
Introduction 6
rather than by patient companions. In addition, my analysis considers that patient companions
tend to take up a broader role than merely acting as an ad hoc interpreter. 4
In addition, the “superdiverse” context considered in this study is quite exceptional in the
existing literature. Carrying out research in super diverse contexts is challenging, as it requires
informed consent procedures to be developed and observations to be analysed in multiple
foreign languages. Tackling language barriers in such a context is more challenging than in a
bilingual context. As no single strategy may be available to fully overcome all challenges, a
combination of different strategies may be required.
This PhD research looks at a unique dataset of audio-recorded transcripts of medical

consultations, complemented by relevant information on contextual elements such as physical
space, gestures, events occurring outside of the consultation and afterthoughts of the involved
physician based on walk-and-talk interviews in between the consultations. In comparison with
most other studies in this field, this represents a particularly rich source of data for the analysis.
Second, the study integrates methods from different angles and disciplines, combining applied
linguistics and medical perspectives, to assess the implications of the dynamics of
communication for the clinical outcome of a linguistically diverse consultation. It does not
restrict itself to specific linguistic aspects of the communication process (such as interpreter
behaviour, non-verbal communication, the use of physical space and others) but rather
considers all these factors in combination with contextual elements in the pursuit of specific
well-defined communicative purposes that are characteristic of the medical consultation. The
analysis explores the incidence of miscommunication, its potential causes, its characteristics,
the strategies used by participants to overcome it, and determines whether the communicative
purpose is reached or not. The multitude of elements considered calls for the adoption of a
relatively open methodological stance.
Third, in order to present the analysis of this multitude of different elements in a structured
way, a conceptual framework is developed that considers different levels of miscommunication
(drawing heavily on earlier work by Coupland et al. (1991)); an extensive range of different
potential causes of miscommunication; and a set of possible tools and strategies to prevent,
accommodate and/or repair instances of miscommunication.
1.4 Structure of this PhD

This PhD study is structured as follows. Chapter 2 reviews the existing literature on doctor-
patient communication within the realm of language and social interaction research, with a
special focus on settings with language barriers. It then adopts a broader perspective to review
the existing literature on language barriers in the ED, which has as yet mostly focused on
4
This is why the title of this study deliberately refers to “language discordant multiparty” interactions rather than
to “ad hoc interpreted” interactions.
Introduction 7
outcomes rather than on the communication process through which those outcomes are
obtained.
Chapter 3 explains the open methodological stance and interdisciplinary nature of my analysis,
which draws on different subdisciplines within the domain of language and social interaction
research, and integrates some key medical perspectives as well. Chapter 4 presents the
conceptual framework that is guiding my analysis to define and characterize
miscommunication, covering the different levels of miscommunication, the origins of
incidences of miscommunication, and the strategies used by participants in an interaction to
prevent, accommodate or repair miscommunication. Chapter 5 describes the research
procedure used for this study. It explains how the data were collected and which challenges
were encountered in the process. Next, it discusses the broad structure of the data, which is to
a large extent driven by the typical structure of a medical consultation, and how the data were
prepared and analyzed.
Chapters 5.4.36, 6 and 1 present the main analysis of this study through three illustrative case
studies. Chapter 9 concludes by reflecting on the main findings of this PhD and the ensuing
implications for practice and further research.
Miscommunication in medical settings 8
2 Miscommunication in medical settings
This chapter discusses in more detail findings from past studies that are relevant to my research,
including research on doctor-patient communication, first from a very broad perspective, and
then zooming in on various aspects of this literature: studies in the research domain of language
and social interaction, and studies focusing on language barriers in medical interactions in
general and in the emergency department in particular. I also briefly review the literature on
how language barriers are typicaly addressed in medical settings.
2.1 A brief historic overview of research on doctor-patient communication

It is widely agreed and officially recognised by the world’s leading journals and medical
schools that what doctors and patients say, and how they say it, has a major effect on the welfare
of patients and the organisation of healthcare services (Robinson 2008). Research on this matter
has a long history: interest in the relation between communication and medical care can be
traced back as far as Hippocrates, almost 2500 years ago (West and Frankel 1991).
Research on doctor’s perceptions of patients took off in the 1960s and was primarily
ethnographic. This was at least partially attributable to technological limitations (Robinson
2008). A seminal study by Becker et al. (1961) studied via observation and interviews how
young doctors behaved and how they looked at patients and how their behaviour and views
were influenced by the medical schools and institutions in which they studied and worked.
Miscommunication in medical settings became a focus of research in subsequent empirical
studies. Studies on patient satisfaction set in the UK (such as Cartwright 1964; Hugh-Jones,
Tanser, and Whitby 1964) as well as in the US (e.g. Korsch, Gozzi, and Francis 1968; Korsch
and Negrete 1972) concluded that patients were often unhappy with the communication flow
between them and the doctors, and that this reduced the quality of care (see West and Frankel
(1991) for a more detailed review). For instance, Cartwright (1964) found that patients who
had recently been discharged from the hospital reported trouble in obtaining information about
their illness. In a similar vein, Korsch and Negrete (1972) report that surveyed patients in a Los
Angeles Emergency Department felt that they did not understand what the doctor said and that
they were not sufficiently invited to explain their complaints.
The key message from this strand of literature in the 1960s and the 1970s was that medical
jargon used by doctors in their communication with patients had a negative impact on patient
understanding; that miscommunication was primarily a result of the lack of a common code or
shared knowledge between the doctor and the patient; and that doctor-patient communication
had to be as clear and understandable as possible (West and Frankel 1991, 170). However, not
everyone agreed on the conclusion that the use of medical jargon was to be avoided for acting
as a hindrance in the communication between doctors and patients. Even the study by Korsch
and Negrete (1972) already hinted at the fact that there were also some patients who were not
dissatisfied by the use of jargon. In contrast, they found the doctor’s use of jargon flattering
and impressive.
Most research in the 1960s and the 1970s approached doctor-patient communication primarily
from a transmission model perspective, in which information is sent in some form from a
sender (the doctor) to a receiver (the patient). A medical consultation was described as an
interaction where topics were mostly initiated by doctors (e.g. by asking questions) and patients
mostly responded to those. In such a model, the basic assumptions were that communication
was good if clear and understandable, and that instances of miscommunication could be
attributed to characteristics of the involved individuals, be it the senders or the receivers.
However, the idea that communication was good if it was clear and understandable to the
patient was challenged by research on communication with cancer patients (West and Frankel
1991). Based on a review of the literature, McIntosh (1974) concluded that, while most patients
wanted information on their diagnosis, doctors often preferred not to tell them. In fact, doctors
were reluctant to disclose sensitive information and bad news, not only because they feared
negative emotional reactions from patients unable to cope with such news, but also because of
the high degree of uncertainty which tends to surround the diagnosis, treatment and control of
many serious diseases, and the unwillingness to admit to being uncertain, for fear that the
patient would lose confidence in their ability to treat them (Glaser and Strauss 1965). Davis
(1960) found that physicians would deliberately hedge, avoid answering the question, and
couch bad news to the parents of children admitted to a children’s hospital. One of his
respondents put the reason behind doing so as follows: “We try not to tell [the parents] too
much. It is better to find out for themselves in a natural sort of way.” (Davis 1960, 44).
Consequently, a particular school of thought developed, advancing the argument that good and
thus clear communication is not always good for the patient (West and Frankel 1991, 171). Ley
and Spelman (1967) argued that patients did not want to have full information on their medical
conditions, and other scholars recommended telling patients only what they asked for, or what
would be needed to ensure their compliance to the proposed treatment (Shands et al. 1951;
Bard 1970). Sontag (1979) describes how at the time of her writing, doctors in France and Italy
were unlikely to communicate cancer diagnoses to patients. Instead, they would tell it to their
family, and leave it to their discretion whether or not to pass on this information to the patient.
A competing school of thought (that was popular particularly in Scandinavia) believed that the
patient should be told the truth, no matter what (Richards 1978, 265–66).
The consideration that miscommunication could be traced back to individual characteristics of

the sender and the receiver, sparked a literature discussing how to strengthen communication
skills of doctors with a view to avoiding miscommunication, on the one hand, and which patient
characteristics are conducive to miscommunication on the other hand (West and Frankel 1991).
As for doctors, interventions such as (psychological) selection tests and the incorporation of
communication skills training in medical curricula was debated and researched (Innes 1977).
As patients could obviously not be selected or trained for better communication, research tried
to describe the sociodemographic characteristics of patients that were more likely to be
involved in medical misunderstandings than others.
Some studies found that miscommunication was more likely to arise with less educated and
socio-economically weaker patients, as these were less knowledgeable on preventive
healthcare, had more difficulties with understanding medical terminology and were less likely
to listen to their doctor (see e.g. studies by Coope and Metcalfe 1979; Samora, Saunders, and
Larson 1961; Plaja, Cohen, and Samora 1968; Bochner 1983). Zola (1963) pointed at ethnicity
as a determinant of medical misunderstandings as he argued that patients from different
ethnicities tend to present their complaint differently. More particularly, in Zola’s study,
Italians were more likely to “overact” when presenting their complaint and to put strong
emphasis on the psychosocial problems associated with their physical medical conditions. This
would encourage doctors to recognise psychosocial factors more often as contributing to a
medical condition as compared to the case of patients from other ethnic groups (Irish and UK
origin) in the study.
Davis (1963) however challenged the position that patients’ weaker socioeconomic status
would in itself be a driver of miscommunication, as he argued that communication was not so
much driven by actual patient characteristics (such as socioeconomic status) as by the doctor’s
perception of the latter. Along similar lines, Roth (1963) found that the way patients presented
themselves in hospitals influenced the way they were given information by doctors. These
studies provided further support to the observation that doctor-patient communication was
more complex than what could be captured or described by a simple sender-receiver model.
The easier access to and increased availability of audio and video recording devices in the late
1970s and 1980s enabled researchers to record doctor-patient interactions. As a result, attention
was increasingly shifted to the processes of communication rather than outcomes alone:
researchers started to look at when, and under what conditions misunderstandings arise in order
to explore what can be done about it (Beckman, Kaplan, and Frankel 1989). Through studies
that analysed interactions based on naturally recorded data, doctor-patient communication
research gradually merged into the broader area of Language and Social Interaction research.
2.2 Language and social interaction research on doctor-patient communication

Language and Interaction (LSI) research brings together different research disciplines that
consider that language is an integral part of a social activity, rather than only a medium of
communication. 5 Within this strand of literature, some studies have focused on medical
settings. A starting point for language and social interaction research on doctor-patient
communication was the seminal study by Byrne and Long (1976), who analysed 2500 audio-
recorded GP consultations in New Zealand and the UK in search of recurrent patterns in
5
Appendix 1 provides an overview of the theoretical foundations of LSI research. Getting familiar with
methodological concepts in this domain was greatly facilitated by training materials such as Stef Slembrouck
(2017)’s online syllabus “What is meant by discourse analysis?”, Marc van Oostendorp (2015)’s online course
“Miracles of Human Language”, materials provided through “The Virtual Linguistics Campus” by Marburg
University (VLC 2017) and course materials from the Summer School “Ethnography, Language and
Communication” organised by King’s College London.
medical consultations. Based on this analysis, they described six main stages in the
consultation, notably the opening, in which the doctor creates a rapport with the patient, the
problem presentation phase, in which the main problem is elicited from the patient, the history
taking and physical examination, the delivery of the diagnosis, the proposition of possible
treatments and the closing. Later research inspired by this study showed that the way the doctor
organises these stages has an impact on the patient’s acceptance of treatment recommendations;
and this was a crucial insight for improving training of healthcare professionals (Mondada
2013).
Byrne and Long (1976) also observed another recurrent pattern in the consultations, notably
that if doctors left little silence or pauses between their own utterances to allow the patient to
speak, they were less successful in obtaining information from the patient. Moreover, they
noted that closed-ended questions (including yes or no questions) on the doctor’s behalf limited
the patients’ opportunity to talk; and that doctors who relied more on open-ended questions,
inviting the patient to elaborate and to give his own perspective, saw their patients less
frequently coming back. While Byrne and Long’s pioneering work was still rather doctor-
centred as it concentrated attention on the doctor’s communication characteristics, later
research analysed the structure of the consultation also focussing on the communicative styles
of the patient.
Studies by Greenfield et al. (1985) and by Roter (1977) tested patient education interventions
to strengthen patient involvement and participation in medical consultations, describing both
the ensuing processes of communication but also their outcomes. Greenfield et al. (1985) found
that patient education increased patient involvement, reduced the impact of disease on patients’
functional ability six to eight weeks after the medical appointment, but did not have a
significant impact on patient satisfaction. In Roter’s study, patient education also had a positive
impact on patient participation. However, she found that this reduced patient satisfaction with
the received care, and raised anger and anxiety levels experienced during the doctor-patient
interaction. Still, patient compliance with follow-up appointments increased, which could
suggest that patient participation was successful in reducing the occurrence of
misunderstandings. Roter hypothesised that the negative feelings elicited by stronger patient
involvement could be attributed to the fact that neither the patient, nor the doctor were prepared
or comfortable with an active patient role at the time of the survey. She was not able to establish
based on her research whether these feelings resulted from the doctor’s resistance to the
increased question-asking, or rather from increased awareness of information gaps. In her view,
these negative feelings were not necessarily to be avoided, as they could help to bring about
behavioural change towards the acceptance of stronger patient participation.
Heritage and Maynard, two sociologists specialised in the study of medical interactions, have
been promoting the use of conversation analysis to analyse doctor-patient interactions since the
1980s (see e.g. Heritage and Maynard 2006), to study issues such as the varying dynamics of
mutual understanding and the power dynamics between patients and doctors. They described
a medical consultation as an interactive process organised in turns, in which the doctor and the
patient alternately take the floor and as such jointly construct the interaction.
Conversation analysis has by now indeed become a popular methodology to study medical
interactions. For example, Robinson (2003) analysed problem-solving activities between the
patient and the doctor during acute care visits using conversation analysis. Focussing on the
organisation of the interaction, he concluded that doctors should allow for more time at the
beginning of a consultation, so that the patient can explain his or her concerns, and after having
delivered a diagnosis, give patients the opportunity to ask questions and share concerns rather
than moving directly to the treatment discussion.
Other studies accounted more explicitly for the context outside of the direct interaction.
Seminal work in this area was done by Goffman (1967), who observed doctor-patient
communication in psychiatric consultations. He concluded that in a consultation where only a
patient and a psychiatrist are present in the room, other powers may play a role as well. His
“telephone booth bias” theory posits that what patients say or do not say during a medical
consultation is influenced by others (Goffman 1967, 139). Patients were often referred to
psychiatry by family members who had their own view on the patient’s condition. Their views
on the patient’s condition, as well as their views on good or bad behaviour in society, would
act on the patient’s talk during the medical encounter in a similar way as “noise on a telephone
line”.
Tannen and Wallat (1987) analysed videotapes of triadic paediatric medical consultations that
were recorded for educational purposes. Building on Goffman (1981b)’s seminal work, they
noted that the paediatrician managed three different “interaction frames” during the considered
medical encounter. 6 They also illustrated how these frames interact with “knowledge
schemas”: structures of knowledge that shape participants’ expectations about events, settings,
and actors within a specific frame, and how sayings and doings should be interpreted within
this frame. They showed that differences between the doctor’s and the mother’s medical
knowledge schemas encouraged the patient’s mother to frequently interrupt the doctor’s talk
with questions, prompting her to shift frames. The frame shifts could be identified through
changes in footing, consisting of linguistic and paralinguistic cues, for example when the doctor
switches from medical language and a didactic tone to adult talk adjusted to the mother’s
knowledge schema. Tannen and Wallat (1987, 210) concluded that such intensive “juggling of
frames” between the mother, the child and the camera, and between the contents of the activity
and the associated registers or ways of speaking, made communication complicated and
cognitively and socially challenging for the doctor.
The field of social psychology has also contributed strongly to the study of medical interaction.
In his seminal book “The Discourse of Medicine: Dialectics of Medical Interviews”, Mishler
(1984) made a distinction between “the voice of the lifeworld”, that is, everyday life and
concerns of the patient, and “the voice of medicine”, representing the medical agenda and
6
In particular, a “social encounter frame”, in which the doctor entertained the mother and the child and ignored
the video camera; an “examination frame”, in which she examined the child and explained what she was doing
towards the video camera; and a “consultation frame”, in which she talked to the mother and ignored the child
and the videotape.
reasoning of the doctor, building on Silverman and Torode (1980)’s notion of “voices”,
comprising a set of assumptions about the relationship between appearance, reality and
language. Mishler then explored how these two voices interact and enter into conflict with each
other during a medical consultation. He observed that the “voice of the lifeworld” frequently
tries to interrupt the dominant “voice of medicine”, for example when a patient tries to bring
up additional personal concerns which may or may not be related to the problem for which he
is seeing the doctor. Faced with such interruptions, doctors can react in different ways: either
by opening discourse to the “voice of the lifeworld”; or by keeping discourse closed,
suppressing the voice of the lifeworld, and re-establishing dominance of the “voice of
medicine”.
As a professor in social psychology, Mishler argued that psychosocial topics are all too often
ignored by doctors because of their medical agenda, and he invited doctors to pay more
attention to the patient’s voice of the lifeworld in the medical consultation. Even though other
studies have shown that the distinction between these two worlds is not always clear cut (see
for instance Atkinson 1995), Mishler’s work has brought an important theoretical contribution
to the understanding of misunderstandings (Roberts 2006, 743).
Another contribution of social psychology research to the medical interaction literature was
provided by Street (1991). Building on communication accommodation theory, Street (1991)
described how doctors and patients use convergence and divergence strategies in medical
encounters. Convergence occurs when a doctor tries to adapt his communicative style to what
(s)he thinks will match with the patient’s style or mood. Divergence arises when a patient,
despite being asked for his/her opinion, keeps a low key, and leaves the talking to the doctor,
possibly because (s)he perceives him/herself as powerless. Street also spoke of the high risk of
misperceived convergence, as a result of the multitude of different frames and knowledge
schemas that often come into play in a medical consultation (see also the discussion of related
work by Tannen and Wallat (1987) above).
Doctors and patients may have different feelings of distress and ideas about treatment, which
makes the medical consultation highly prone to misunderstanding. One example of
misperceived convergence in a medical interaction is when doctors believe that patients (or
their companions) have little understanding of medical terminology, and therefore use
simplified language that is far below the actual patients’ understanding (Street 1991). Another
example, which may occur in a language discordant interaction, is when doctors start to speak
excessively loud to a non-native patient.
The view that verbal and non-verbal behaviour is used for convergence or divergence is related
to Goffman’s notions on frames and footing in the sense that participants to a conversation shift
frames by changing footing to reach convergence with their primary addressees.
Communication accommodation theory refines these ideas by exploring more extensively the
motivations of participants to an interaction to pursue convergence (or divergence).
2.3 LSI research on language-discordant medical interactions

Some studies within the LSI literature have focused specifically on language discordant
medical interactions. For example, studies by Roberts et al. (2004; 2005) applied an
interactional sociolinguistics approach to linguistically and culturally diverse general
practitioner consultations in South East London, and described the way in which non-native
patients linguistic (e.g. pronunciation, word stress, intonation) and broader communicative
behaviour (e.g. presentation of self) leads to misunderstandings, 7 reduces the orderliness of the
medical consultation, and increases interactional uncertainty.
Still, most research on language-discordant medical interactions that fall within the scope of
LSI research focus on possible interventions, such as professional or informal interpreters,
multilingual phrasebooks or phone or video interpreting (see e.g. a review by Roberts 2006).
A substantial literature has developed in the field of interpreting studies, focussing on the
deployment of professional interpreters. This literature draws on a variety of methodologies,
including sociolinguistics, the ethnography of communication and conversation analysis. In
what follows, a few examples are reviewed.
Wadensjö (1992, 1998) was one of the first researchers to study interpreter-mediated medical
consultations. She applied Goffman’s participation framework theory to tape-recorded
consultations complemented with ethnographic field notes. Based on her observations and
analysis, she dismissed the (at that time) traditional view that interpreters were to serve as
“invisible” language conduits, and not as active participants in the interaction (Reddy 1979).
She found that interpreters take an active role in the mediated consultation, including by
relaying (conveying meaning between the doctor and the patient, but not necessarily by always
literally translating utterances as if they were a “translation machine”) and co-ordinating (e.g.
by taking an active role in turn design, by signalling non-comprehension from either of the
parties, asking for additional clarifications or offering additional explanations).
This view aligns closely with Angelelli’s (2003) perspective on interpreters as “visible co-
participants” to an interaction. Angelelli (2004) documented her findings on the roles
professional medical interpreters take on within the hospital’s context, the societal context, and
relevant social factors, based on data which she collected while spending 22 months observing
in a Californian bilingual hospital. The role of the medical interpreter has been explored in a
broader set of sociolinguistic studies, including Leanza (2005), Greenhalgh et al. (2006), and
Krystallidou (2012, 2013, 2014). Greenhalgh et al (2006) identified a variety of roles an
interpreter may be expected to take, including as a translator, an interpersonal mediator, an
educator or the patient’s advocate. Leanza (2005) discussed how an interpreter can fulfil the
role of managing cultural differences in four ways: the system agent (who focusses on the
dominant culture and tries to deny or assimilate cultural differences), the community agent
(who is an advocate for the minority culture and considers it as equally valid as the majority
culture), the integration agent (who looks for a way “in between” that brings the minority and
7
See Chapter 0 in this dissertation for more details.
the majority closer to each other), and the linguistic agent (who attempts to maintain a neutral
position and not to intervene on any other level than that of language). Finally, Krystallidou
(2012, 2013, 2014) drew on Goffman (1981b)’s participating framework to explore how the
professional medical interpreter role is perceived and negotiated through verbal and non-verbal
elements (such as gaze and body orientation), and how it can contribute to a patient-centred
approach.
Other aspects of professional interpreter-mediated consultations have been studied by

sociolinguists, such as the way interpreters influence the power balance between doctors and
patients. Davidson (2000) observed that interpreters often support the doctor’s medical agenda
during the encounter, rather than the patient’s “voice of the lifeworld” (Mishler 1984). This
implies that interpreters sometimes reduce the quantity of information as to keep up the pace
of the medical interview. Davidson (2000)’s results contrasted with other studies (e.g. Kaufert
and Koolage 1984) which suggested that interpreters often take on the role of the patient’s
advocate. A possible explanation for these divergent results could lie in the macro-context, in
the way the hospital (and the interpreter service) is organised.
Other LSI approaches have been used as well in research on language discordant medical
interactions. Li (2011, 2013, 2015) drew on conversation analysis to study turn-taking in
interpreter-mediated consultations, and how turn design is co-constructed by the doctor, the
interpreter and the patient. Finally, Collins and Slembrouck (2006) used ethnographic research
to document how a multilingual phrase book was used in medical consultations in a
linguistically and culturally diverse community health centre in Belgium, and how doctors felt
about this.
2.4 Research on language barriers in the ED

Most of the existing literature on language barriers in the ED pertains to the medical literature
and has focussed on their impact on the outcomes of consultations in the ED. In what follows,
the findings of these studies are briefly reviewed.
Research based on US data found that a language barrier leads to increased diagnostic
insecurity, resulting in a higher number of test orderings (e.g. blood sample tests, X-rays),
increased length of stay in the hospital, a higher number of hospitalisations, and in general
additional expenses (Hampers et al. 1999; Garra et al. 2010). Along the same lines, Goldman
et al. (2006, based on records from a Canadian paediatric ED) and Mahmoud et al. (2013,
based on records from an Australian ED) found that language barriers are significantly
associated with variation in length of stay.
On the other hand, Waxman and Levitt (2000), Cossey et al. (2012) and Wallbrecht et al. (2014)
did not find a significant difference in length of stay between patients with limited English
proficiency and English-speaking patients. In the study by Waxman and Levitt (2000), an
interpreter was used whenever there was a language barrier. The interpreter might have
eliminated the language barrier, and this could explain why no differences were found in length
of stay. On the other hand, Wallbrecht et al. (2014) found that using an interpreter actually
increased the length of stay.
Language barriers may have an impact on the comprehension of follow-up instructions. Crane
(1997) investigated differences in comprehension between English and Spanish-speaking
patients, and concluded that the latter scored significantly lower on the comprehension of
discharge instructions than the former. In a quantitative cohort study, Sarver and Baker (2000)
found that Spanish-speaking patients are less likely to receive follow-up appointments than
patients without language problems. However, they did not find a difference in compliance
rates for those who received follow-up appointments.
A study by Carrasquillo et al. (1999b) found that non-English-speaking patients reported more
problems with regard to the discussion of medical conditions, explanations of reasons for and
results of diagnostic testing, and discharge instructions. Regarding the follow-up, Carrasquillo
et al. (1999b) found that non-English-speaking patients understood less well under which signs
they should return to the ED. Also, these patients reported a lack of proper instruction and
understanding of the dose and possible side effects of medication.
Finally, research has explored the impact of language barriers on patient satisfaction.
Carrasquillo et al. (1999b) surveyed non-English speaking (Hispanic) patients’ satisfaction
with ED-provided services. Their study revealed that non-English-speaking patients were less
satisfied and half as likely as English-speaking patients to return to the same ED in the future.
Gany et al. (2007) surveyed Cantonese, Mandarin and Spanish-speaking patients in primary
care and the ED of a New York hospital and found that patients who shared a language with
their physician were more satisfied with the provided services than other patients.
Research that explores the microstructures of physician–patient ED interactions involving

language barriers (David et al. 2006, 357) to assess “actual behaviour” (Priebe et al. 2011, 11)
is much scarcer. A possible contributing factor is the challenge to gain access to the ED
environment as a setting for research (see Section 5.1). The only exception I know of is a
quantitative study by Flores et al. (2012) on the incidence of interpretation errors (such as
omissions, additions, substitutions, editorialisation and false fluency, see Section 4.4.3 for
more details) in cross-linguistic paediatric ED interactions. They found that, although many
errors were encountered in all interactions, the use of professional interpreters led to fewer
errors of clinical consequence (which altered the medical history of the presented illness,
diagnostic or therapeutic interventions, parental understanding of the child’s medical condition,
and/or follow up plans) as compared to intermediation by ad hoc interpreters, or no interpreters
at all.
2.5 Addressing language discordance in a medical setting

A number of studies have addressed the question of how to best tackle language barriers in a
medical setting. The available literature finds that optimal results are achieved when hospital
staff is multilingual, such that the clinician can communicate with the patient in his/her own
language (Gany et al. 2007). Implementing this in a linguistically highly diverse environment
seems, however, less feasible. Therefore, the literature recommends the use of professional
interpreters (Ramirez, Engel, and Tang 2008).
What distinguishes professional interpreters from other interpreters is the training they have
undergone (Cox and Lázaro Gutiérrez 2016). They not only need a deep knowledge of the
languages they interpret into and from, but also of relevant cultural concepts and expressions
(Valero-Garcés 2006) and of institutions. Healthcare interpreters need training on how to
transmit sensitive messages, and on how to manage their own emotions in tense and stressful
situations, for instance to tell patients that they have cancer or are terminally ill (Phelan 2001,
21). Finally, interpreters can be trained on how to act as a “cultural broker” and to intervene
proactively in situations where communication is hampered by cultural differences (Cox
2015; Kaufert and Koolage 1984; Verrept 2000).
Nevertheless, several challenges associated with the use of professional interpreters in the ED
imply that doctors often revert to using non-trained, ad hoc interpreters 8 and that professional
interpreters remain largely underutilised. According to Ramirez et al. (2008), professional
interpreters are often not called in because of perceived financial and time constraints, while
ad hoc interpreters are often readily and freely available. In a study of an urban paediatric
hospital in the US, O’ Leary et al. (2003) found that a major reason for medical staff not to call
in medical interpreters, even if they were present on-site, was that the process of calling and
locating them was perceived as too time consuming and cumbersome. The lack of prior
information on the patient’s language skills complicates this process (Karliner, Pérez-Stable,
and Gildengorin 2004).
Appropriate organisational arrangements are crucial for an adequate provision of professional

language services. For instance, if ED doctors, who are constantly running around performing
different tasks, do not have a contact list with the direct internal phone numbers of the on-site
interpreters readily available, it is hardly possible for them to take the time to find out who to
contact and how. Under time pressure, they tend to look for the closest available solution, which
are often family members, relatives, or colleagues, or not using interpreting services at all.
When professional interpreters are called in, they often spend a lot of time waiting, as the
physician is multitasking, monitoring test results, calling colleagues or seeing different
8
Ad hoc interpreters are persons who accompany the patient (such as family members, friends or bystanders) and
are more proficient in the hospital’s working language than the patient is. Hsieh (2016) refers to family and friends
acting as an interpreter as “family interpreters”. She considers the latter have a distinctive relationship with and
knowledge about patients; as opposed to “chance interpreters”, untrained bilingual bystanders who provide
interpreting services by chance. Patient accompaniment is most frequent in the case of young or elderly patients,
where relatives accompany patients even in language concordant contexts (Turabián and Pérez Franco 2015).
patients. When external professional interpreters are called in, this potentially results in a
relatively high cost due to the time spent waiting.
In some cases doctors are not aware of available professional interpreting services (see e.g.
Bonacruz et al. (2003) for a study set in Australia), which underlines the importance of raising
awareness among hospital staff. Establishing a personal rapport between doctors and on-site
interpreters can encourage doctors to call in an interpreter when needed.
Another reason why professional interpreters remain underutilised is that doctors often do not
realise their importance for the quality of the information exchange (Elderkin-Thompson,
Silver, and Waitzkin 2001). First, correctly assessing the patient’s and his/her companion’s
language skills is complex. Medical staff often overestimate the patient’s language skills, or
judge that ad hoc interpreters suffice for language mediation. Nevertheless, individuals who
seem fluent in daily small talk may still lack the proper repertoire for a medical consultation.
People learn the meaning of words and concepts according to the circumstances they have been
exposed to (Hymes 1974). Even if individuals seem to understand what is being said, they are
not necessarily able to interpret messages within the specific healthcare context that rely on a
body of specific background knowledge. In this context, Zun et al. (2006) found that patients,
doctors and nurses, significantly overestimated effective patient health literacy.
Companions can provide comfort and various types of support to patients, and contribute
positively to the information exchange (Rosland et al. 2011; Hubbard et al. 2010). If patients
are accompanied by family members, the latter may be able to provide valuable information
on the patient’s state of health and history (Lachance 2016; Meyer, Pawlack, and Ortun 2010;
Rosenberg, Seller, and Leanza 2008). Greenalgh et al. (2006) found that family interpreters felt
that they had an advantage over professional trained interpreters in that they could foster trust
between the patient and the doctor and that they were able to keep the consultation more
private.
Nevertheless, if a language-discordant interaction is mediated by a non-professional

interpreter, there is a high risk of interpreting errors (Flores et al. 2012; Karliner et al. 2007).
Based on observations in a paediatric ED, Flores et al. (2012) argue that ad hoc interpreters
more often “altered or potentially altered the history of present illness, medical history,
diagnostic or therapeutic interventions, parental understanding of the child’s medical condition,
and/or plans for future medical visits (including follow-up visits and specialty referrals)” than
professional interpreters. Typical mistakes consisted of omissions, adaptations and
editorialisation (with the interpreter giving his own view rather than translating) (Flores et al.
2012).
Interpreting errors (and in particular those that remain unnoticed) may lead to errors in the
diagnosis and the proposed treatment and/or protracted consultation time. Fagan et al. (2003)
compared the differences in patient’s length of stay in the hospital between cases where use
was made of on-site interpreters, telephone interpreting and ad hoc interpreting. Contrary to
the common perception that calling in professional interpreters costs a lot of time, they found
that the patient’s length of stay increased when ad hoc interpreting or telephone interpreting
was used instead.
According to Ho (2008), family interpreters do not encounter linguistic problems in every

context; depending on the context, patients may benefit from a deeper involvement of family
interpreters in the consultation. Ho (2008) added, however, that this does not justify not calling
in professional interpreters. Along the same lines, Gray et al. (2011) argued, based on research
in New-Zealand medical care, that in some situations, using family interpreters involved
minimal clinical risk; but in most other settings, clinicians acknowledged the extra clinical risk
associated with the use of family interpreters.
Several studies have pointed out that doctors are often unaware of the disturbances induced by
non-professional interpreters in medical consultations. For example Meyer et al. (2010, 298)
argued that “although interactions with ad hoc interpreters often appear to be fluent and
unproblematic at first glance, a closer look reveals significant miscommunication. However,
such miscommunication passes by unnoticed by the primary interlocutors.” In a similar vein,
Pöchhacker and Kadric (1999) concluded from an analysis of ad hoc interpreted mental health
consultations that doctors were often unaware of the loss of information that occurred, although
it resulted in a loss of quality and effectiveness of their service provision. A particular risk is
that if the clinician does not properly identify the validity of the different potential sources of
medical history at the outset of the consultation (as recommended by Bickley 2013), chance
interpreters may be mistaken for valid sources of medical history.
An additional complexity is that, as research in primary care has shown, doctors and patient
companions often have different views on the latter’s role (Boivin, Leanza, and Rosenberg
2013; Rosenberg, Seller, and Leanza 2008). Patient companions tend to assume a much broader
role than pure interpreters. Doctors also often consider the patient’s companion as a full partner
in the conversation and a potentially useful source on the medical background of the patient,
rather than as a mere ‘translation machine’. But at the same time they are wary of the
companion responding on the patient’s behalf and giving his/her own views on the patient’s
condition rather than the patient’s own view (Boivin, Leanza, and Rosenberg 2013).
In some cases, the presence of the patient’s companion makes it more difficult for the doctor
to get across to the patient and/or elicit information from the patient. Some studies have argued
that companions may hamper patient participation in the medical interview or the discussion
of sensitive issues (Shepherd, Tattersall, and Butow 2008; Barone, Yoels, and Clair 1999).
Aranguri et al. (2006) found that the translation of doctor or patient utterances by a non-
professional interpreter was associated with a significant loss of information and that the
amount of small talk, which tends to be crucial for the process of building rapport between the
patient and the doctor, was reduced. Bührig and Meyer (2004) argued that non-professional
interpreters are more likely to jeopardise, rather than facilitate, consent negotiation by not
providing a proper translation of the doctor’s talk.
Communication strategies that have been promoted by the few studies available, include
explicit role negotiation with patient companions, e.g. by highlighting helpful behaviours or by
clarifying and agreeing on respective role preferences for patients and companions (Laidsaar-
Powell et al. 2013).
If there is no patient companion, or if the latter’s linguistic repertoire is not sufficiently solid
to act as an ad hoc interpreter, communication may as well break down entirely. It may be
impossible to build rapport with the patient, perform a verbal history, or even simply to find
out what language the patient speaks. In such a case, a medical interaction may deteriorate to
what is referred to as “médécine vétérinaire” (veterinary medicine) (Clark 1983; S. Bowen
2001), implying that only a physical examination can be performed. As Watt (2008) assessed
that 80 percent of the diagnosis tends to depend on oral communication, a communication
breakdown will unavoidably have a strong negative impact on the diagnosis as well as the
treatment.
Increasingly, doctors are using other tools, such as multilingual phrase books (e.g. Collins and
Slembrouck 2006), medical translation software such as Universal Doctor (see Chapter 8), or
even Google Translate (e.g. Patil and Davies 2014) to cope with language barriers. The use of
these tools offers fascinating scope for further research; yet, little knowledge is available on
their effectiveness.
Research methodology 21
3 Research methodology
This chapter discusses the methodological stance followed in this PhD. Section 3.1 discusses
the worldview and strategies of inquiry adopted in this study. Section 3.2 explains how it
pertains to the research area of linguistic ethnography, which allows us to opt for an open
methodological stance within the realm of Language and Social Interaction research. Section
3.3 discusses the interdisciplinary dimension of my research approach, which incorporates
medical perspectives on communication and assigns a central role to co-construction of the
analysis with medical practitioners. Finally, Section 3.4 motivates the choice for a case study
approach.
3.1 Worldview and strategies of inquiry
“I suppose it is tempting, if the only tool you have is a hammer, to treat

everything as if it were a nail.”(Maslow 1966, 15)
The selection of an appropriate research approach critically depends on the research questions
that are being addressed. Researchers need to consider what are the philosophical worldview 9
assumptions underlying the study; which procedures of inquiry (research designs) are related
to this worldview; and which specific methods or research procedures can translate this
approach into practice (see Figure 3.1 and Creswell 2014). For the purpose of this study, a
pragmatic worldview is applied. The advantage of the pragmatic worldview for the research
questions I wish to address is that pragmatism is a pluralistic, real-world practice oriented
approach that has a “concern with applications – what works – and solutions to problems”
(Patton 1990; McCaslin 2008). All available research approaches can be used to gather relevant
insights and to find “practical and usable solutions” (McCaslin 2008, 675). Researchers are
free to select the data collection and analysis methods that best serve their needs and purposes
(Creswell 2014), rather than being restricted to observing the world through the narrow lens of
a specific profession. 10
A qualitative research approach is chosen. According to Creswell (2014, 2), “qualitative

research is an approach for exploring and understanding the meaning individuals or groups
ascribe to a social or human problem. […] Those who engage in this form of inquiry support a
way of looking at research that honours an inductive style, a focus on individual meaning, and
the importance of rendering the complexity of a situation.” In terms of procedures of inquiry,
this study uses a mix of different designs – most prominently ethnography, case study and
discourse analysis.
9
The term “worldview” is defined as “a basic set of beliefs that guide action”, along the lines of Guba (1990, 17).
Other scholars refer to a similar concept as a “paradigm” (Lincoln, Lynham, and Guba 2011)
10
An example of this pragmatic approach is that my analysis not only considers patient companions as possible
ad hoc interpreters (which is how they have been considered mostly in the area of interpreting studies). Instead, it
accounts for the fact that patient companions may take up a broad range of different roles.
Figure 3.1: The interconnection of worldviews, design, and research methods
Philosophical Worldviews
Postpositivist
Constructivist
Transformative
Pragmatic
Research approaches
Quantitative
Qualitative
Mixed methods
Research Methods Research Designs

Questions
Quantitative (e.g. experiments)
Data Collection
Qualitative (e.g. ethnographies)
Data Analysis
Mixed methods (e.g. explanatory
Interpretation
sequential)
Validation
Source: Creswell (2014, 5)
3.2 Linguistic ethnography
“[Let’s] roll up our linguist sleeves to drill down to the details of social
problems” (Auer and Roberts 2011)
Our study belongs to the field of linguistic ethnography (LE). This term captures a growing
body of research by scholars who combine linguistic and ethnographic approaches in order to
understand how social and communicative processes operate in a range of settings and contexts
(Shaw, Copland, and Snell 2015, 1). LE finds its origins in interactional sociolinguistics and
the ethnography of communication, 11 but it keeps its door also open to broader approaches
from within anthropology, applied linguistics, and sociology (Creese 2008, 229). LE takes an
interpretive look at local and immediate actions of actors from their point of view and considers
how these interactions are embedded in wider social contexts and structures (Creese and
Copland 2015, 2). It is a “disciplined way of looking, asking, recording, reflecting, comparing,
and reporting” (Hymes 1980, 105).
Even though LE can be seen as belonging to the broad field of linguistics, it differs to some
extent from mainstream linguistics. Empirical procedures in linguistics are usually relatively
11
For a brief review of these research areas, see Annex 1.
standardised and taken for granted in some schools and paradigms (Rampton, Maybin, and
Roberts 2015, 17). This should be seen in view of the goals of linguistic research, which usually
seeks to identify general patterns in language structure and use.
Linguistic ethnographers, in contrast, combine different types of data to gain insights in the
complexity of social events (Blommaert 2007). A distinctive feature of ethnographers is that
they typically undergo a period of “immersion” in the context they are studying. Their presence
in the field, and the associated processes of learning and adaptation to the research setting play
a major role and can be expected to have relevant implications for the resulting analysis
(Rampton, Maybin, and Roberts 2015, 17). As Rampton et al. (2015) put it, “Ethnography
seeks to produce theoretically ‘telling’ (rather than typical) cases, using ‘the particular
circumstances surrounding a case ... to make previously obscure theoretical relationships
suddenly apparent’ (Mitchell 1984, 239), and it demands our attention for the ‘delicacy of its
distinctions [rather than] the sweep of its abstractions’ (Geertz 1973, 25)”.
Our analysis uses linguistic ethnography as a stepping stone, but draws, where relevant, on
other subdisciplines within the field of language and social interaction research, such as
discourse analysis, interactional sociolinguistics, interpreting studies, L2 research, and social
psychology and on seminal work by Gumperz, Goffman and Hymes (see Annex 1). Following
Rampton’s distinction of “doing” versus “using”, I will “use” the insights emanating from these
disciplines, rather than “doing” these disciplines. They will “suggest directions along which to
look” rather than “prescriptions of what to see” (Blumer 1969). This choice of a relatively
broad methodological stance rests on the conscious realisation that, to study the joint
significance of different communication aspects, contextual elements, and the medical
perspective (and their respective interactions) for the outcome of the communication process,
one cannot restrict the analysis to specific linguistic aspects of the communication process
alone. My aim is to make “optimum use of the sensitising frameworks available in the (sub-)
disciplines focusing on language” (Rampton 2017, 1).
3.3 Linking discourse to medical practice: an interdisciplinary approach
“Nothing underscores the importance of the [medical] history more than

being unable to talk with the patient, an increasingly common experience.”
(Bickley 2013, 81)
Besides drawing on different subdisciplines within the domain of LSI research, this PhD also
builds on insights from communication models developed in the medical “praxis” literature,
which approaches the doctor-patient interaction as a “conversation with a purpose”
(Bickley 2013).
To date, research studies that combine insights from medical “praxis” literature with insights
from discourse literature remain scarce. The brunt of the doctor-patient communication
research pertains to the “praxis” literature, which focuses on talk as a tool for giving
information and displaying emotions such as affection (Ainsworth-Vaughn 2001). In this

literature, language tends to be considered as a transparent vehicle for communication, and
interactions are analysed predominantly with a view to identifying recurrent functional patterns
for activities that are characteristic of medical consultations such as collecting information or
creating rapport. Hence, the emphasis is on the functionality of language, and the outcomes of
the communication process. The more scattered research on doctor-patient communication that
pertains to the “discourse” literature focuses more on the talk itself, e.g. how are sequences
organised and what are the observed power relations during the conversation, and give less
attention to the global functional outcomes of the communication process.
This study analyses talk from a discourse perspective, describing the communication process,
and which strategies are used by the interactants to negotiate meaning. However, these are
linked to the global functional outcomes of the communication process, as I explore the impact
of the communicative dynamics on the flow of the medical consultation and the ensuing
medical care provision. In other words, the medical purposes that pertain to the communicative
goals of the doctor-patient interactions take a central position in my analysis (see Section 5.2.1).
More generally, interdisciplinarity has an increasing impact on the research field of linguistic
ethnography, as Creese and Copland (2015) have highlighted.
The presented analysis has come about in collaboration with both the communities of practice
(the medical practitioners involved in the consultation) and the communities of speech (the
language specialists that helped with the translation of the transcripts), which led to the creation
of a “community of resources” (Katz et al. 2000 as cited by Candling and Candling 2003). This
partly responds Candling and Candling (2003) who argue that studying healthcare
communication in an intercultural context requires the collaboration of applied linguists with
health professionals and researchers from other social sciences; but that such inter-professional
collaboration has been the exception rather than the rule in applied linguistics. To promote
inter-professional collaboration in order to better understand the context of the interaction, and
to promote practical relevance of the results of the analysis, Candling and Candling (2003, 146)
say:
“We can, however, make a beginning; we can adopt a more open

methodological stance; we can avoid using healthcare data merely to serve
refinements in our theories; we can learn to code our results conceptually and
in terms of language understandable to our professional colleagues and to the
end users; we can begin from the problematic as defined by our collaborators
and address not just questions of how but also questions of why and to what
purpose.”
The medical perspective is strengthened by co-constructing the analysis (Candling and

Candling 2003) with medical practitioners in a heuristic, iterative approach through processes
of member-checking, using practitioners as a soundboard. The three data analysis chapters have
also been read by the clinicians who featured in it as to warrant the correctness of the
assumptions made on their medical and communicative activities during the considered
consultations and make the analysis more objective. Finally, this study is deliberately written
in a way as to make both the theory and findings accessible and understandable to both linguists
and non-linguists (such as medical practitioners).
3.4 A case study approach

Our use of linguistic ethnography involves that different types of data are collected using
ethnographic methods (such as field notes from participant observation, multimodal audio-
recordings, and interviews after spending a considerable time of immersion in the research
setting, see Chapter 0); and that these are used to develop a “thick description” of social
behaviour (Geertz 1973) and to produce highly contextualized case studies which are analysed
using a heuristically developed conceptual framework, drawing on insights from discourse
analysis.
Case studies rely on multiple sources of evidence, that are triangulated with each other. The
use of highly contextualised case studies is a common approach in linguistic ethnography and
interactional sociolinguistics (see e.g. Stubbe 2010; Heinrichsmeier 2016; Maryns 2004;
Rampton 2006). According to Yin (1994), case studies are the preferred strategy for research
questions that consider the “how” and “why” of complex contemporary social phenomena.
Case study research allows an exploration of the holistic and meaningful characteristics of real-
life events, over which the investigator has little or no control. It is typically applied in cases
where contextual conditions are considered as highly relevant to the phenomenon under study,
and where the boundaries between the phenomenon and its context are not clear. Other research
strategies such as experiments or surveys either deliberately isolate a phenomenon from its
context (by “controlling” for the context), or have a limited ability to investigate the context as
they can only consider a limited number of variables in their analysis (Yin 1994). In contrast,
case studies typically consider many more variables of interest than there are data points.
Case studies can be used for exploratory, descriptive or explanatory purposes (Yin 1994).
Exploratory case studies try to answer general questions, to prepare the ground for further
examination of a phenomenon under study. They take a broad perspective and allow the
researcher to explore any phenomenon in the data that may be of interest. Descriptive case
studies typically start with a descriptive theory to support the description of a phenomenon
based on collected data. Finally, case studies can as well be explanatory, for instance if the
researcher juxtaposes competing theoretical explanations for a single set of events and
compares each theory with the actual course of events (e.g. as in Allison 1971). The case study
approach in this PhD is in between exploratory and descriptive. On the one hand it is
exploratory in the sense that the available literature does not yet provide a conceptual
framework that describes well the observations made. In such case, as Yin (1994) argues, “any
new empirical study is likely to assume the characteristic of being an “exploratory” study”. On
the other hand, the approach taken is descriptive, as it develops and applies a theory to describe
the data.
Case studies benefit from the prior development of theoretical propositions to guide the data
collection and analysis (Yin 1994). This is a point of possible divergence with ethnographic
research, which according to some schools should refrain from prior commitment to any
theoretical model in order to avoid the pitfalls of structural determinism (Wilson and Chaddha
2009). On the other hand, the separation of ethnography from theory has as well been criticized
from within the field. For example, Wacquant (2002) argues that an “unwarranted empiricist
disjunction of ethnography from theory” leads to analytical weakness in ethnographic research;
Wilson and Chaddha (2009) conclude that “good ethnography is theory driven”.
The theoretical framework that is used in this study (and that will be discussed in the next
chapter) is neither strictly deductive nor inductive, but represents a combination of both.
Concepts from existing theories and available literature have been used as “sensitising
concepts” to guide reflections on empirical data, and to lay a foundation for the analysis and
development of a conceptual framework drawn from the data (G. A. Bowen 2006).
Case study research has sometimes been criticized for its lack of rigor, offering scope for biased
interpretations of the evidence. Skewed interpretation is also a common risk of ethnographic
studies, where “the strong attachments one develops with one’s subjects, […] can lead to
emotions that make the idea of social science less than realistic” (Duneier 1992). This
highlights the need to pay particular attention to objectivity in research. 12 The external validity
and scope for generalization of case study research has been questioned as well. As Yin (1994)
put it, a single case study (like a single experiment) cannot be generalized to populations or
universes. Instead, researchers use case studies to formulate theoretical propositions, and to
expand and generalize theories.
12
See Section 5.5 for some strategies I have applied in my research to strengthen its objectivity.
Conceptual framework 27
4 Conceptual framework
This chapter presents the conceptual framework that is guiding the analysis in this research
with a view to exploring the dynamics of (mis)communication in language discordant
multiparty interactions in the ED, drawing on different research disciplines. The aim of my
analysis is to describe how doctors, patients, and third parties that are present at the medical
consultation negotiate meaning in language-discordant interactions in order to achieve their
communicative goals, and to what extent they are successful.
A first prerequisite for the development of a framework to analyse miscommunication, is

understanding what this concept means. To this extent, Section 4.1 explains how
miscommunication is defined in this PhD. Next, Section 4.2 discusses how I went about
developing a conceptual framework based on deductive and inductive insights. The resulting
framework has three main levels of miscommunication. Section 4.3 shows how different levels
of miscommunication can be discerned in my data, going from simple, localized
misunderstandings to broader, globalized occurrences, drawing on an existing framework that
was developed by Coupland et al. (1991). Section 4.4 presents the different sources of
miscommunication that are considered in my analysis. Examples of interactional dynamics
which may lead to miscommunication are lack of engagement, violation of interactional rules
and misalignments in frames and/or goals. Finally, Section 4.5 reviews different
communicative strategies commonly used by participants in interactions similar to the ones
under study to prevent, accommodate or repair miscommunication. Examples of such strategies
are convergence, confirmation checks, and non-verbal behaviour such as gesturing and other
body movements.
4.1 Defining miscommunication

As Coupland et al. (1991, 11) have argued, miscommunication is a broad and multidimensional
but also “slippery” concept which may cover a range of different phenomena depending on the
definition. Different researchers use different terms for the same phenomenon, on the one hand,
and similar terms for different phenomena on the other hand (Gass and Varonis 1991).
While some use the term “miscommunication” to refer to any interactional problem that arises;
others use it to point quite narrowly at localised problems of misunderstanding (Stubbe 2010,
39). Miscommunication happens when a mismatch occurs between the speaker’s intention and
the hearer’s interpretation (Milroy 1984); and can then originate in a hearer-based
misunderstanding or in a speaker-based mis(re)presentation of meaning. But
miscommunication can also be less significant, consisting of minor mishaps such as slips of
the tongue, omissions or substitutions which do not interfere with the smoothness of an
interaction.
Miscommunication also well be more severe, when communication breaks down completely.
As Coupland et al. (1991) argue, it is illusionary to think of miscommunication simply as a
deviation from “perfect” communication, as language use and communication are inherently
flawed, partial, and problematic: even clear, concise and honest communication may be the
cause of difficulties between individuals who interact with each other (Redding 1972).
A misunderstanding is a specific type of miscommunication. It may emerge from the difficulty

to understand what is literally being asked; or from the difficulty to understand what is meant
by what is said, or how one is supposed to act in response to what is said. A misunderstanding
may be very overt, such as in the extreme case where there is insufficient understanding to
proceed with a consultation at all (Roberts et al. 2005). However, it may as well be subtler, and
even pass unnoticed – to be (or not) revealed at a later point in time during the interaction. In
this context (Milroy 1984, 15) makes a distinction between an incomplete understanding, in
which at least one of the participants has noticed that something has gone wrong, and a
misunderstanding that remains undetected. For the sake of clarity, in this study I will refer to
Milroy’s ‘misunderstandings’ as undetected misunderstandings, and to Milroy’s ‘incomplete
understandings’ as ‘detected misunderstandings’.
Gass and Varonis (1991, 125) argue that the major difference between detected and undetected
misunderstandings is that if participants recognise a problem, they may act in a way to try to
solve it, while in the case of an undetected misunderstanding, this is not the case. In some cases,
misunderstandings are detected, but no attempt is made for repair.
Like miscommunication, misunderstandings may have different causes, outcomes, as well as

consequences (Tzanne 2000, 2). In a medical interaction, they may be inconsequential, but they
may as well hamper the achievement of the medical communicative purpose. If the medical
communicative purpose of an interaction is not achieved, this may have clinical consequences.
In other words, miscommunication can alter the medical history of the presented illness,
diagnostic or therapeutic interventions, the patient’s own understanding of his/her condition,
and/or the follow-up plan, with serious risks to the quality of healthcare (Flores et al. 2012).
4.2 Developing a conceptual framework

The conceptual framework that is applied in this PhD has been developed through a
combination of deductive and inductive insights. It is deductive in the sense that it starts off
from an existing theoretical framework, notably Coupland et al.’s (1991) conceptual
framework that distinguishes different levels of miscommunication. It is inductive in the sense
that this framework is used a starting point for the analysis; not as an end point: it is further
enriched with insights from interpreting studies, interactional sociolinguistics, second language
learning, and medical communication based on observed patterns emerging from the data. The
audio-recordings and corresponding ethnographic data generated a ‘site of encounter’
(Rampton 2007, 585). Their multidimensional character (including variables such as language
barriers, role dynamics, stress and unpredictability) could not be confined to one single
academic discipline (Phillips and Pugh 2010, 59), and called instead for the consultation of
different lines of research and methodologies. Different strands of literature were examined
and added whenever the “circumstances of the data” demanded it (Rampton 2007, 585).
Existing theories are used as “sensitizing concepts”, that are further improved and refined
through an iterative and heuristic process of data analysis and interpretation (G. A. Bowen
2006). All these different methodological approaches are integrated into a comprehensive
framework. Hence, the design of the conceptual framework was determined by the nature of
my data, and developed along the way of my analysis, instead of largely being established
beforehand. For clarity and readability purposes, however, the analytical framework is
presented chronologically before and not after the analytical chapters.
The approach of this PhD is exploratory in that still little is known about the interactional
dynamics of miscommunication in language-discordant consultations, and even less so in the
specific context of the ED. It does not pertain to the domain of “testing out research” which is
aimed at testing and challenging previously proposed generalisations and theories (Phillips and
Pugh 2010, 59). It is problem-solving to the extent that better insights in their dynamics are a
first step towards a solution to communication problems. My approach can also be considered
as interdisciplinary, as different disciplines are applied in an integrated way to address one
single problem. This exploratory and problem-solving approach goes well with linguistic
ethnography as described by Hymes (2004, 11).
In what follows, I discuss in more detail the result of this approach, which takes the form of a
framework that distinguishes different levels of miscommunication (Section 4.3), different
sources of miscommunication (Section 4.4) and different strategies to prevent, accommodate
and repair miscommunication (Section 4.5).
4.3 Levels of miscommunication

In order to stimulate more in-depth and cross-disciplinary discussions of miscommunication at
a conceptual level, seminal work by Coupland et al. (1991) proposes a framework that
distinguishes between different levels or “faces” of miscommunication, integrating insights
from different strands of the literature (see also, for instance, Stubbe 2010). A summary of this
framework is presented in Table 4.1. While the different levels are presented in a hierarchical
structure, ranging from “localized” occurrences of miscommunication such as simple
misunderstandings that are repaired within a few turns of an interaction, to “globalised”
occurrences of miscommunication that involve structural power imbalances, distinct social
identities and value systems and may lead to problematic intergroup relations. Nonetheless,
Coupland et al. (1991, 16) emphasise that, given the complexity of miscommunication, the
same communicative acts can be interpreted in highly diverse ways, and hence several of the
following levels may apply simultaneously.
Level I miscommunication considers those flaws in meaning transfer that remain unnoticed and
therefore unrepaired.
Level II miscommunication refers to minor misunderstandings that are recognisable, either by

participants in an interaction or by researchers. Participants may become aware that the
interaction becomes more effortful, due to the occurrence of “equivocations, disqualifications
(Bavelas 1983), 13 small deceptions, minor misunderstandings, interrupted turns at talk, and
slips of the tongue” (N. Coupland, Wiemann, and Giles 1991, 13). In response, they will try to
act so as to “avoid undue clarity, unpleasantness, threat and confrontation.” This may involve
attempts at repair (often immediately, e.g. through third or fourth turn repair, or in a later turn,
see e.g. Schegloff (1977, 1992, 2000); but participants in an interaction may as well decide to
“let it pass” (Ragan and Hopper 1984), for attempts to repair may come with problems of their
own (such as highlighting communicative inadequacy).
In this context, it is important to note that communication has some inherent mechanisms to
prevent or accommodate misunderstandings. As Pinker (1995) argues, the meaning of a
message often remains unchanged even if specific parts of it are removed. This can be referred
to as “the redundancy of language” (Seife 2007). The main implication of linguistic
redundancy is that it is not always needed to understand everything that is being said to get the
message. This aligns with Brown (1995)’s view that a full comprehension between different
interactants of what is said or meant is not possible, but also not required to understand each
other. She proposes to speak instead of adequate interpretation of what is meant or said. If
grammatical mistakes do not prevent the hearer from understanding the message, one can speak
of grammatical redundancy. Redundancy can also be applied to information content in a
broader sense, implying that not every single word or utterance needs to be understood in order
to achieve a communicative purpose.
Level III miscommunication considers flaws in communication processes that can be attributed
to certain personal deficiencies of the speaker or the hearer, more specifically to a lack of
communication or social skills, personal communication style or attitude (which can often be
repaired through skills training or the implementation of protocols) or to ill will, unwillingness
to communicate, or personality problems. In conversations involving non-native speakers,
language skill deficiencies may also generate Level III miscommunication, and can be
prevented through language skills training.
Level IV miscommunication, on the other hand, recognises the strategic value of interactions
and defines miscommunication as flaws in conversational goal attainment, in other words,
failures to achieve desired social outcomes in terms of relational goals (managing relationships
between interactants, such as power differences or the degree of closeness), identity goals
(reflecting an individual’s desire to present themselves or the conversational partner in a certain
way), and instrumental goals (the achievement of specific task-related outcomes, including
providing or seeking information, persuasion, offering comfort or support) of interactions. At
this level, repair is possible through repeated or modified attempts to achieve these goals, either
13
Disqualification is non-straightforward communication – messages that are ambiguous, indirect, or evasive to
some degree. Speakers use such messages primarily when placed in a “bind”, e.g. when addressing persons with
conflicting interests or opinions, or in situations requiring tact, notably “when it would be unkind to be honest but
dishonest to be kind” (Bavelas 1983).
in the same, or in a later interaction. In the context of the analysis of misunderstandings in

medical consultations, the strategic goal of the interaction includes the medical communicative
purpose, which mostly consists of exchanging information between the doctor and the patient.
During the history taking sequence of the consultation, for example, the communicative
purpose is to elicit medical and social history from the patient (Bickley 2013).
Level V miscommunication focuses on those instances of miscommunication which relate to

group dynamics and/or originate in intercultural differences in behaviour, beliefs, perceptions,
and interpretations. At this level, identity is defined in social rather than personal terms (Tajfel
and Turner 1979) and communication difficulties may arise from “lack of understanding of
differences, suspicion or fear of the outgroup, or threatened social identity”. Such instances of
miscommunication may be repaired through an increased understanding of social processes.
Finally, Level VI miscommunication considers the potential of interactions to reinforce or

constitute a societal value system and its associated social identities (see also Henley and
Kramarae (1991) or Coupland et al. (1991)) by implicitly or explicitly disadvantaging
individuals or groups, while proposing themselves as “morally correct”. Repair of such
miscommunication involves political or socio-structural change.
Table 4.1: Integrative model of levels of analysis of miscommunication
Level Characteristics Problem status Awareness level Repairability

attributed to
“miscommunication”
1 Discourse and Unrecognised Participants are Not relevant
meaning transfer are unaware
inherently flawed
2 Strategic compromise; Possibly, not Low for participants; Relevant at local
minor necessarily easily identified at local level only
misunderstandings or recognised. level by researchers
misreading are routine
disruptions to be
expected
3 Presumed personal Problems attributed Moderate for Deficient people
deficiencies to individual lack participants; directed can be “fixed”
of skill or ill will towards other (or (e.g. by skills
(or both) sometimes self) training)
4 Goal-referenced; Problems High; participants may Repair is an
control, affiliation, recognised as be fully aware of ongoing aspect of
identity and failure in strategic implications of everyday
instrumentality in conversational goal behaviour. interaction with
normal interactions attainment. relationship
implications
5 Group/ cultural Problems mapped Moderate; group Acculturation or
differences in onto social identities taken for outgroup
linguistic/ identities and granted and differences accommodation;
communication norms, group- seen as natural sociocultural
predisposing memberships reflections of groups’ learning
misalignments or statuses
misunderstandings
6 Ideological framing of Participants Participants are Only through

talk; socio-structural perceive only status typically unaware; critical analysis
power imbalances quo researchers hyper- and resulting
aware, galvanised by social change
their own ideology
Source: Coupland, Giles, and Wiemann (1991, 13)
In my analysis, I will focus mostly on the first four levels of miscommunication, as it is difficult
to discern very broad sources of miscommunication such as those falling under level V (inter-
group dynamics) and level VI (societal values) based on a single, relatively “localized” event
such as a medical consultation. I verified with the translators and transcribers who belonged to
the same community of speech and country as the respective patients and companions whether
they detected any instance of miscommunication that might potentially be attributed to these
levels. However, they did not. This does not mean that these levels were not present at all. It
rather means that to identify these broader communicative dynamics, a different analytical
approach would be needed, notably one that has a wider focus, looks at communicative
dynamics at the macro-level, and involves considerably more observation, to be sufficiently
representative for specific “groups” of patients, companions or even doctors.
4.4 Sources of miscommunication

This section describes the way in which my analysis will discern different sources of
miscommunication, focusing on those causes that are most frequently observed in contexts
such as those considered in my analysis. For instance, does miscommunication stem from a
lack of linguistic skills only, or are there other factors at play, such as glitches in the application
of interactional or interpreter rules, or the lack of shared background knowledge? Also broader
issues can lie at the roots of miscommunication, such as role dynamics, lack of engagement,
and misalignment in communicative frames and/or goals. Where possible, I also relate these
sources to the levels of miscommunication they are most likely to provoke.
4.4.1 Language barrier effects

The most obvious source of miscommunication in language-discordant doctor-patient
consultations is the lack of language skills of one or more of the interactants (Gass and
Varonis 1991). This may result in speaker-based mis(re)presentations as well as hearer-based
misinterpretations. Roberts et al. (2005) discern the following three types of language-related
speaker-based mis(re)presentations based on their analysis of language-discordant GP
consultations in London:
x Errors against grammar, vocabulary, and lack of contextual information: This involves
word-finding difficulties as well as the use of wrong terms for concepts; producing
vague language, juxtaposing key words with little “syntactic” glue, and wrong use of
pronouns. Also, the wrong use of time markers such as “I have worked this morning”
as opposed to “I have worked since this morning”, or confusion between “since when”
and “when” may cause confusion. In some cases, the way in which “yes” and “no” is
used can be misleading as well.
x Pronunciation and word stress: Mispronunciation of words (such as “brist” instead of

“breast”) and differences in word stress (such as a non-native English speaker who
pronounces the English “hypothesis” as “hypothesis”) may cause confusion for a native
English-speaking hearer.
x Intonation and other features of speech delivery: Tunes, rhythm and stress help to
organise information units, help to distinguish between old and new information and to
establish the speaker’s perspective. When this part of the communicative style differs
between two interactants, confusion may arise. The choice of where to put the stress in
a sentence can also vary between languages and/or specific users of language. In British
English, stress is usually put on the part the speaker finds most important. Non-native
speakers and/or speakers of another variety of English, may, however, unintentionally
put stress elsewhere. Some even speak without prosody, which makes it difficult for
the hearer to determine which element of his message the speaker wants to draw
attention to.
Similar inferences can be made about miscommunication resulting from a lack of language
skills of the hearer, in other words, about hearer-based misunderstandings. Roberts and Sarangi
(2005) refer to moments when patients are having difficulties to understand the doctor as
“moments of trouble”. Li (2013) touches upon hearer-based misunderstandings when he refers
to the interpreters’ potential lack of “textual comprehension”, which may affect their
performance but may remain unnoticed if interpreters give the impression of being fluent in
the language of the interaction (Meyer, Pawlack, and Ortun 2010).
In these cases, the cause of miscommunication can be attributed to one or more participants in
a conversation, and may be prevented through language training, these instances of
miscommunication can be considered as Level III miscommunication (see Section 4.3).
Smaller instances of miscommunication arising from language deficiencies that are
immediately repaired or do not even need to be explicitly repaired (e.g. because of the
redundancy of language) can be classified as Level II miscommunication.
4.4.2 Violation of interactional rules

Miscommunication can also be triggered by infringements of the rules of interaction. While
these rules should by no means be considered absolute or universal, they offer an interesting
perspective.
x Turn-taking: If the rules of turn-taking (Schegloff 1987; Schegloff 2000) are violated,
e.g. if different interactants are talking at the same time (overlapping in turn-taking) or
if there is no space left for the patient to answer (too little processing time),
misunderstandings may occur. Research by Roberts et al. (2005) already described
“overlapping or interrupting talk” as a violation of turn-taking rules. Even though there

is some evidence that turn-taking rules are universal or at least similar in different
languages (Stivers et al. 2009), individual or group behaviour may still vary (Gumperz
and Tannen 1979; Tannen 2011).
x Questions and answers: There are several implicit rules in question-and-answer based
conversations. If a question is not well understood, it should be repeated. Respondents
need processing time after each question (Martin 2015). Asking multiple questions at
the same time may be confusing and leave the respondent puzzled as to whether
questions are separate or the same, and which question to answer. Suggestive questions
may lead to false recognition. If the respondents have relatively weak language skills,
(closed) binary questions may be easier to answer than open questions; but if
respondents’ language skills are very weak, also binary questions may be confusing by
increasing the risk of false recognition.
As these sources of miscommunication can be mitigated to some extent by providing doctors

with appropriate training on interactional rules during an interview, miscommunication arising
from these origins can be considered as level III miscommunication. Of course, it is not
reasonable to expect, even after appropriate training, that doctors would behave according to
textbook models under any circumstances. Violations of interactional rules are unavoidably
fostered by time constraints, chaos, stress and tiredness, which are the typical conditions under
which care is provided in the ED.
4.4.3 Interpreting errors

Language-discordant consultations are often mediated (or partially mediated) by an interpreter,
be it a professional interpreter or a non-professional ad hoc interpreter, mostly a family member
or friend of the patient who is more proficient in the hospital language. While interpreters are
a valuable tool to overcome language barriers, they may also make errors in interpreting, which
lead to misunderstandings. Table 4.2 provides an overview of the most common interpreting
errors based on research by Wadensjö (1998, 107–8) and Flores et al. (2012, 546–47).
Table 4.2: Common interpreting errors
Rendition type Definition
Close rendition Approximately literal translation of the original (including style)
Expanded rendition More explicitly expressed information in the rendition than in the original.
or addition The interpreter added a word/phrase not uttered by the clinician or patient.
Reduced rendition The interpreter did not interpret a word/phrase uttered by the clinician or
or omission patient.
Substitution A combination of an omission and an addition. The interpreter substituted a

word/phrase for a different word/phrase uttered by the clinician or patient.
Multi-part rendition Two interpreter utterances correspond to one original, which is split into
parts by the interjection of another original.
Summarised Text that corresponds to two or more prior originals.

rendition
Non-rendition Interpreter-initiated utterance that does not correspond to any original

utterance by the clinician or patient
Zero rendition Original utterance not translated.
Editorialization The interpreter provides his or her own views as the interpretation of a
word/phrase uttered by the clinician or patient.
False fluency The interpreter uses a word/phrase that does not exist in that particular
language or an incorrect word/phrase that substantially altered the meaning.
Source: Wadensjö (1998, 107–8); Flores et al. (2012)
If a language-discordant interaction is mediated by a non-professional interpreter, the risk of

interpreting errors increases (e.g. Flores et al. 2003, 2012; Karliner et al. 2007). Apart from
linguistic mistakes, non-professional interpreters are more likely to violate broader rules of
interpreting such as rules regarding the role distribution (for instance by speaking for the
patient, instead of just translating what the patient says). Patient companions may have
insufficient knowledge of medical terminology, and of the importance of the accuracy and
completeness of messages they convey. Moreover, they are not always impartial interpreters,
for instance as they might have a personal stake in the reason for or the outcome of the medical
visit. Such behaviour can as well be referred to as misalignment of goals (see Section 4.4.8) or
as flaws in role dynamics (see Section 4.4.5).
4.4.4 Lack of background information or pretextuality

Interactional uncertainty can arise from a lack of mutual foreknowledge or pretextuality
(Maryns and Blommaert 2002). The latter is, however, characteristic of the ED context (Knopp
et al. 1996). As interactants in the ED usually meet for the first time, there is often no mutual
foreknowledge on issues such as the medical history, the nature of the relationship between the
patient and the companion, their educational, linguistic, socioeconomic and migration
background, each other’s expectations, and each other’s levels of understanding. These gaps in
pretextuality can be a source of miscommunication.
For instance, the lack of a medical record complicates the process of history taking. Unclarities
regarding the relationship of the companion to the patient generate ambiguities regarding the
reliability of the companion’s knowledge on the patient’s medical history. Incorrect
assessments of the interactants’ level of understanding may lead to the selection of ineffective
strategies to address language barriers, undetected miscommunication or mistargeted attempts
to accommodate and repair miscommunication.
In the context of medical communication with non-native patients, the lack of shared
background knowledge often becomes evident as well in the patients’ lack of institutional
knowledge of local social and health infrastructure. In instances where the patient’s companion
takes up the role of ad hoc interpreter, possible gaps in the companion’s institutional contextual
knowledge are likely to negatively impact on the quality of the latter’s interpreting performance
(Li 2013).
4.4.5 Problems of role dynamics

Effective doctor-patient communication usually implies the uptake of specific roles by the
doctor and the patient. A doctor is expected to establish a caring relationship with his patient,
to gather the necessary information to facilitate an accurate diagnosis, counsel appropriately,
and give the correct therapeutic instructions (Ha and Longnecker 2010). Conversely, a patient
is expected to communicate actively, openly, and collaboratively with the doctor, and ask for
the necessary clarifications where needed/desired. At the same time, while doctor-patient
communication tends to involve asymmetric power relations by nature (Pilnick and
Dingwall 2011), this is even more so in the ED (Rosenzweig 1993; Le and Ro 1991;
Lachance 2016).
The preferred rules for doctor-patient communication (at the level of power distance, trust,
empathy, directness, flexibility and ability to listen) may vary across countries, cultures and
over time (Gao et al. 2009; Rimondini et al. 2015; Mazzi et al. 2015). Van de Poel et al. (2013)
argue that patients from more egalitarian cultures prefer to be treated on an equal footing,
implying that the doctor is considered mostly as an advisor, and expected to provide rationales
and explanations for his/her recommendations. While patients from more hierarchical cultures
would expect a more authoritarian style of interaction, do not necessarily want to be involved
in discussing treatment options, and prefer to be told what is best.
Richard and Lussier (2014) argue that once the main reason for the visit has been established,
the patient’s silence, which is often misinterpreted as non-involvement in his/her own care
(Britten and Ukoumunne 1997; Stevenson et al. 2000), can also be seen as an implicature (in
Grice (1975)’s terms) 14 to request the doctor to take action. In the context of a language-
discordant ED consultation, however, the implicature of silence may be more difficult to assess,
as the patient’s silence could just as well stem from a linguistic barrier.
Intermediation by professional interpreters in healthcare is also subject to a set of rules on role

dynamics. Their basic objective is to provide a linguistic conversion of a message spoken in
one language to another one. However, there are many other aspects of interpreter roles that
continue to be debated (see e.g. Avery 2001; Cox 2015), such as the rigidity with which an
interpreter should adhere to interpreting literally messages, rather than conveying affective
content as well; the acceptability of proactive interventions by the interpreter, e.g. to request
14
See Annex 1 for more details on Grice’s theories
clarifications if (s)he notes the patients’ needs are not being met; or the degree of formality that
is most desirable when it comes to the patient/interpreter relationship.
When the patient’s companion takes up the role of the ad hoc interpreter, it is not uncommon
that role confusion arises between the companion and the physician, as both parties may have
opposing agendas (Leanza 2007; Leanza 2005; Li 2013). Moreover, the role of the companion
may as well change over the course of the consultation, for instance from family member,
husband or neighbour, to the translator, spokesman, system navigator or author of what is being
said (Greenhalgh, Robb, and Scambler 2006; Sarangi 2015).
The physician may also take up various roles or change roles during the ED encounter (Lussier
and Richard 2008; Lachance 2016). For instance, Lussier and Richard (2008, 1090–91)
describe how, when a patient’s life is at risk, a doctor may act as expert in charge, and take
unilateral action without directly considering the patient’s psychosocial conditions. In less
acute situations, the doctor may take up the role of an expert guide, in which he takes the lead
in decisions, but explains the patient carefully what will happen next. In case of minor or
chronic conditions, the doctor may act as a partner, providing the patient with different
treatment options, inviting the patient to participate in the treatment decision, and paying
substantial attention to psychosocial aspects. Finally, the doctor can also take up the role of a
facilitator, helping the patient to get a follow-up appointment or showing him/her the way to
the pharmacy. Each of these roles goes together with different Goffmanian frames and their
associated communicative styles (Tannen and Wallat 1987; Goffman 1974).
In general, ED physicians are less likely to take on the role of a facilitator than for instance
general practitioners (Richard and Lussier 2014; Lachance 2016). However, in the case of non-
native patients, the need for facilitation may be higher and hence even ED physicians may take
up this role to help the patient find his/her way through the institution.
When participant roles in an interaction are not pre-established and mutually agreed, confusion
and misunderstandings are likely to arise, especially in the case of multiparty multilingual ED
encounters where people meet each other for the first time. Effective communication in such a
context requires “communicative flexibility” (Gumperz and Cook-Gumperz, 1982) and
implicit or explicit role negotiation even when the choice of roles often takes place in the
margins of the patient’s and doctor’s awareness (Richard and Lussier 2014). Medical
interaction research finds that, in general, it is the physician who decides which role should be
adopted by whom. While the patient and/or his/her possible companion may propose or imply
other roles, the physician appears to have wider discretion in deciding which role to adopt
(Richard and Lussier 2014).
4.4.6 Lack of engagement in the conversation

At the most basic level, miscommunication may arise from a lack of communication. Lack of
involvement in a conversation (non-engagement). This is an important predictor of
misunderstandings. It occurs for instance if a person who is being spoken to, does not recognise
that he or she is being spoken to (Gass and Varonis 1991, 127–128). Possible reasons for this
range from hearing loss, over language barriers, to unwillingness to communicate. While
studying general practitioner consultations in the presence of language barriers, Roberts et al.
(2004) observed that in some cases, patients would say little during the consultation, but instead
offer non-verbal cues, such as empty jars of pills or letters, and wait for the physician to infer
what the problem is. They refer to such behaviour as “low self-display” and relate it, inter alia,
to patients’ possible sociocultural assumptions about the hierarchy in a doctor-patient
consultation and the power of diagnosis based on very little patient-given information.
In multiparty interactions, the risk of non-engagement is particularly high, as it is relatively

easy for some interactants to disengage from the conversation and leave the communication to
the others. Sometimes, patients who are accompanied to a doctor’s visit by family members or
friends who are better-versed in the doctor’s language, have been observed to leave the talking
to the companion, whose footing then switches from “animator” to “author” in Goffman
(1981b)’s terms, as (s)he ceases to act as a mediator/interpreter between the doctor and the
patient.
Given the diversity of reasons that can lie at the basis of non-engagement in a conversation, be
it individual deficiencies in social, communication or language skills, or different cultural
norms about how to behave in a consultation, the case of non-engagement can be analysed
either as Level III or Level V miscommunication.
4.4.7 Misalignment of frames

Even if a hearer understands the words spoken by a speaker, he does not necessarily understand
what the speaker means semantically or pragmatically with his/her specific utterances,
linguistic behaviour and/or conversational style. When a hearer decodes a message conveyed
by a speaker (s)he needs to infer the speaker’s intention by relying on contextual information.
Sometimes, glitches can occur at this level: if the contextual information the speaker relies on
differs from that relied on by the hearer, a “misalignment of frames” occurs which often gives
rise to misunderstandings.
These misalignments of frames can be very simple, e.g. when a professor asks his student: “can
you send me your paper?” and the student sends him a different paper than the one he had in
mind. It can also occur when a doctor asks his patient several questions in a row, without
leaving space for a reply in between. The risk is then high that he receives an answer to one
question, but interprets as the answer to another question. The resulting misunderstandings
would most likely pertain to miscommunication at level IV; as they are not straightforwardly
remediated through training or behavioural changes. They interfere with conversational goal
attainment at its most basic level of the exchange of information.
Misalignments of frames can also be broader and more complex, especially in conversations
between speakers from different backgrounds or cultures, where a lack of shared contextual
information frequently gives rise to communication difficulties and misunderstandings
(Gumperz 1982b, 1982a; Gumperz and Tannen 1979). In such case, misunderstandings are
referred to as “pragmatic failure” (Thomas 1983).
Thomas (1983) distinguishes between two types of pragmatic failure, notably “pragma-
linguistic failure” and “socio-pragmatic failure”. Pragma-linguistic failure occurs when the
pragmatic force mapped by the speaker onto a given utterance is systematically different from
the force most frequently assigned to it by native speakers of the target language, or when
speech act strategies are inappropriately transferred from a non-native speaker from his own
native language to a foreign language. This usually relatively easy to overcome through
additional language skills training and/or immersion in the community of speech (Saville-
Troike 2003).
Socio-pragmatic failure refers to the social inappropriateness of communicative behaviour

(Thomas 1983) and occurs when a wrong interpretation results from incomplete fine-tuning of
speaker behaviour to the sociocultural or socio-psychological context of interaction. For
instance, the way in which a request is formulated can be very telling about the social
relationship between interactants. For example, a doctor can requests a patient to undress in
different ways. If he addresses an old schoolmate, he may say “Hey, can you take off your shirt
buddy?”. If, on the other hand, he addresses a middle-aged man he has not seen before, it would
be more appropriate to say “Would you mind taking off your shirt, so I can examine you?” If
the doctor would have used the first formulation for the second patient, the latter could have
considered this as inappropriate. In such event, a pragmatic failure would have occurred, which
would possibly need to be repaired by the doctor afterwards.
Pragmatic failure may be inconsequential, but it may as well result in interactional conflicts
(Padilla Cruz 2013). It may lead to interactive conflict, friction, unjustified attribution of
negative personality traits and stereotyping, and to communication breakdown (Level V
miscommunication). Pragmatic failure is not restricted to interactants of different linguistic
backgrounds. Pragmatic norms are also subject to regional, ethnic, political and class
differences (Thomas 1983).
4.4.8 Misalignment of goals

Given the strategic value of interactions, miscommunication may arise from a failure to achieve
interactional alignment when interactants have different, sometimes conflicting, goals or
agendas. When different participants are aiming for different outcomes for a particular
interaction, communication becomes less effective as both might try to steer the conversation
towards the accomplishment of their own agenda, and this could mean that they are less
attentive for the interactive needs of their conversation partners. Misunderstandings resulting
from misaligned agendas are salient illustrations of level IV miscommunication.
In this context, Mishler’s work in the area of social psychology is highly relevant. He observes
that the patient’s everyday life and concerns (“the voice of the lifeworld”) often interferes with
the medical agenda (“the voice of medicine”) during medical interactions (see Section 2.2 for
more details). Looking more closely at the ED context, Lachance (2016) has found that
discrepancies between the doctor’s and the patient’s expectations in the ED are relatively
common. Some examples of possible conflicts between doctors’ and patients’ agendas in the
the ED are outlined in Table 4.3.
Table 4.3: Possible conflicts between patient and physician agendas in the ED
Patient’s agenda ED doctor’s agenda
The patient is often anxious and wants to be The doctor’s priority may not be to find the
heard and understood (Wisten and exact cause of the patient’s complaints, but
Zingmark 2007). rather to exclude possible serious diseases and
symptoms (Lachance 2016)
Patient often wants to show the doctor that (s)he
is in pain and to be relieved accordingly The more chronic the symptoms appear, the
(Fosnocht, Hollifield, and Swanson 2004; more the doctor will be assured about the
Fosnocht, Swanson, and Bossart 2001) patient’s condition (Lachance 2016).
Patient wants to know the cause of his/her pain The ED doctor will only address the patient’s
and to be reassured subsequently (Britten and concerns if it fits within his/her medical agenda
Shaw 1994) (Scheeres et al. 2008; Slade et al. 2008).
Patients coming to the ED often consider their

condition much worse than the symptoms may
indicate (Lachance 2016)
Source: Lachance (2016)
As companions are typically with the patient to offer support, they will usually also support the
patient’s agenda and try to push it forward. Companions may bring their own agendas, which
can be an additional source of miscommunication. It is important for doctors to be aware that
the patient’s agenda and the companion’s agenda do not always coincide.
4.5 Prevention, accommodation, and repair

Social interaction allows interactants to negotiate meaning in a conversation, in other words,
to prevent, accommodate or repair miscommunication and ultimately to better understand one
another. The existing literature describes several strategies that interactants use to improve
mutual understanding. This section discusses a series of strategies that are relevant in language-
discordant doctor-patient interactions.
4.5.1 Generic communication strategies in medical interviews

Some of the generic communication strategies that doctors use in language concordant
consultations can enhance communication in language discordant communication as well. An
overview of some of these strategies is provided in Table 4.4.
Table 4.4: Generic communication strategies in medical interviews
Strategy Description
Avoiding Leading questions that include a desired response such as “Is your pain like a
leading sprain?” may prompt a patient to just reply with a "yes’. In language divergent
questions interactions a patient may just say yes or no, without having understood the
question, giving rise to false recognition and misunderstanding. Often such as
misunderstandings pass by unnoticed. Instead, a more neutral question such
“Can you tell me what your pain is like?” will invite the patient to explain what
his/her pain is like. While this may be challenging for patients in language
discordant consultations, inviting them to reply to such a question at least
provides the doctor with a basis on which to assess the extent of the patient’s
comprehension. If the question is unsuccessful at eliciting the requested
response, other types of questions may be tried out.
Using questions In cases where quantitative information is required, open-ended questions such
that elicit a as “How many meters can you walk before get short of breath?” are better than
graded response closed-ended questions such as “Do you get short of breath when you walk for
100 metres?” that require only a yes or no answer. In a language discordant
consultation, it can be troublesome for a patient to provide the doctor with a
graded answer (e.g. in cases where patients have difficulties with numeric values
in the second language), but again, such a strategy avoids false recognition and
makes potential miscommunication more visible.
Asking a series Instead of asking, “Is your pain, stabbing, tightening, throbbing, or burning?” it
of questions, is better to say “Can you tell me what your pain is like? Stabbing? [pause]
one at a time Tightening? [pause] Burning?”. In this way, the patient has more time to process
and understand the different possible answers. In a language discordant
encounter, it often happens that as a result of communication barriers, when
several questions are asked at the same time, at least some of them remain
unanswered and snowball into a series of different questions that patients are
expected to answer all at once. Offering more processing time in between
subsequent questions can help reveal miscommunication on questions or words.
Offering To help patients describe their symptoms, questions with multiple answer
multiple choices options such as “Is your urine yellow, or rather red?” may facilitate
for answer understanding.
Encouraging Backchannels or continuers (such as nodding, or interjections such as hmmm,

patients with okay, go on), can display the doctor’s attentive hearing (Goodwin 2007) and
backchannels encourage the patient to go on with his/her story. In language discordant
consultations, such backchannels can also be useful, for instance to encourage a
patient with word-finding difficulties to keep on trying to say something.
Echoing Repeating the patient’s words can encourage the him/her to expand his/her story.
In language discordant consultations, repetition can also function as a self-
accuracy check (see later).
Summarising Summarizing at different points in the consultation what the patient has said so
far can help to get oversight on what is known and what is not known yet. In a
language discordant consultation, this technique can act as an additional self-
accuracy check and help to reveal potential miscommunication.
Source: Adapted from Bickley (2013, 59–62)

The above listed strategies illustrate in particular how the way in which questions are asked,
may have a significant impact on how (much) information is revealed by the patient
(Watt 2008, 3). Some of the usual question strategies may present additional challenges in a
language discordant environment. If so, they can be varied with other types of questions.
4.5.2 Communication strategies in a second language

Long (1983) describes a set of interactional features often used in conversations with non-
native speakers to negotiate meaning. Hearers may use clarification requests (“What do you
mean by X?”) or confirmation checks (“Do you mean X?”). Conversely, speakers may use
comprehension checks (“Do you understand?”). The use of these features often acts as a signal
of communication difficulties, as interactants will only use these features if they believe a
misunderstanding is likely to occur. Other interactional features include input modifications
(e.g. stress on key words, decomposition, partial self-repetition) and semantically contingent
responses (e.g. recasts, repetition, expansions) (Long 1996).
Dörnyei and Scott (1997) further elaborate a list of more than thirty communication strategies
used to avoid (and/or repair) misunderstandings in interactions that involve non-native speakers
including for example the use of circumlocution or paraphrasing, approximations, all-purpose
words, self-repair or self-rephrasing, and direct or indirect appeals for help. Their taxonomy
draws on earlier research by Tarone (1977), Faerch and Kasper (1983), Bialystok (1983, 1990);
Paribakht (1985); Willems (1987); Poulisse (1987; 1993); and finally Dörnyei and Scott
(1995b, 1995a). An overview is provided in Table 4.5.
Table 4.5: Inventory of communication strategies in a second language
Strategy Description
1. Message Leaving a message unfinished because of some language difficulty

abandonment
2. Message Reducing the message by avoiding certain language structures or topics

reduction/topic considered problematic languagewise or by leaving out some intended
avoidance elements for a lack of linguistic resources
3. Message Substituting the original message with a new one for not feeling capable
replacement of executing it
4. Circumlocution Exemplifying, illustrating or describing the properties of the target object

(paraphrase) or action
5. Approximation Using a single alternative lexical item, such as a superordinate or a related

term, which shares semantic features with the target word or structure
6. Use of all-purpose Extending a general, “empty” lexical item to contexts where specific
words words are lacking
7. Word coinage Creating a non-existing word in the non-native language by applying a

supposed non-native language rule to an existing non-native word
8. Restructuring Abandoning the execution of a verbal plan because of language

difficulties, leaving the utterance unfinished, and communicating the
intended messages according to an alternative plan
9. Literal translation Translating literally a lexical item, an idiom, a compound word or

(transfer) structure from the own native (or another) language to the non-native
language
10 Foreignizing Adjusting a word from the own native (or another) language to the non-
native language phonology or morphology
11 Code switching Including a word from the own native (or another) language in non-native
(language switch) speech (ranging from single words to complete turns)
12 Use of similar- Compensating for a lexical item whose form the speaker is unsure of with
sounding words a word (existing or non-existing) which sounds more or less like the target
item
13 Mumbling Muttering inaudibly a (part of a) word whose correct form the speaker is
uncertain about
14 Omission Leaving a gap when not knowing a word and carrying on as if it had been
said
15 Retrieval In an attempt to retrieve a lexical item, saying a series of incomplete or

wrong forms or structures before reaching the optimal form
16 Self-repair or other Making self-initiated corrections in one’s own speech or in the

repair interlocutor’s speech
17 Self-rephrasing Repeating a term, but not quite as it is, but by adding something or using
paraphrase
18 Over-explicitness Using more words to achieve a particular communicative goal than what
(waffling) is considered normal in similar native speaker situations
19 Mime (non- Describing whole concepts nonverbally, or accompanying a verbal

linguistic/ strategy with a visual illustration
paralinguistic
strategies)
20 Use of fillers Using gambits to fill pauses, to stall, and to gain time in order to keep the
communication channels open and maintain discourse at times of
difficulty
21 Self-repetition and Repeating a word or a string of words immediately after having said to
other- repetition them, or after the interlocutor has said to them, to gain time.
22 Feigning Making an attempt to carry on the conversation in spite of not

understanding understanding something by pretending to understand
23 Verbal strategy Using verbal marking phrases before or after a strategy to signal that he
markers word or structure does not carry the intended meaning perfectly in the
non-native language
24 Direct or indirect Turning to the interlocutor for assistance by asking an explicit question
appeal for help concerning a gap in one’s knowledge of the non-native language; or trying
to elicit help indirectly by expressing lack of a needed item either verbally
or non-verbally.
25 Asking for Requesting repetition when not hearing or understanding something

repetition properly
26 Asking for Requesting explanation of an unfamiliar meaning structure

clarification
27 Asking for Requesting confirmation that one heard or understood something correctly
confirmation
28 Guessing Similar to the confirmation request but with a lesser degree of certainty
regarding the message heard
29 Expressing non- Expressing that one did not understand something properly, either
understanding verbally or nonverbally
30 Interpretive Extended paraphrase of the interlocutor’s message to check that the

summary speaker has understood correctly
31 Comprehension Asking questions to check that the interlocutor can follow you
check
32 Own-accuracy Checking that what you said was correct by asking a concrete question or
check repeating a word with a question intonation
33 Response: repeat, In response to a signal of miscommunication by the other (e.g. other-

repair, rephrase, repair, a request for clarification, …), the speaker can repeat the original
expand, confirm trigger (possibly in a corrected form), provide other-initiated self-repair,
rephrase the original trigger, put it into a larger context, or confirm what
the interlocutor said or suggested.
Source: Dörnyei and Scott (1997, 188–94)
4.5.3 Linguistic accommodation

Communication accommodation theory, as developed mostly by Giles and his co-authors
(Giles, Coupland, and Coupland 1991a; Giles et al. 1987) posits that when interacting,
individuals may adapt linguistic, prosodic, and non-verbal communicative behaviour to each
other, to reduce differences. This is referred to as convergence. According to Street (1991),
convergence is used by interactants in order to win approval, affiliate, establish rapport, and/or
communicate effectively. In a language-discordant doctor-patient consultation, convergence
may be evident in language-switching. Rampton (2005) describes the activity of performing
utterances in a different language as a means to socialise and create rapport as “crossing”. Non-
verbal behaviour such as expressing involvement through gaze, gestures, body orientation and
facial expression can help as well to establish rapport if a language barrier complicates verbal
communication. In a medical context, the literature prescribes that doctors and patients (and
possible third parties when present) align their levels of discourse to each other as to promote
mutual understanding (Richard and Lussier 2014, 2016a).
Conversely, divergence refers to the way in which such differences are accentuated by
speakers, and may particularly be observed in situations where speakers feel “threatened” in
their social identity (Giles, Coupland, and Coupland 1991b, 9). In doctor-patient interactions,
divergence may as well be evidenced by hierarchical behaviour by the doctor, if this runs
counter the patients’ wishes (Street 1991). Similarly, if doctors try to finish the consultation
quickly, exercise control by shifting topics or switching style abruptly and leave too little space
for patients to express concerns or discuss treatment alternatives, this can also be referred to as
a divergence (Dryden and Giles 1987).
Using baby talk in conversations with young children may be appropriate and effective
instances of convergence. However, excessive convergence (over-accommodation) is often
perceived as downgrading, patronising, or irritating, for example when nurses use baby talk to
elderly, irrespective of their capabilities (e.g. Caporael 1981). Another example is when doctors
believe that patients (or companions) have little understanding of medical terminology, and
therefore use simplified language that is far below the actual patients’ understanding
(Street 1991).
4.5.4 Non-verbal communication

Gestures and their concurrent words are often co-expressive and meaningfully related
(Gerwing and Allison 2009, 312). In face-to-face interactions, participants commonly use
prosody, facial and hand gestures to make meaning or to convey a message (Bavelas, Coates,
and Johnson 2000; Holtgraves et al. 2014). Such gestures are not just redundant movements or
signs of a lack of language fluency. On the contrary, as argued by Holtgraves et al. (2014, 17),
they make out an integral part of communication. According to Gerwing and Dalby (2014,
308), gestures hold a semiotic potential although they lack contextual meaning and their
interpretation mainly relies on accompanying speech and the context of the interaction. Even
though gestures and non-verbal communication often take place beyond the awareness of the
speakers, they sometimes have a potential to prevent misunderstanding by enhancing
communication.
Non-verbal behaviour (such as body movements, the display of specific objects) can be
redundant, if it only confirms or strengthens what has already been communicated verbally.
However, it can be non-redundant as well, in case it substitutes for verbal communication by
adding necessary information. In language-discordant interactions, it is common to see
speakers use non-verbal communication instead of verbal communication (Gerwing and
Allison 2009).
4.5.5 Meta-communication
Van de Poel et al. (2013) also point out the importance of meta-communication in preventing
or solving conflicts with patients. Meta-communication literally means communication about
communication, and usually means that the doctor explains what he will be doing next, why,
and how. This can happen in a verbal or a non-verbal way. It can help the patient feel more
secure, at ease, and in control. Meta-communication may also be used for role negotiation. If a
patient or his/her companion does not spontaneously take up the role which is preferred by the
doctor, the doctor may use meta-communication to explain why he prefers the patient or his/her
companion to take up that particular role in the interaction.
In a time-constrained environment such as the ED, the physician may have to frequently
interrupt the patient (or the companion)’s narrative as to manage the information flow
(Lipkin 2008). To avoid threatening the patient’s face when interrupting his/her narrative flow,
the doctor can use meta-information. He/he can say, “I’m sorry for interrupting, but can you
explain this a bit further? How long have you been having this pain?” By making it explicit
that the physician knows that (s)he is interrupting, the patient or companion are less likely to
feel cut off (Lachance 2016; S. A. Cole and Bird 2000).
Description of research procedure 47
5 Description of research procedure

This Chapter explains how the data were collected and which challenges were encountered in
the process (Section 5.1). Next, it discusses the broad structure of the data, which is driven by
the typical structure of a medical consultation (5.2) and how the data were prepared for analysis
(Section 5.3). Section 5.4 discusses the structure of and the selection of the case studies; and
Section 5.5 explains what was done to strengthen their validity.
5.1 Data collection

For this PhD project, data have been collected through participant observation, the most
common data collection strategy in ethnography (Saville-Troike 2003), in a Brussels public
hospital Emergency Department (ED). While the official languages were French and Dutch,
the ED was linguistically and culturally super-diverse.
Over a period of 6 months, 129 clinician-patient interactions in the ED were audio-recorded.

For each of these, contextual information was collected with the help of a smartpen. Working
with the smartpen allowed the simultaneous recording of sound and of relevant contextual notes
and drawings. Most of the interactions were accompanied by walk (or run)-and-talk interviews.
These interviews were recorded before and in between the audio-recorded consultations and
provide additional contextual information to the recorded interactions. The triangulation of
field notes, audio-recordings, interviews, and member checking by the hospital staff supported
the collection of a robust dataset.
Out of the 129 interactions, 55 included a foreign-speaking patient and 15 took place with a
non-native (Dutch/French) speaking doctor. Moreover, 22 cases concerned multiparty
interactions, in other words, they involved (at least) a third person (friends, family, staff…)
next to the doctor and the patient. In many cases, this third person acted as an ad hoc interpreter
for at least part of the interaction.
5.1.1 Context of the data collection
The city
The data were collected in Brussels. The Brussels healthcare sector provides an interesting case
for research on language barriers given its multicultural and multilingual setting. In 2015, 34%
of the Brussels population held a foreign nationality (BISA 2016), and Petrovic (2012) suggests
that 62% is of foreign origin. Contributing factors include the presence of European institutions
and related lobby organizations, but also the fact that Brussels hosts Belgium's largest reception
centre for asylum seekers, with approximately 720 beds (Fedasil 2011). In addition, Brussels
has a large number of undocumented immigrants (Lacourt 2007). According to Vandermotten
et al. (2011), many ethnic minorities move to the capital, looking for solidarity and security;
and big cities are a refuge for those who are excluded from the regular economy. Many newly
arrived immigrants do not speak Dutch or French, and are unfamiliar with the Belgian
institutional context.
The hospital
Data were collected in the Emergency Department of a public inner city hospital belonging to
an officially bilingual (French-Dutch) public hospital network in Brussels. The asylum centre
Klein Kasteeltje and the municipal Public Social Action Centres (OCMW/CPAQ) have
agreements with the hospital under study for medical treatment of asylum seekers. The
extensive diversity among patients as well as staff reflects the city's multicultural and multi-
ethnic character.
The Emergency Department

The inner city ED under study treats on average 55,000 patients per year (Cerf 2012). About
40 percent of these patients hold a foreign passport and four percent are homeless. The ED
employs 80 persons of different ethnic and linguistic origins, including medical staff, clerks,
and maintenance and cleaning staff. As is usual in hospital EDs, foreign patients are
overrepresented as they tend to bypass primary care providers (Debosscher 2012; Rivadeneyra
et al. 2000). During the day, the hospital to which the ED belongs benefits from five on-site
professional medical interpreters, covering the major foreign languages encountered among
patients (Classical Arabic, Moroccan Arabic, Turkish, Polish, Russian and Romanian), and
from external community interpreters catering for other languages. The latter are, however,
seldom called in.
To obtain first-hand information on the number of foreign language patients in the emergency
department under study, a quantitative survey was carried out at triage. Over a 3-month period,
doctors working at the triage in the ED kept information on the patient’s mother tongue, the
language used, and whether professional or ad hoc interpreters were used in the consultation.
The survey also comprised a subjective assessment by the doctor on the quality of
communication during the consultation on a 1 (“Impossible”) to 5 (“Excellent”) scale. The
results of this survey showed that of the 1360 patients, only 69% were native French, Dutch or
English speakers. The remaining 31% of the patients held a wide variety of mother tongues:
42 other languages were counted. The most common languages were Arabic (11% of patients),
Portuguese (5%), Spanish (3%), Polish (2%), Romanian (2%), and Italian (1%).
In 28% of the consultations with non-native patients, ad hoc interpreters were used (such as
family or friends who accompanied the patient or - in exceptional cases - local medical or
hospital technical staff of the same origin). In less than 2% of the consultations with non-native-
speaking patients, a professional mediator or interpreter was called in. For less than half (48%)
of the interactions with non-native patients, clinicians judged the quality of the communication
as "excellent" or "easy". In 15% of the interactions with foreign-language patients, the
clinicians assessed the communication as “difficult" or "impossible".
5.1.2 Entering the field

According to Hammersley and Atkinson (2010: 41), “the problem of obtaining access to data
looms large in ethnography.” Given relatively stringent ethics regulations, hospitals in
particular have proven to be difficult places to gain access for research on communication
(Slade et al. 2008), and especially research involving audio-recording of doctor-patient
interactions (Valero-Garcés 2002: 470).
Finding a gate-keeper
The easiest way to gain access to a hospital for research purposes is by identifying a
“gatekeeper” that helps the researcher “enter the field”: an insider with a personal interest in
the research project that can open doors to give the researcher a foothold in the site of study
(Mack et al. 2005). I was lucky to have the hospital network’s language policy coordinator
taking great interest in my project. She was instrumental in identifying other hospital staff who
could provide me with guidance and support to obtain the necessary permissions to collect data.
Clearly communicating the research topic and objectives was a crucial step in the process of
negotiating access (Hammersley and Atkinson 2010: 42). In particular, the head of the most
diverse ED in the hospital network volunteered to act as the project’s principal investigator in
the research proposal for the ethics committee and allowed me to collect data in his ED unit.
His ED unit was coping with a large group of patients with no or limited command of Dutch,
French or English. He hoped that a study like ours would shed new light on the question of
how to facilitate communication under such conditions.
Complying with ethics regulations

For any form of data collection, approval from a formal ethics committee is required. This
committee ensures that ethics regulations are complied with, and that patient rights to privacy
and quality care are guaranteed. As in most hospitals, the ethics approval procedure was a
rigorous and bureaucratic process, which relied on strictly scheduled and infrequent meetings,
resulting in a long waiting list of research proposals. This delayed the collection of audio data
by approximately one year. 15 Nevertheless, the long preparation time allowed me, as a field
researcher, to gradually get to grips with and identify some structure in the hectic atmosphere
of the ED, to join up with staff, to learn about the logic of a consultation and the main
communication problems experienced by the hospital staff and their patients.
Most hospital-based research draws on biomedical and psychosocial surveys (Hammersley and
Atkinson 2010: 43), and the ethics committee approval protocol is geared towards such studies.
Health research has to abide by strict ethics regulations, with stringent rules on informed
consent (IC) procedures (Pope and Mays 2006) which require patients to sign an IC form prior
to participation. In principle, ICs are based on the concept of autonomy, and should allow
patients to independently make an informed decision on whether to participate in a study or not
(Berg and Appelbaum 2001).
15
The application form for the ethics committee is presented in Annex 2.
5.1.3 Obtaining informed consent

The first draft IC form that was submitted to the ethics committee was based on a standard IC
form used for clinical trials. 16 It comprised an extensive list of issues: project relevance,
objectives, methodology, expected outcomes, the funding agency, and potential risks for
patients involved. Using such a long IC form in a context of language barriers, often aggravated
by a low degree of literacy, did not seem feasible. Therefore, the ethical committee decided to
give their approval for observations, conditional on a feasible strategy to obtain IC from the
concerned patients. Participant observation without audio-recording did not require the
patient’s IC, and could therefore be initiated before having found a solution to the problem of
obtaining IC.
Obtaining IC from foreign language speaking patients to audio-record their interaction with
doctors is relatively new in ED communication research. Pioneering Australian research on ED
communication by Slade et al. (2008) focused on patients that had at least a satisfactory
command of English, excluding patients who needed an interpreter. Flores et al. (2012) also
collected audio-records from medical interactions in an ED in the US for their research.
However, as they focussed on Spanish/English interactions, one bilingual collaborator sufficed
to facilitate the consent negotiation process.
In the context of this study, invoking professional interpreters was challenging: first, the ED
under study operated in a highly multilingual environment. Second, from necessity, patients
were attending ED services without prior appointment. Thus, this would require an unlikely
high number of standby professional interpreters. The use of ad hoc interpreters (such as family
members) was unlikely to suffice to overcome the language barrier in this case as discussing
the IC form requires complex and high language skills (Schenker et al. 2007). A second
problem involved potential side effects of providing patients with complex information. Bhutta
(2004) reviewed different international guidelines used for obtaining IC in research projects in
developing countries. She argues that lengthy and complex IC documents may cause distrust
among potential study participants whose imperfect understanding leads to an overestimation
of the involved risks.
Such challenges are not uncommon in the literature. Bührig and Meyer (2004) have found that
ad hoc interpreters may jeopardise, rather than facilitate, the process of consent negotiation by
not providing a proper translation of the doctor’s talk. Schenker et al. (2007) found in a US-
based study that, despite the presence of on-site interpreters, non-English-speaking patients
were less likely to get documentation of IC for common invasive interventions. Language
barriers are also a common reason for exclusion from clinical trials (Van Spall et al. 2007).
This may contribute to explaining why, in the majority of studies on the understanding of IC
forms, language barriers have not been a topic (Schenker et al. 2010, 170).
16
This initial consent form is presented in Annex 3.
Different strategies were then tried out to settle on what was, in my view, the most pragmatic
procedure to collect data from patients in a way that respects their rights, without jeopardising
their quality of care. First, a simplified one page IC form was developed with a narrower focus
on key aspects of the study, starting from the idea that ours was a negligible risk study (Hale
and Napier 2014) and eliminating superfluous information such as risk, assured benefits and
funding (Bhutta 2004). The one pager drew as much as possible on typical simple French
vocabulary used by doctors in language-discordant consultations, as identified through
participant observation. The document explained that data were collected for a research project
on communication with the purpose of improving it, requested permission to audio-record the
consultation, guaranteed the participants anonymity and emphasised the right to refuse
participation at any time. This simplified consent form was approved by the ED’s head on
behalf of the ethics committee. 17
For the pilot of this consent form, it was agreed that for each new patient, the MD treating the
patient in question would first assess whether the context (intuition, pathology, specific
context) was favourable for a recording. Then he would introduce the researcher to the patient
as a colleague doing research on communication. To engender trust, the researcher would wear
a medical staff-like white coat. The researcher would then give and explain the document to
the patient and ask him/her to sign it if (s)he agreed with its content. Unfortunately, in most
cases, patients and ad hoc interpreters did still not seem to understand the information provided,
especially within the time limits available, as the ED staff’s priority was the provision of
(urgent) medical care to the patient. One reason was that many patients and/or their companions
were largely functionally illiterate.
Therefore, it was decided to develop audio-recorded versions of the IC in a wide range of

different languages. For the translations, the collaboration of field specialists was sought.
During the daytime, the hospital benefited from five on-site professional medical interpreters,
covering the major foreign languages encountered among patients (Classical Arabic, Moroccan
Arabic, Turkish, Polish, Russian and Romanian), and from external community interpreters
catering for Pashtu, Dari, Urdu, Mandarin and Portuguese. The majority of these interpreters
originated from the same countries as the patients they interpreted for and were thus familiar
with the general background of potential study participants (Marshall 2007). This was crucial
for communication, as difficulties are known to arise in goal-oriented talk (such as IC
negotiation) when interlocutors do not share the same background (Gumperz 1982a).
These professional interpreters were asked to translate the simplified IC form prior to the audio
recording. They were given a full explanation on the research project, and were (from their
daily practice) already familiar with the issues it addressed. Interpreters were not asked to
translate the contents of the simplified text literally, but rather to explain its contents in their
language. Cox et al. (2013) refer to such a “personalized translation” as a “transexplanation”.
Interpreters were also asked to address patients in the first person, and to introduce them to the
17
The simplified IC form is presented in Annex 4.
researcher mentioning his name and to imagine they were standing in front of the patient to
enhance direct communication.
The audio-files ranged from 30 to 50 seconds depending on the language. For example,
Moroccan Arabic has no formal status, and does not exist in formal writing. People who only
speak Moroccan Arabic are highly likely to be functionally illiterate. In this case, the interpreter
took additional time resulting in a clear and detailed colloquial explanation of the project. On
the other hand, the Polish transexplanation only required 20 seconds as, according to the Polish
interpreter, most of the Polish immigrants in Brussels are literate.
Each audio-file was uploaded to a tablet computer, marked by the foreign name of the language.
For the pilot, as before, doctors were asked to first assess the context for recording and then to
introduce the researcher. Next, the researcher would ask the patient’s language or show the
tablet screen asking the patient to point at his/her language. The audio-file was then played. A
smartpen (which can write and record audio simultaneously) was used for recording (see
Section 5.1.4). The pen was clearly showcased to the patient to ensure (s)he would understand
what it was used for. Finally, the patient asked whether they agreed to participate in the study,
using elementary expressions such as “yes/no”, “okay/not okay”. In case of doubt, or in case
no audio-version of the IC was available in the patient’s language, the interaction was not
recorded.
After having agreed to the audio-version, participants had to sign the simplified IC forms. It
took time to explain the written document corresponded to the audio-file. Patients who did not
understand the written document (and undocumented patients who were afraid to give up
medical discretion rights) were often reluctant or hesitant to sign. This is why, after
consultation with the ED head, I relied on the ethical principle that if consent cannot be
obtained in writing, an official oral consent should be documented (WMA 2013). 18 Hence, at
the end of each consultation I asked the patient again, with the recording device running, to
confirm (s)he agreed with inclusion of the recorded data in this study.
An interesting feature of the audio-recordings is that hearing a message in their mother tongue
or other familiar language may have a calming effect on patients, as it appeals to their cultural
identity (Gumperz 1982a). As such, the recording can to some extent play the role of what
Leanza (2007) referred to as “welcomer”. As visiting the ED can be very stressful, especially
in a foreign country, this was an important positive side effect of using the audio-files. To
promote understanding, it is crucial that a patient be in a “calm frame” (Bhutta 2004, 772).
Still, a caveat is due. First, as the audio-recorded IC is a unidirectional tool, there is little room
for additional questions by the patient. Neither was it possible to check whether the patient
understood all fine details of the “transexplanation”, given time constraints and linguistic
barriers. This study implied only negligible patient risk. However, for complex clinical studies,
18
World Medical Association Declaration of Helsinki Ethical Principles for Medical Research Involving Human
Subjects
the proposed strategy may not suffice. More research is needed on patient understanding of
complex IC forms (especially for clinical research) in which linguists and medical practitioners
work together to seek practicable solutions (Bhutta 2004). With increasing migration and
societal complexity, the importance of transdisciplinary research can only be expected to
increase. For more complex clinical IC procedures, professional medical interpreters should be
called in.
In the context of this project, the ethics regulations required IC to be asked only from the
patients. However, as the patient companions were involved in the procedure of asking IC, they
had ample opportunity to express their possible objections too or to influence the patient’s
decision. No audio was recorded against the will of the companion. Doctors were also involved
only on a voluntary basis, which implied their consent.
5.1.4 Enabling research by reassuring patients

Initially, a classic Dictaphone was used to record interactions. However, the intrusive presence
of this tool seemed to intimidate patients or cause anxiety. Moreover, given the hectic context
of the ED, in which many patients were seen simultaneously, it was difficult as an observing
researcher to keep up with the note-taking on relevant contextual elements. Multitasking was
compulsory to keep track of the different simultaneously unrolling communicative events.
A practical solution was found in the use of a smartpen 19 that allowed for multimodal
recordings (Figure 5.1). This tool consists of a digital pen with a micro-camera on top and a
special paper notepad. With the micro-camera, the writing was filmed and thus digitalised
along with the simultaneously recorded audio. The smartpen allows one to write down a set of
key words, while the full details of the conversation are audio-recorded. One can draw the
movements that take place in the examination booths, and record relevant positions and gazes
of participants in the interactions. This facilitated note-taking on extra-linguistic elements such
as gaze, position, appearance, and movements (Bezemer 2012).
Another advantage was that the smartpen was more discreet than a classic Dictaphone. By
using the smartpen to note down contextual elements during the doctor-patient interaction, the
“staring” effect of a researcher ostentatiously holding a recording device was avoided.
19
In this project, a smartpen from the company Livescribe (www.livescribe.com) was used.
Figure 5.1: Using the smartpen for note-taking
Source: Author’s own photography
5.1.5 Collecting “broad” contextual elements
“A fieldworker will need to give proof of all the good qualities in life:
patience, endurance, stamina, perseverance, flexibility, adaptability,
empathy, tolerance, the willingness to lose a battle in order to win a war,
creativity, humour and wit, diplomacy, and being happy about very small
achievements. Put that in a job advertisement and you will never find a
suitable candidate.” (Blommaert and Dong 2010: 24)
A first period of observation, before obtaining the permission to audio-record clinician-patient

interactions, served for acclimatisation. The purpose of “acclimatising” (Burns 2000) was to
get a “thick description” (Geertz 1973) of the ED context and to become familiar with the staff.
In particular, this research stage allowed me to understand “as fully as possible” the working
conditions of the clinical staff in the specific context of a highly diverse ED; which is a
prerequisite for ethnographic fieldwork (Rampton, Maybin, and Roberts 2015, 43).
As a field researcher, by “hanging out” in the research setting (Agar 1996), I gradually
integrated into the ED team and graduated from a “pure outsider” to an “acceptable
incompetent” (Fielding 2001). This first period allowed me to collect contextual information
and identify practitioners who were willing to contribute to the research project based on their
expertise.
Framework for data collection

Data collection relied on qualitative methods from the field of linguistic ethnography or
interactional sociolinguistics, with a view to achieving a detailed and holistic description of the
micro-dynamics of miscommunication in the context under study. These methods require the
researcher to look at communicative events in a holistic way, investigate the context (Rampton,
Maybin, and Roberts 2014, 4), and take as many as possible relevant elements into account
(Angelelli 2004, 2000; Roy 1999). Possible sources of misunderstanding are sought both at the
level of the interaction (local inferencing), and at the broader context of the interaction and its
participants (global inferencing), which may include organisational aspects, institutional
aspects, individual elements, practical or logistic aspects, and situational elements.
Such an approach has been advocated, among other scholars, by Cicourel (1987) who argues
that a mere focus on the “narrow” context, limited to what can be inferred from the audio-
transcripts of interactions, such as internal syntactic structure and lexical choices, might lead
to missing out on elements relevant for a proper understanding of a particular talk. Also Roberts
(1996) argues that elements such as the social relationships between actors constitute relevant
“contextualisation clues” that have an impact on understanding and a great explanatory power
in communication analysis. So far, however, applied linguists have often failed to take up these
elements in the study of healthcare interactions (Candling and Candling 2003).
Doing ethnographic research means that the researcher, in an attempt to get a holistic view on
the study object, often spends considerable time (generally at least six months) in the study
setting (Fetterman 2008). Data are collected via a combination of different techniques, such as
observation, formal and informal interviews, and audio-recordings. Contextualization, a non-
judgemental orientation, and accounting for culture and multiple realities are crucial aspects of
data collection (Fetterman 2008, 288).
The selected approach allows the researcher to grasp the highly complex medical and social
context of the ED that surrounds the audio recorded clinician-patient interactions (Paltved and
Musaeus 2012, 772); to register the involved actors’ thinking, feeling and acting; to capture
organisational and team processes; and to support the development of theory with clinical and
organisational implications (Paltved and Musaeus 2012, 772).
A similar approach was taken by Angelelli (2004) who spent 22 months observing in a
Californian bilingual hospital to analyse the roles of professional medical interpreters within
the hospital’s context, the context of the society, and relevant social factors. However, she did
not focus on the ED; and neither did she explore the particular role of third parties
accompanying patients. Slade and co-authors also applied a similar strategy in their
sociolinguistic research on the communicative dynamics of EDs in Australia and Hong Kong
(Scheeres et al. 2008, 14; Slade et al. 2008, 294; Slade, Chandler, et al. 2015, 71). However,
they do not consider language barriers: patients who needed an interpreter were excluded from
their Australian study; and only native Cantonese speakers were included in their Hong Kong
study.
The following framework was used for the collection of relevant contextual elements:
x Individual elements: Concerns personal elements such as education, personal context,
knowledge, assumptions, experience, ideas of doctors, patients and companions about
each other. These psychological and sociocultural contextual aspects will have an
impact on how the participants in an interaction perceive each other and may have an
impact on their mutual interaction (Richard and Lussier 2016b). These elements also
include individual social factors identified by Angelelli (2004), autonomous factors
identified by Li (2013), and the broader socioeconomic and cultural background of the
speakers discussed by Duranti (2005, 26–27).
x Situational elements: comprises elements that arise during an encounter or gathering

(Goffman 1964) which may have an impact on the negotiation of meaning or the
relationship between interactants. They include interactional factors (Li 2013) and
elements such as pitch or tone (Goodwin and Duranti 1992). Other situational elements
include the role dynamics between the interactants; which can change throughout the
interaction (Richard and Lussier 2016b). Also non-verbal behaviour can have a relevant
impact on the interaction (Gerwing and Allison 2009; Gerwing and Landmark Dalby
2014; Goodwin 2000).
x Organisational elements: refers to elements (location, technology, administration) of

coordination, management, organisation, and procedures in the ED that may have an
impact on communication.
Gathering organisational contextual information

I spent the first weeks of acclimatisation in the clinicians’ dispatch room where test results
were analysed and patient files were filled out in between patient visits. Doctors orally agreed
to give a sign when they would go to see a patient, but in practice this rarely happened. I did
not want to disturb them either. Although “just sitting there” felt a bit uncomfortable, the
“backstage” observation of the place “behind the scenes” where the staff worked out of the
sight of patients (Goffman 1959) helped me to record relevant organisational contextual
information of the ED. I saw doctors multitasking, analysing, ordering blood samples, graphs,
X-rays, calling colleagues in other units for hospitalisation of patients and keeping patient files
up to date. Very often, they were seeing different patients simultaneously. They were also often
called for other issues on their internal mobile phones. This aligned with earlier findings by
Laxmisan et al. (2007) that multitasking is a typical characteristic of work in the ED.
Gathering individual contextual information

In general, the staff were very welcoming. They invited me for lunch, and whenever they had
time, they would answer questions. These conversations granted space for explaining them the
purpose of my research, to get to know clinicians and nurses on a personal note, and to learn
about their background, experience, and opinions.
When they first heard I was involved in applied linguistics research, the staff assumed it
focused on bilingualism, which is a rather sensitive issue in Belgium. The hospital was
officially bilingual French-Dutch. Some of the native French-speakers felt uncomfortable with
me (who had been introduced as a Dutch native speaker) for not speaking Dutch fluently. Other
researchers have reported as well that the way in which they present themselves or are
introduced may determine whether individuals under study feel ”threatened” (see e.g. Angelelli
2004: 45; Rampton, Maybin, and Roberts 2014, 19). In overt research it is not unusual for study
subjects to define or place the researcher according to their experience in an initial response
(Hammersley and Atkinson 2010: 63). Once the staff knew that the study focussed on
communication with patients who did not speak (properly) any of the official languages, and
that their own language skills were not a topic of interest for this study, they seemed to become
more relaxed and forthcoming.
Many doctor trainees came from the same university as I did. This helped me to blend in.
During informal chats, some of the staff members explained why, despite their interest in the
research topic, they were hesitant to have the company of a researcher when they were seeing
patients. They feared that this study might report on and criticise their behaviour and judge
their way of communicating with patients. Once they learnt that this study was bound by a duty
of professional confidentiality and that no personal information regarding the observations
would be documented publicly or reported to their hierarchy, their hesitation disappeared.
Thanks to my informal encounters backstage, I gradually became more relaxed in each other’s
presence; and increasingly I was invited to join in when they were seeing patients “front stage”
(Goffman 1959).
For privacy reasons, patients were not interviewed but their individual contextual information
could be derived from doctor-patient interviews (e.g. where they are living/working; how long
they have been in the country; marital status; insurance).
Gathering situational contextual information

As it was often difficult for doctors to predict whether relevant communicative events would
arise, I started to “shadow” individual doctors. This enhanced the building of rapport, and
helped to deepen the understanding of patient trajectories in the ED and the situational context
of clinician-patient encounters. I gained insight into the logic of medical consultations and
treatments. The staff were used to having medical trainees following them, and hence also to
explaining medical rationales and procedural logic. Accompanying doctors in their walks to
and from the dispatch room provided time to interview them. These instant interviews
generated extra contextual, social, and pathological information regarding patients, procedures,
as well as the staff’s view on the communicative event that had just taken place.
5.1.6 Participants as co-researchers

Given the importance of specialised and technical information for communication in the ED,
the “participant as co-researcher” approach was used (Boylorn 2008). 20 An advantage is that
participants contribute additional insights from their own perspective on the study subject
(Boylorn 2008: 599). This “emic” approach is common in ethnographic research (Fetterman
2008).
20
This is a participatory research method that invites participants) to collaborate in data collection and analysis.
Some doctors showed a particular interest in the research project, and were ready to devote
time and effort to contribute formally. The collaboration with doctors was crucial to overcome
methodological challenges. One doctor was instrumental in finding a strategy to overcome
language barriers in obtaining patient IC; another doctor helped collect quantitative data on
patient diversity in the ED. Previously undocumented expertise among medical staff about
communication and language barriers was recorded.
5.2 Nature of the data

The nature of my data (transcripts of ED consultations) implies a specific structure that broadly
follows the typical structured approach of a medical interview. This takes the form of a
sequence of different stages as has been described by Heritage and Maynard (2006, 14). Real
life communication usually proceeds less orderly than textbook communication models. This
implies that communicative events often need to be traced across different stages of the
interaction to establish whether the goals have been achieved. Section 5.2.2 puts forward a
model of a medical consultation as an event with a non-linear structure.
Finally, the data look at multiparty consultations, in which in addition to the doctor and the
patient another participant is present, which I refer to as the patient’s companion. The latter can
take up a multitude of different roles. This adds another layer of complexity to the data. This
is discussed in more detail in Section 5.2.3.
5.2.1 The different components of the ED consultation

Drawing on seminal work by Byrne and Long (1976), Heritage and Maynard (2006, 14)
describe how a medical interview in primary care in general consists of six set sequences,
notably the opening, presenting complaint, the examination sequence (which includes both
history taking and the physical examination), the diagnosis, the treatment, and closing (see
Figure 5.2). A similar structure is presented in the Calgary Cambridge guide to the medical
interview (Kurtz et al. 2003). Each sequence has a communicative purpose, which determines
the nature of the information that is exchanged between the doctor and the patient, as well as
some typical contextual elements and associated communicative features. While a linear
structure is suggested, in practice these stages can take place in another order, or overlap
(Rhodes et al. 2004; Meeuwesen 2003).
Figure 5.2: Components of the medical consultation
Source: Based on Heritage and Maynard (2006, 14)

ED consultations broadly follow the structure of a primary care medical consultation, although
the different stages are likely to be more mixed up, merged or intertwine. 21 Therefore, the
structure of the medical interview above will not be used in my analysis as a normative
framework to evaluate the communication process, but rather as a contextual framework to
describe the nature of the ED consultation, the different communicative goals that are pursued,
and the characteristics of each sequence (Heritage and Maynard 2006).
The latter are determined by the specific approach doctors are expected to follow, and the skills
that they require, to build a relationship with the patient and the patient’s companion(s) where
applicable, and to manage the information flow and structure throughout the consultation.
Examples of how the contents of the medical interview (notably the different stages) interact
with the communication process are discussed in major international reference manuals for
practitioner training such as the Calgary Cambridge guide to the medical interview (Kurtz et
al. 2003) and the Bates' guide to physical examination and history taking (Bickley 2013). In
what follows, the different stages of the medical interview, applied to the ED context, are
reviewed, and I will comment on the communicative characteristics of these stages.
The opening
In this stage, the clinician, the patient, and the companion establish a relationship. This phase
is crucial, as “the success of the medical consultation depends not only on the doctor’s clinical
knowledge and interview skills but also on the nature of the relationship that exists between
doctor and patient” (Watt 2008, 1). The doctor greets the patient, introduces himself to the
patient, and verifies the patient’s name. The doctor seeks the patient’s major identifying data
(such as age, gender, occupation), her personal and social history (e.g. education level,
household composition etc.), and identifies the available sources of medical history (is there a
medical record, was the patient referred and if so, by whom, and is there a letter of referral, is
the patient accompanied, and if so, how does the companion relate to the patient and how
reliable is (s)he as a source of history) (Bickley 2013, 8; Kurtz et al. 2003). Contrary to
ambulant care or booked consultations at for instance a GP’s cabinet, patients arriving in the
ED generally do this without prior notice, without having chosen for a particular physician.
They may be accompanied by family members, police officers or ambulance workers who have
decided to go to the ED (Lachance 2016). The doctor will also assess the reliability of the
patient (relating to his/her memory, engagement level, mood and so on).
In the case of language-discordant medical interactions, the clinician will try to get insights at
this stage into linguistic and communicative skills of the patient and his/her companion. If the
clinician and the patient share no common language, the need for an interpreter is clear; but if
one of them is partially proficient in the other person’s language, determining the extent of the
language gap can be challenging (Schenker et al. 2008, 264). Under tight time constraints, there
is often little attention or time for a careful assessment. It therefore sometimes takes place at a
later sequence of the interaction, or not at all.
21
The typical context in which ED medical encounters occur is described in more detail in Section 1.2.
In the particular context of the ED, patients are likely to be scared and anxious during this
“intake” moment, as they might not be aware of what is wrong with them, of the severity of
their complaint, or of what will happen next (Engel et al. 2010; Scheeres et al. 2008; Slade et
al. 2008; Slade and McCarthy 2011; Le and Ro 1991; Lachance 2016). They may therefore
need to be reassured by the ED staff. A special case concerns undocumented patients who need
urgent medical care, and who may even be more anxious because of their legal status and
potential distrust of public institutions. Again, however, there is often little occasion for this
reassurance due to time constraints and patient crowding (Engel et al. 2010). The process of
informing and reassuring might be further challenged by language barriers between the doctor
and the patient.
Presenting complaint
In this phase, the physician gathers information on the symptoms or concerns that have urged
the patient to seek care (Bickley 2013, 7). This stage is characterized by open questions such
as “What has brought you here?” These questions are crucial, as they offer patients the
opportunity to answer in their own way and shape the content of their response (Watt 2008, 3;
Kurtz et al. 2003). The patient should not be interrupted at this stage, as crucial information is
to be gathered in order to properly conduct the following sequences. If the doctor has received
prior information on the patient’s vital signs and main complaint based on data collected by a
nurse during the intake, the doctor may as well cut this stage short and move directly to the
description of the main complaint (Lachance 2016).
Patients expect the physician to show his/her understanding of the their worries and suffering
and seek reassurance from him/her accordingly (Fosnocht, Swanson, and Barton 2005; Britten
and Shaw 1994). Even if the physician can attend the first concerns by providing a listening
ear to the patient, given the specific context of the ED, when the patient or companion starts to
explain his/her worries, doctors often gently interrupt this process in order to drive the medical
agenda further (Lachance 2016). If patients are not allotted sufficient space to explain their
concerns, these “unvoiced” concerns may become part of a “hidden agenda” that may or may
not surface later on in the medical interview (Bickley 2013, 825). For an efficient
communication process, it is in the interest of the doctor to deal (even if briefly) with all the
patient’s concerns before moving on to the next stage (Lachance 2016).
In the case of a language discordant consultation, communication may be challenged and thus
this initial sequence may be protracted or re-appear at a later point in the consultation. Roberts
et al. (2004) show that, when patients do not master a particular language, they may have
difficulties in properly stressing those parts in their utterances that are most important, for
example during problem presentation. In addition, ways of describing symptoms and oneself
(on a personal note) may differ across cultures. The lack of shared background may not only
lead to misunderstandings but can also produce embarrassing and uncomfortable moments and
frustration.
Examination
The examination stage relies on two major strategies to gather information. On the one hand,
information is extracted from patients and their companions (“history taking”), on the other
hand information is gathered through a “physical examination” of the patient.
The stage of history taking is to collect information on a range of relevant issues (Bickley 2013,
7). The doctor will gather “a clear, and chronologic account of the problems prompting the
patient to seek care” by means of a verbal interview (Bickley 2013, 8). This often follows a
structured Review of Systems approach, during which the presence or absence of common
symptoms related to each major body system is assessed. Here a mixture of open-ended
questions (e.g. tell me more about your headache, sleeping problems) and closed-ended
questions (e.g. when do you have headaches, how many times did you go to the toilet last night,
how many aspirins did you take, what is the colour of your urine…) are needed to gather
information on the patient. The doctor will also ask the patient about his/her health history
(allergies, medication, smoking and drinking habits, past illnesses), and his/her family history
with respect to major illnesses.
Bickley (2013) prescribes that a full description should be obtained for each symptom that the
patient may have, even when several symptoms are presented simultaneously. For a full
description, the physician needs to get details on the seven attributes of a symptom. These refer
to the relevant dimensions of a symptom which need to be discussed in order to obtain a full
description, notably the location; the quality (e.g. continuous or intermittent); the quantity or
severity (often on a scale of 1-10); the timing (including the onset, the duration and the
frequency); the setting in which it occurs; remitting or exacerbating factors and finally the
associated manifestations (Bickley 2013, 70). Typically, these dimensions are addressed in the
medical interview by asking questions such as “where does it hurt?”, “what is the pain like?”
and so on. In Watt’s (2008, 3)view, “Gathering information on the patient’s problems is one of
the most important tasks to be mastered in medicine”… “and the way the questions are asked
can have a profound effect on how the patient reveals information.”
The physical examination is meant to confirm or refute the conclusion from the verbal
interview (Watt 2008, 13). Prior to starting, the doctor is expected to ask the patient permission
to carry out the physical examination and also to explain how (s)he will do this (Kurtz et al.
2003). In case of language-discordant doctor-patient interactions, this phase can sometimes
compensate for the information gaps that remain after the previous stages. From a linguistic
perspective, this sequence predominantly features the imperative (e.g. breathe deeply, stretch
your leg...). From a communicative perspective, it is interesting to note the use of multimodal
communication during this stage, involving both verbal and non-verbal (e.g. palpation)
activities. Moreover, gestures and words are complements rather than substitutes, in the sense
that they depend on each other (Goodwin 2000; Gerwing and Landmark Dalby 2014).
Diagnosis
Diagnosis is one of the most challenging and complex tasks of the physician (Kuhn 2002, 740).
In this sequence, the doctor needs to evaluate the patient’s condition. Next, he needs to explain
his findings, while comparing the results from history taking with the results of the physical
examination and of possible additional tests. This process is to result in a treatment plan or
referral (Bickley 2013, 51). It is crucial that the patient understands what the doctor says about
the pathology and possible test results to allow for patient participation in the next step.
Treatment plan or referral

In this sequence, the physician shares the treatment plan with the patient. As opposed to the
diagnosis, to which patients generally tend to listen passively rather than participating actively
in the conversation, the final treatment decision often involves explicit or implicit negotiation
between the physician and the patient (Stivers 2006a, 279). The treatment can consist of a
referral, medication, discharge, or hospitalisation to which the patient is to agree. The doctor
needs to be sure that the patient understands this information and consequently can confirm or
not the information and to agree or not to a possible treatment or referral. Patient participation
is crucial at this stage; and this requires a smooth flow of information. Linguistic skills that
allow for bargaining are needed.
The different sequences are interdependent. Roberts (2010) finds, based on an analysis of the
treatment decision process in a family practice in a linguistically and culturally diverse
neighbourhood in London, that this bargaining process is not an act as such, but rather the
natural result of an “environment” which is created at the beginning of the consultation. When
such a common environment is not created from the beginning, problems may arise further
down the consultation line as private issues need to be discussed or physical examinations,
which she calls “face-threatening moments”, need to be carried out.
Closing
The closing of the consultation is difficult as the patient still may have many questions of which
some might not have been answered (Bickley 2013, 72). Here it is important that the patient
has agreed upon the treatment plan or diagnosis. In case of a language barrier, such agreement
can be problematic as many issues are likely to remain blurry. It is important that the physician
clearly asks whether the patient has some questions left prior to communicating that the end of
the consultation has come.
5.2.2 The non-linear structure of the ED consultation

Several studies have shown that there is a significant divergence between medical
communication as described by textbooks, and real life situations (Peräkylä and Vehvilfinen
2003; Sarangi 2010, 175; Atkins et al. 2016). As a result, the notion that a medical encounter
is a unique event, which requires constant adjustments on the part of the doctor, the patient and
a possible companion is gaining prominence (Richard and Lussier 2014; Zoppi and McKegney
2002).
Haidet (2007) compares the medical encounter to a jazz ensemble in which the different
participants need to attune to each other to communicate and understand each other’s views
and ways of functioning. As every consultation and every patient is unique, as in jazz, a doctor
needs to improvise by adjusting and departing from the prescribed patterns whenever needed.
While communication skills training and guidelines are essential, Haidet (2007) argues that
doctors should incorporate this knowledge in their own personal styles in the way jazz
musicians do. Salmon and Young (2011) and Richard and Lussier (2014) suggest that medical
communication is inherently creative because medical encounters are characterized by
uncertainty and situatedness. To embrace the complexity of the medical encounter, Richard
and Lussier (2014) and Li et al. (2017) promote the use of dialogic approaches to study and
teach medical communication.
To capture part of this complexity, my analysis approaches the ED consultation as a set of

communicative events in pursuit of different communicative objectives, rather than as a single
communicative event. These objectives comprise not only the medical communicative
purposes that pertain to the respective sequences of the medical interview (such as identifying
the patient’s main complaint or obtaining a description of the attributes of a particular
symptom); they also comprise the communicative purposes that are latently present in the
background such as managing the information flow and building a relationship with the patient.
My analysis focuses mostly on the doctor’s communicative objectives, as doctors are more
likely to follow a structured approach to the information flow (based on their clinical training)
and to set the agenda of the consultation rather than patients or their companions.
This implies that my analysis does not take a sequential approach to the medical interview,
which would imply that speech events are explored in the same order as the one in which they
are produced. Instead, I approach the consultation as an event with a “non-linear” structure and
present the medical interview as a string of speech events in the pursuit of specific
communicative goals which may span several sequences and present multiple interruptions. A
string start from the instance the pursuit of a communicative goal is initiated in the medical
interview, in other words, when a topic is raised. It ends with (in the best-case scenario) the
accomplishment of the communicative purpose, or (in the worst-case scenario) the
abandonment of that purpose.
My approach is illustrated in Figure 5.3. The left panel shows a medical consultation with a
linear structure. Every sequence of the consultation includes a set of communicative purposes,
which are pursued through a series of communicative acts (the black bars of varying length),
until the communicative purpose is reached (marked by a black dot) or abandoned (market by
a white diamond), possibly after several unsuccessful attempts. When the communicative
purposes pertaining to a specific stage have been achieved, the interaction shifts to the next
stage.
Figure 5.3: The structure of a medical consultation
(a) Linear structure (b) Non-linear structure
Communicative purposes Communicative purposes

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
භ STAGE 1
භ STAGE 1
භ භ
STAGE 2 STAGE 2
භ ԉ
ԉ
ѕ ԉѕ
T භ T ԉ භ
I STAGE 3 I ԉ STAGE 3
M භ M ԉ භ
E ѕ E ԉ ԉ
ԉ ԉ STAGE 4
STAGE 4 ԉԉ
භ ԉ
භ ԉභ
ԉ
භ ԉ භ
STAGE 5 ԉ STAGE 5
ѕ
භ භ
Communicative acts pertaining to a chain of acts in the pursuit of a specific local medical communicative purpose
භ Successful finalization of a chain, implying that the local medical communicative purpose has been reached
ѕ Unsuccessful finalization of a chain: the chain is aborted before the local medical communicative purpose has been reached
ԉ Temporary interruption of a chain of communicative acts

Source: Author’s own elaboration
The linear structure is the typical textbook example of a medical interview. However, reality
is typically more complex. The right-hand panel of Figure 5.3 shows how, according to the
non-linear structure, communicative goals may be pursued across several stages of the medical
interview. The string of communicative acts pursuing a medical communicative purpose may
be interrupted (marked by white circles) through a topic shift towards the pursuit of another
communicative purpose; and picked up at a later sequence. The reiteration of a communicative
purpose in a later sequence may lead to the successful achievement, but it may as well be
unsuccessful, and eventually end in the abandonment of the communicative purpose. This
analysis will illustrate how, in language-discordant interactions, these strings of
communicative events are often like an obstacle run, in which participants move from one
hurdle to another in their attempts to retrieving information from, and/or getting their own
messages across to the other participant(s).
The proposed approach is particularly suitable for the context of the ED, as this is a context
where many different activities happen at the same time, and doctors often switch back and
forth between different sequences of the medical interview.
5.2.3 The medical consultation as a multiparty interaction

Our analysis is further complicated by the presence of third parties, in addition to the doctor
and the patient. In in the language-discordant interactions presented in the case studies, each of
the patients have brought a companion to the consultation. This support may take the form of
language support, but it may as well take the form of emotional support (Laidsaar-Powell et al.
2013; Ellingson 2002). If the companion is more proficient in the language that is spoken
during the medical interview than the patient, (s)he is often referred to as an “ad hoc
interpreter”, although that is not necessarily the role (s)he was expecting to fulfil.
In the field of interpreting studies, traditional views of interpreters who have no participatory
role in the interaction have started to evolve in response to research picturing them as essential
partners and co-constructers of the interaction (Angelelli 2012, 432). Based on research on
professional on-site hospital interpreters, Davidson (2000) finds that, rather than neutral
translators, interpreters are active participants in the diagnosis process. Related research by
Bolden (2000) finds that interpreters’ participation in a consultation is guided by their
understanding of the specific biomedical goals of an interaction and not so much by the
supposed task of an interpreter.
Ad hoc interpreters, who have not been trained to follow certain professional guidelines during
their mediation of interactions, are more likely to be active participants in and co-constructors
of the medical interview. In the cases under study, companions are observed to take up different
roles (ad hoc interpreter, advocate, carer, spokesperson, provider of moral support, bystander
…). There is no pre-established or a priori mutually agreed role distribution, and the companion
can change roles over the course of the consultation. This means that very often, instead of
acting as an ad hoc interpreter, or in Goffman’s (1981b) terms as an ‘animator‘ who voices the
patient’s concerns and transmits the patient’s thoughts and answers to the doctor, companions
take up the role of ’author‘ who speaks on behalf of the patient and expresses his/her own
beliefs on what is the patient’s problem. It is important to note that companions do not always
have better language skills than the patient. For all these reasons, I prefer to refer in this PhD
to linguistically discordant multiparty consultations rather than to ad hoc interpreted
consultations.
5.3 Transcription and translation

To prepare the data for the analysis, the recorded doctor-patient interactions were transcribed
using common transcription symbols from interactional sociolinguistics (Roberts 2015;
Gumperz and Berenz 1993). 22 Next, native speakers of the patient and his/her companion’s
22
For an overview of the transcription symbols that were used, see Annex 5.
language, familiar with the medical community of speech for that language, were consulted to
translate and contextualise the utterances and corresponding meanings, and to support the
analysis of the communicational dynamics (Hansell and Seabrook Ajirobutu 1982).
In order to verify the validity of my analysis, I conducted member checking with the physicians
involved in the consultation (Ballinger 2008) and elicited their feedback on the results of my
analysis. Details on how the community of speech and practice were involved in each case, are
presented in Table 5.6.
Table 5.1: Involvement of communities of practice and of speech in data preparation
Case study Community of practice Community of speech
Midnight mastitis The MD who featured in A Spanish professor in linguistics

the interaction A Moroccan intercultural mediator
A second MD A Belgian linguist of Moroccan origin specialist in
Spanish
A Moroccan professor in Spain who is a native
Berber speaker who is specialised in Arabic
languages and speaks French
A native Berber speaker who also speaks Spanish
A Belgian MA student who majored in Spanish
A native Spanish speaking gynaecologist
The patient with a kidney The MD who featured in A Pakistani person with a PhD in pharmaceutical
stone the interaction sciences who speaks both Punjabi and Urdu; familiar
A second MD with both the Belgian and Pakistani context; and has
experience with public health projects in Pakistan.
A British person with a PhD in applied linguistics
who is of Pakistani origin and speaks Urdu and
Punjabi.
A British/Pakistani community interpreter who
speaks both Urdu and Punjabi.
Tuberculosis and The MD who featured in A Polish person with a PhD in pharmaceutical
technology the interaction sciences who is familiar with healthcare in both
A second MD Belgium and Poland experienced with interpreting in
healthcare settings
5.4 Analysis of case studies
5.4.1 Structure of the analytical chapters

In each analytical chapter, I focus on a set of relevant illustrative excerpts of a particular
medical interaction. The selected excerpts present “apt illustrations” in view of the research
question I wish to address (Gluckman 1961). They are illustrative in the sense that they
manifest the same types of communication problems, repair, and accommodation that were
encountered in the whole of the consultation in question and other consultations with similar
contextual features regarding language skills, role dynamics, and miscommunication.
The structure of the three analytical chapters is designed accordingly. First, the broad context
of the interaction is explained. This includes information on individual, situational, and
organizational elements (see Section 5.1.5). In particular, information is provided on relevant
issues such as the amount of foreknowledge the doctor has on the patient’s medical condition,
the background and the language skills of the patient and the companion, and the time of the
day at which the interaction took place.
Next, each of the excerpts is contextually described and analysed in detail using the conceptual
framework set out in Chapter 0 (Rampton, Maybin, and Roberts 2014, 3). The analysis implies
that for each of the excerpts, I specify its communicative goal. Then, the most relevant global
and local contextual observations from a communicative and medical perspective are described
in detail, based on the transcript and on corresponding information gathered through
ethnographic observation interviewing and the study of medical textbooks. This represents the
“communicative context” of the interaction. At the end of each string of conversation excerpts
pursuing a particular medical communicative purpose, I verify whether the predefined
communicative goal has been achieved.
In a next step, the analysis recapitulates how these observations fit into my conceptual
framework, by reviewing the different levels of miscommunication that are observed; and by
summarizing the most important sources of miscommunication and major strategies to
accommodate, prevent, and overcome instances of miscommunication as identified based on
the selected excerpts. Each of the three analytical chapters ends with a conclusion that reviews
the main insights emerging from the analysis regarding how miscommunication arises in a
language discordant multiparty interaction in the ED, how it is addressed, and what its
implications are.
5.4.2 Brief presentation of the selected case studies

The first case study discusses a consultation in which there was a quasi-absolute language
barrier between the patient and the doctor and a partial language barrier between the companion
and the doctor. The patient’s mother tongue was Berber. Her companion, who like her
originated from Morocco, spoke a bit of Spanish. This was the only shared language between
the doctor and the companion. As a result, communication between the doctor and the
companion was very problematic from the beginning. In addition to this, there was a
misalignment in agendas between the doctor and the companion which seemed to reduce the
latter’s engagement in the doctor’s medical agenda. While the language barrier was very
obvious due to its strength, the difference in agenda remained undetected (possibly masked by
the language barrier), and therefore unaddressed.
The second case study discusses a consultation with a male patient from Pakistan who appeared
not to speak Dutch, French, or English. His companion, also from Pakistan, spoke English quite
well. Hence, communication took place between the companion and the doctor in English at
L2 level and between the companion and the patient in Urdu and Punjabi. Doctor and patient
had to rely on the companion for translation. Initially the companion, despite having been asked
to do so, did not take up the role of ad hoc interpreter. After repeated efforts by the doctor,
extensive negotiation and the use of meta-communication, the companion finally took up the
role of interpreter. However, later analysis showed that he made considerable mistakes in his
translations, leading to several instances of miscommunication that remained unnoticed at the
time.
The third case study discusses a consultation with a female patient and her female companion,
both from Poland. Both seemed to be quite fluent in French at first in casual talk, but when the
doctor’s questions became more detailed and technical, their relevant repertoire turned out to
be limited. Each interactant showed, however, a strong willingness to communicate and a
genuine commitment to the interaction, as evidenced by their pro-active display of
understanding and non-understanding. In addition, in this interaction the doctor tried to use for
the first time a multilingual history taking software application to mitigate the language barrier.
5.4.3 Motivation of the selection of the cases

The three case studies differ from each other at the level of different relevant variables, such
as the local context, the patient's complaints, individual communicative characteristics of the
interactants and the nature of the relationship between them, the nature of the language barrier,
the strength of the interactants’ linguistic repertoire, the level of engagement of different
interactants and the discursive flow and strategies they apply. This allows for a comparison
across different contexts of how language discordance can impact on the flow of
communication and the resulting medical care.
Despite these differences, all three cases share some features: first, they are reflections of
language discordant medical consultations in the same ED. Second, in each case, the patient is
accompanied by another person to provide (linguistic) support. Third, the overarching objective
of each of the interactions was to establish an accurate diagnosis of the patient’s complaint and
to propose the most effective treatment or referral. These commonalities help us to show how
different the way or the struggle to accomplish this goal can be, and how this relates to the
context in which the interactions took place.
5.5 Transparency and credibility

In ethnographic fieldwork, information is filtered through the researcher’s impressions and
his/her theoretical research strategy and theoretical orientations (Schwimmer 1996). This
challenges the objective description of the phenomenon under study. Several strategies were
deployed to ensure the credibility of the research that was undertaken. I will discuss these in
what follows.
5.5.1 Transparency
As to enhance transparency, I have attempted to explain in detail what the objectives are of the
research, how I collected the data and how the conceptual framework was developed. I have
also explicitly discussed why I have used case studies and how I define these. Throughout these
discussions, I have sought to explain in detail and explicitly the steps and other procedural and
methodological issues as to foster understanding and avoid vagueness. Also during the data
collection process, I have been transparent towards the research participants, the ethics
committee and other individuals involved. This facilitated the collaboration with the field.
5.5.2 Credibility
Field immersion
As described in section 5.1, I have spent 500 hours in the ED to get familiar with the context
and the people, and to collect the data. Prior to starting the observation in the ED, shadowed
the intercultural mediators of the hospital in question (Cox 2015). Although they are rarely
called in in the ED, this helped me to understand the context of the hospital as such and the
issues surrounding immigrant patients.
In the ED I worked closely with the clinicians to conduct a quantitative survey, to observe, and
to find an ethically sound way to audio-record consultations. I also joined the clinicians during
their lunches and dinner breaks. For ethical and practical reasons, I did not spend time outside
the consultations with the patients and their companions. I got to know the social workers,
security staff and cleaning staff working in the hospital. After a while, when I entered the ED,
I did not enter a completely strange place, as staff members would greet me and have small
talk conversations.
Combining different research angles

The conceptual framework of this research was designed with a view to get an as holistic as
possible perspective on the communication process in the case studies. It combined
perspectives from interactional, medical, and contextual perspectives. Why and how this was
done, is explicitly discussed.
Consulting community of practice, speech, and other specialists in the field

For each of the case studies language specialists, doctors, and fellow linguists were consulted
to get their views on the interaction, and on my analysis of that interaction (see Section 5.3).
Furthermore, parts of the data and analyses have been discussed in data workshops and
presentations at the Centre for Language Discourse and Communication at King’s College
London and at medical and linguistic conferences, to get objective feedback from scholars
external to the research procedure.
Multimodality
The data used in the case studies originate from different data sources, including ethnographic
participant observation, audio recordings, interviews, conversations with clinicians at the ED,
and document analysis of textbooks on medical communication.
Case study: Midnight mastitis 70
6 Case study: Midnight mastitis
This chapter presents the analysis of the first case study, in which a female patient from
Moroccan origin and her companion visit the ED in the early morning because of acute breast
pain. Section 6.1 describes the broader context in which the interaction took place. Section 6.2
explores a set of selected excerpts and identifies the most relevant observations from a
communicative perspective, with a focus on communicative purposes and instances of
miscommunication. Section 6.3 then summarizes the most relevant insights at the level of the
levels of miscommunication; its sources of miscommunication, and the strategies used to
overcome it. Finally, Section 6.4 concludes.
6.1 Context of the interaction

The interaction took place at 2:30 am during a Friday night shift and involved three participants.
The physician (DOC) was a male internist in his early thirties. He started work at 8 am and he
had been working non-stop since then (for almost 20h). Little information was available on the
patient’s background: the physician did not know what language(s) she spoke or how long she
had been in Belgium. The patient’s companion (COM) appeared to have lived in Spain prior
to coming to Belgium. He spoke some Spanish; this was the only lingua franca between him
and the physician. Little information was available on his identity or his relation to the patient.
During the conversation, a variety of languages were spoken, notably Moroccan Arabic,
French, Spanish and Berber. Only after the interaction, the mother tongue of the patient was
identified as Berber, by studying the audio-recordings more closely together with other
language experts. According to the Berber and Moroccan Arabic language specialist who was
consulted for the translation of this transcription, the patient did not understand anything of
what was being verbally communicated in French or Spanish. She seemed to be able to
understand some expressions in Moroccan Arabic, but only to talk in Berber herself.
Table 6.1: Languages used by the interactants in case study 1
DOC French, L2 Spanish

COM L2 French (limited), L2 Spanish (limited), Berber, Moroccan Arabic
PAT Berber
The candidate diagnosis was mastitis, a very painful infection of the breast which often, but
not always, goes together with breastfeeding (NHS 2014). In the case of lactation mastitis,
typically, only one breast is affected. Other common symptoms include breast tenderness,
warmth to the touch, a general feeling of illness (malaise, vomiting), breast swelling, pain or a
burning sensation (continuously or while breastfeeding), skin redness (often in a wedge-shaped
pattern), and fever (NHS 2014).
6.2 Analysis
For the analysis in this chapter, eight excerpts were selected from a longer transcript. The full
transcript of the medical interaction is provided in Annex 6. The excerpts present “apt
illustrations” of the medical consultation under consideration (see Chapter 5). As has been
explained in Chapter 4, a medical interview can be broadly structured into six sequences (see
Figure 6.1). The selected excerpts are taken from stages 2 (Presenting complaint: Excerpt 6.1),
3 (Examination) and 4, 5 and 6 (Diagnosis, Treatment and Closing: Excerpt 6.8) of the medical
consultation.
Within the Examination stage, the doctor attempts to obtain a “full description” of each
symptom the patient may have by gathering information on the seven attributes of pain (its
location; its quality; its quantity or severity; its timing; the setting in which it occurs; remitting
or exacerbating factors and finally its associated manifestations (Bickley 2013, 70). The
selected excerpts address particularly the timing of the pain (Excerpt 6.2 and Excerpt 6.3), its
setting (Excerpt 6.4, Excerpt 6.5), and its location (Excerpt 6.6 and Excerpt 6.7).
Source: Based on Heritage and Maynard (2006, 14) (see Section 5.2).
6.2.1 Presenting complaint

The doctor’s communicative goal in this sequence was to gather information on the concerns
that had urged the patient to seek care (Bickley 2013, 7). Upon entering the examination booth,
the doctor greeted the patient and her companion who were waiting for him. The patient was
sitting on a bed with a grimace on her face and her hands around her right-hand breast, and her
companion stood next to her. By enumerating the languages he was proficient in himself to the
companion, the doctor and the companion identified Spanish as the language they had in
common. The patient did not actively participate in this language negotiation process. It was
difficult however for the doctor to accurately assess the companion’s level of Spanish language
skills. He opened the consultation through an open-ended question (“What is the problem?”),
inviting the patient and her companion to describe their main concerns (Excerpt 6.1).
Excerpt 6.1: Presenting complaint
1. 00:07.0 DOC ¿cuál es el problema? What is the problem?

…
¿usted me va a explicar? Are you going to explain it to me?
2. COM (incomp)
3. COM ..hmm?
4. 00:10.0 COM = = ah oui, Oh yes

... aquí problema/ Problem here
((points at the right-hand breast of patient))
5. PAT = [groans]
6. 00:13.0 COM el...el pecho The... the breast
…
7. 00:16.0 DOC La... la..../ The… the…
como se llama, How does one call this...
= = el pecho? the breast?
8. 00:19.0 COM = pichu..pichu// Brist... brist....
.. = breast.. breast=
9. DOC [ hmm ]
10. COM = = pichu* problema/ Brist problem
… = problem with the breast=
11. DOC Y cùanto tiemp (tiempo)- And how long
= since how long? =
Without awaiting the response to his question, the doctor swiftly turned to the patient’s
companion and asked him whether he would speak on behalf of the patient (turn 1: “Are you
going to explain it to me?”). This can be interpreted as an attempt by the doctor to clarify the
role division in the consultation. Rather than taking a clear lead in the role assignment; the
doctor seemed to leave some discretion up to the companion as to what role he preferred. After
some initial confusion, the companion replied affirmatively and showed the doctor where the
patient’s pain was located by pointing at her breast (turn 4: “Oh yes. Problem here.”). These
first utterances by the companion provided a first illustration of the challenged communication
flow between the doctor and the companion. His pointing gesture was non-redundant (Gerwing
and Allison 2009) as it added necessary information to the word “here” and to the patient’s
non-verbal behaviour.
In general, the patient’s involvement in the consultation was particularly weak. She did not
speak. She sat in an inclined position on the examination table while holding her painful breast
in her hand. Her posture, grimace and occasional groaning seemed to hold the implicature
(Grice 1975) that she was in pain and that she wanted to be relieved from her pain (Richard
and Lussier 2014). To be relieved from their pain is indeed the main concern of most patients
entering the ED (Fosnocht, Hollifield, and Swanson 2004; Fosnocht, Swanson, and Bossart
2001). The patient might have had other items on her agenda, however, if so, she was unable
to express them. She did not undertake any significant effort to be involved in the interaction.
This reflects low self-display. In the literature, low self-display has been related to possible
sociocultural assumptions on the hierarchy and power relations between the doctor and the
patient (Roberts and Sarangi 2005). Based on this case study alone, however, it is impossible
to conclude whether the patient’s low self-display is indeed culture-driven.
The companion hardly took up the role of ad hoc interpreter. In Goffman (1981b)’s terms, his
role was mostly that of an author (expressing his own beliefs) rather than an animator
(conveying the patient’s thoughts and feelings). As a result, the physician only got an indirect
account of the patient’s symptoms.
In turn 6, the companion latched on the patient’s groaning and was looking for a way to express
himself. It was clear however that he encountered severe word-finding difficulties. After a few
moments, he found the word “breast”. When the doctor tried to reply, he encountered similar
word finding difficulties (turn 7: “the… the… how does one call this… the breast?”). Later the
doctor explained that he was looking for the term “mama”, which is more commonly used to
refer to a specific breast in Spanish-spoken medical consultations. The companion replied
affirmatively by repeating the word “pichu” several times (turn 8 and turn 10). The
companion’s limited language skills were reflected in his mispronunciation of the word breast
in Spanish (“pichu” instead of “pecho” in Spanish or “brist” instead of “breast”). The
companion’s utterances presented little syntactic glue (e.g. turn 8, turn 10) (Roberts et al. 2005).
This could partly be strategic, as message reduction is a typical communication strategy for
second language speakers (Dörnyei and Scott 1997).
While the companion participated much more actively in the interaction than the patient, the
doctor indicated during the walk-and-talk interviews that the frequent use of the interjection
“hmm” by the companion (e.g. turn 3) was confusing. It was unclear to the doctor whether this
implied confirmation, or rather a lack of understanding by the companion. In contrast, the
doctor’s own interjection “hmm” in turn 9 is used in a confirmatory way (in the sense of “all
right”). After finding out about the patient’s main complaint or problem, he wanted to move
his medical agenda forward, by closing the problem presentation sequence and moving on to
the examination (Beach 1995; Robinson and Stivers 2001). He did this through a topic shift
towards the timing of the pain (turn 11).
As the doctor found out that the patient’s main complaint was a painful right-hand breast, his
main communicative goal was achieved. The non-verbal behaviour of the patient and the
companion had substantially compensated for the strongly challenged verbal communication
flow. The main complaint was an indication for a first candidate diagnosis, notably mastitis.
Nevertheless, this hypothesis was surrounded by considerable uncertainty because of the
companion’s weak language skills and his sometimes limited and ambiguous replies to the
doctor’s questions. The diagnostic insecurity was reinforced by the doctor’s inability to speak
directly with the patient. The risk of such insecurity is that it requires additional time and
resources to be invested to confirm the diagnosis.
6.2.2 Timing of the pain

Excerpt 6.2 immediately followed upon the previous one in the interaction. As the doctor had
identified the patient’s main complaint, he could move on to the examination sequence. These
first indications hinted at a first candidate diagnosis that the patient might be suffering from
mastitis. Based on this hypothesis, the doctor asked questions to find out more about the
attributes of the patient’s illness, starting with questions on its timing. In the excerpt below,
which pertains to the examination sequence, the physician gathered information on the timing
of the patient’s symptoms. Important aspects included not only the timing of the onset of the
pain, but also whether it came suddenly or gradually, and whether the patient experienced it on
a continuous or rather on an intermittent basis (Bickley 2013).
Excerpt 6.2: Timing of the pain (Part 1)
11. DOC y cuanto tiemp (tiempo)- And how long

= for how long? =
12. 00:27.0 COM = = cuando el dolor* When the pain
cuando el dolor el pichu*, When the pain, the breast...
el dolor y la cabiza::, la barriga::, la todos* The pain and head, the stomach, the
everything
= she has a headache, stomach ache, she has
pain all over=
13. 00:32.0 DOC oké, Okay
<1>
¿desde cuando? ¿cuánto tiempo? Since when? how long?

.. ¿cuántos días*? How many days?
14. 00:35.0 COM = = (pero si,):: But yes
casi casi un año // pero- Almost, almost one year, but
15. 00:35.5 DOC = ¿un * año? One year?
16. 00:38.0 COM = = un año, oui* One year, yes

..
17. 00:38.3 COM pero que me yo cuando la traelo aqui*// But when I bring her here
= but when I brought her over here ==
18. 00:38.6 PAT ((incomprehensible))*
19. 00:43.0 COM y una pastilla* et c’est fini*, c’est fini* and a tablet and it’s finished, it’s finished
= she took a tablet, and the pain was gone =
<1>
20. 00:44.0 DOC a ver/ Let’s see
21. 00:54.0 COM = = un año cuando está, One year ago
..
le duele/ It hurts her
=?????=
22. DOC [ hmhm ]
23. 00:56.0 COM = = pero yo me la he traido aquí* he, But I have brought her over here
..
he pasado un (TAC)…he pasado y (que bebe I went through a CAT, she drank a tablet and
un) una pastilla y c’est fini// it’s finished
<1>
= they did a CAT scan, gave her a tablet and
the pain was gone =
C’est fini// no hay le duele// It’s finished. She has no pain

= she did not have pain anymore=
24. 00:57.0 DOC = = oké Okay
25. 01:03.0 COM = = ahora un año que pasa otra vez/ Now, one year that it happens again.
.. = now, one year on, it happens again=

y cuando el picho mm el bebé, And when the breast mm the baby
… = she is in pain when she is breastfeeding her
baby=
tiene bebé a la maison She has a baby at home.
26. 01:04.0 DOC = = ¿ella tiene bebé? Does she have a baby?
This excerpt illustrates how the flow of communication between the doctor and the patient’s
companion was heavily impaired. The doctor repeatedly asked for information on the onset of
the pain, but the companion always gave different answers; sometimes seemingly irrelevant, at
other times seemingly contradictory with previous answers. For instance, when the doctor
asked for the first time, “for how long?” the companion responded with an enumeration of
possible locations of pain (turn 12). The second time the doctor asked, the companion replied
“almost one year” (turn 14). This confused the doctor. He found it difficult to believe that the
patient had borne a painful condition like mastitis for a full year. Thinking that the present
symptoms were a repetition of what happened last year, the doctor suspected a
misunderstanding. The companion seemed to be in a different frame, talking about something
that had happened a year ago, whereas the doctor was in the frame of wanting to know when
the present illness or pain had started.
As he suspected that the companion did not understand him well, the doctor used common L2
strategies to make his message clear to the companion, frequently repeating and rephrasing his
question (turn 11: “And how long?”; turn 13: “Since when? How long? How many days?”),
message reduction, and echoing the companion’s reply as an accuracy check (turn 15: “one
year?”). His rephrasing of questions in turn 13 could at the same time generate confusion for
someone with limited language skills, as three questions that were not semantically identical
were asked right after each other, without leaving room for an answer in between, potentially
raising doubts over whether these were meant as distinct questions or as rephrasings of one
another.
In the subsequent turns, a misalignment of goals between the doctor and the companion was
revealed. The doctor tried to steer the interaction towards those questions and topics he
considered most relevant to achieve his communicative goals. At the same time, the companion
tried to pursue his own agenda. Such a misalignment between the medical agenda and the
patient (or the companion)’s own agenda is not uncommon in medical consultations; but it can
impact negatively on the communication flow. The first signs of this misalignment were the
companion’s interjections: in turn 14, the companion responded: “but yes...”, implying that he
wanted to say more. Then, after a short pause, he continued (“almost, almost one year, but...”),
which again suggests that he wanted to say more, but he was interrupted by the physician (“one
year?”, turn 15). Hence, while the physician wanted to elicit information on the onset of the
patient’s condition, the companion seemed to want to talk about something else.
The companion’s agenda was revealed more clearly when he suddenly interrupted the doctor
and said “But when I bring her here […] and a tablet and it’s finished, it’s finished”; turn 17,
19). This indicated that the companion (and probably the patient) had mainly come to the
hospital with a view to getting a prescription for similar medication, potentially antibiotics, to
relieve the patient’s pain. The doctor brushed off the companion’s attempt to pursue his
personal agenda through the interjection “let’s see” (turn 20) and continued with his medical
agenda in turn 21. The companion made another attempt in turn 23: “I went through a CAT,
she drank a tablet and it’s finished.” The doctor acknowledged this (“okay”, turn 24). Another
indirect reference to these pills was made at a later stage in the consultation (see Section 6.2.5).
The companion’s repeated use of the French expression “c’est fini’ can be considered as an
illustration of “language switch”, a common phenomenon in language discordant contexts. In
this context, he did probably not use it strategically to strengthen rapport with the doctor (see
Section 4.5.3 on linguistic accommodation), but rather because it was a familiar expression in
everyday talk. The practice of mixing French and Spanish terms did however contribute to
confusion later in the conversation (see Excerpt 6.5), as, in utterances with weak syntactic
glue, unclarity over the linguistic origins of particular words (French or Spanish) led to
ambiguities over their correct interpretation.
The doctor did not react to the companion’s reference to the scan the patient had undergone in
the past. “TAC” is Spanish for a computerised axial tomography (CAT) scan, but belongs to
the rather specific Spanish hospital repertoire. While this was potentially a relevant piece of
information, the physician did not enquire for further details on it. This can to some extent be
attributed to the language barrier. During the member checking session, the doctor involved in
the interaction informed us that he is familiar with the Spanish word TAC, but that he did not
notice the companion was referring to it, likely because of the companion’s peculiar accent in
Spanish and because he was very tired at the time of the interaction. This shows how different
factors may exacerbate the impact of language barriers on the communication flow, even if
someone is familiar with the terminology used.
In turn 25, the companion hinted at the fact that his previous reference to “one year” indeed
implied that the patient had had a similar problem one year ago, and not for the past one year
(“now one year that it happens again”). At the same time, he mentioned that the patient had a
baby at home. As mastitis most commonly affects women who are breastfeeding (Mayo Clinic
2015), the doctor wanted to know more about this baby.
The doctor’s confirmation check in turn 26 can be considered a signal of his understanding,
according to the second turn proof procedure that is often applied in conversation analysis. This
approach is based on the simple principle that interactants display an understanding of each
other in their turns at talk and interaction (Jenks 2014, 279). While interactionally this can be
seen as a successful mutual understanding, the doctor perceived the quality of communication
as low. In the after-action interview on this consultation the doctor said that he was frustrated
by the fact that the companion would not answer his questions and would change topics. The
doctor also doubted the companion’s understanding of his questions, which increased the
diagnostic insecurity. Despite the doctor’s perceived low quality of the communication, he
assessed the fact that there may be a baby as a crucial element for the history-taking. While he
may not have gotten an answer on the timing of the pain, he wanted more information on the
relation between the baby, breastfeeding and mastitis.
He therefore shifted temporarily to another topic (the setting of the pain, Excerpt 6.4). After a
few detailed questions on the patient’s baby, the doctor switched back to the timing of the pain.
This is where Excerpt 6.3 starts.
Excerpt 6.3: Timing of the pain (Part 2)
33. 01:16.0 DOC oké Okay

…
entonces, So
la señora le duele el pecho, The lady’s breast is hurting?
34. 01:16.5 COM == mmh
..
35. 01:19.0 DOC ¿desde ya hace un año? Already since one year?
...
o es lo mismo* que hace un año? Or is it the same as one year ago?
<1>
36. 01:22.0 COM Mis ..mo The same
…
mismo un año// The same one year
.. = the same as one year ago=
37. 01:26.0 DOC oké, ahora es nuevo*? Okay, now it is new?
.. = Now, she is in pain again? =
ahora nuevo? New now?
= Has the pain started over again?=
38. 01:26.5 COM ahora * está nuevo// Now, it is new
39. 01:27.0 DOC = = oké,…¿de cuánto días, cuántos días? Okay, how many days, how many days?
= For how many days has she been in pain? =
40. 01:29.0 COM = = sólo hoy Only today
41. 01:31.0 DOC = = hoy, hoy empezó?... comienza hoy? Today? Today it started? Does it start today?
= did it start today?=
42. 01:32.0 COM = = claro après de midi/ Of course, afternoon
43. 01:34.0 DOC = = y ¿es el mismo* que hace un año? And is it the same as one year ago?
44. 01:35.0 COM = = claro Of course
45. 01:36.0 DOC eso es? That’s it?
<1> = are you sure?=
46. 01:44.0 DOC o::ké// Okay
<1>
y está dando el pecho, and she is giving the breast;
¿está dando la leche a su bebé desde hace un Has she been giving the baby the breast since
año? one year?
...
¿es el mismo* bebé? Is it the same baby?
The doctor first recapitulated the facts (turn 33: “Okay, so the lady’s breast is hurting?”) and
then returned to the previous topic. In turn 35, he made a new attempt to repair the
misunderstanding by highlighting explicitly the difference between the two different meanings
that the companion’s utterance “one year” could have. More specifically, he asked whether the
patient had been in pain for one year (“already since one year?”) or whether the pain she was
experiencing was similar to the pain she had last year (“or is it the same as one year ago?”).
The companion replied by echoing the doctor’s utterance “The same. The same one year” (turn
36). Echoing is a common L2 communication strategy that can be useful to mitigate word-
finding difficulties. However, it sometimes also masks false fluency.
To promote understanding, the doctor continued to perform confirmation checks: “Okay, now
it is new? New now?” (turn 37), through message reduction, self-repetition and self-rephrasing.
The companion responded by echoing “Now it is new” (turn 38). Next, the doctor asked “Okay,
how many days, how many days?”. After the companion replied “Only today”, the doctor
rephrased his question (turn 41: “Today? Today it started? Does it start today?”) as an accuracy
check. This was confirmed again by the companion. Then the doctor returned to the previous
issue with another accuracy check (“And it is the same as one year ago?”). Upon the
companion’s confirmation (“Of course”), the doctor again performed a confirmation check
(“This is it?”).
The high degree of repetitiveness in the doctor’s questions and utterances can be considered a
reflection of the high degree of uncertainty surrounding the conversation, which triggered a
series of accuracy and confirmation checks to ensure the absence of misunderstandings.
Gradually, it got clearer that the doctor could reasonably assume the patient suffered from the
same problem as one year ago, but not continuously since then.
It still puzzled the doctor that the patient had had the same problem one year ago as year was a
relatively short interval between breastfeeding two different babies. Therefore, he eventually
shifted the conversation back to the topic to the patient’s family circumstances, to further
enquire about the setting of the pain (turn 46).
These two excerpts illustrate the string of communicative events in pursuit of information on
the onset of the patient’s pain. Only after considerable effort and prolonged consultation time
the doctor was somewhat assured that the patient’s pain had started that same afternoon. The
language barrier did not only generate a high degree of uncertainty; it also reinforced the
misalignment of frames and of goals between the doctor and the patient’s companion.
Moreover, the doctor still had to rely on what the companion said; while he was unable to
assess whether the companion was a reliable source of history and diagnostic information in
the first place (Lachance 2016; Bickley 2013).
6.2.3 Setting of the pain

As noted above, the doctor shifted the conversation to the setting of the patient’s pain, prompted
by the companion’s casual remark about the patient’s baby (turn 25). The setting of the pain
generally concerns possible environmental factors, personal activities, or other circumstances
that might have contributed to the patient’s symptoms. As there are two common types of
mastitis, one of which (lactation mastitis) is typically associated with breastfeeding, it was
important to confirm whether the patient was indeed breastfeeding a baby. The other type,
referred to as periodical mastitis, may occur with non-breastfeeding women (NHS 2014).
Excerpt 6.4: Setting of the pain (Part 1)
25. 01:03.0 COM ahora un año que pasa otra vez/ Now, on year that it happens again.
y cuando el picho mm el bebé/ And when the breast mm the baby
.. = she is in pain when she is breastfeeding her
baby=
26. 01:04.0 DOC = = ¿ella tiene bebé*? Does she have a baby?
27. 01:04.3 COM = =ah, oui Oh yes
..
28. 01:04.6 DOC ¿le da leche*? Is she giving it the breast?
<1>
29. 01:08.0 COM lichi, cuando está lichi y el le duele// Mealk, when the mealk is there and it hurts
<1> her
= when she breastfeeds, it hurts=
30. 01:08.5 DOC Oké Okay
…
31. 01:13.0 COM me entiendes? Do you understand me?
32. 01:13.5 DOC …más o menos More or less
<1>
((speaks in Moroccan Arabic)) A little
[ laughs ]
To ensure he had understood the companion correctly, the doctor replied to the companion’s
reference to the patient’s baby with an accuracy check, repeating what the companion had said
(turn 26: “Does she have a baby?”). The companion replied affirmatively (“Oh yes”). The
physician asked for additional details using closed-ended questions: “is she giving it the
breast?”. This time, the companion displayed active collaborative construction by responding
(albeit in poor and ungrammatical Spanish): “mealk, when the mealk is there and it hurts”. The
companion mispronounces the word “milk” (“leche” in Spanish). However, as they were
talking about breastfeeding, the speech pattern was predictable and the doctor could easily
guess or understand the companion’s implicature. This implies there was a certain degree of
redundancy in the conversation which compensated for possible ambiguities originating from
pronunciation difficulties (Pinker 2007).
The doctor acknowledged this response briefly in turn 30 (“okay”). Consequently, the
companion performed a comprehension check: “do you understand me?” (turn 31), indicating
that the companion suspected miscommunication (Long 1983) and was concerned about
achieving shared understanding. The doctor responded “more or less” in Spanish, and then
added a short utterance in Moroccan Arabic (“a little”) (turn 32). Strictly speaking, this
utterance in Moroccan Arabic was redundant, as it implied the same message as his previous
utterance “more or less”. However, speaking the language of the patient and/or the companion,
even if it concerns only a very basic expression, can be a tool for convergence and contribute
to strengthening rapport (see Section 4.5.3). The doctor further reinforced this gesture by
adding a smile to his utterance: humour is also a tool for convergence.
Then he switched the conversation briefly back to the timing of the pain to fill the remaining
information gaps on that issue (see Excerpt 6.3) before pursuing a deeper exploration of the
setting of her pain in Excerpt 6.5. During this deeper exploration, the doctor not only enquired
further about the baby, but also about other typical symptoms (fever, vomiting). Finally, he
asked about the possibility of the patient being pregnant, to verify whether her vomiting could
be explained by other factors than mastitis.
Excerpt 6.5: Setting of the pain (Part 2)
46. 01:44.0 DOC o::ké// Okay

(1)
y está dando el pecho, and is she breastfeeding;
¿está dando la leche a su bebé desde hace un Has she been giving milk to the baby for one
año? year?
...
47. 01:51.0 COM no*, no*, no:: No no no
y ahora es (tres,). and now it is (three/a lot) =?=
y hace un año le duele.. and one year ago she was in pain
y ahora le bastante le duele.. and now she is having a lot of pain
48. 01:53.0 COM no poder comer, no poder leche a bebé no/ Not able to eat, not able to milk to baby no
= she is unable to eat, unable to breastfeed the
baby=
49. 01:55.0 DOC = = le duele mucho? Is she having a lot of pain?
50. 01:56.0 COM = = demasiado* Too much
51. 01:59.0 DOC = = Oké.. ¿cuántos años tiene su bebé? Okay, how old is the baby?
...
52. 02:02.0 COM euh:: un año mmh un año casi casi/… Euh one year mmm one year almost, almost
53. 02:04.0 DOC = = oké *… y ¿tiene fiebre? Okay, does she have fever?
<1>
54. 02:06.0 COM ha fiebre* .. oui She has fever, yes
55. 02:09.0 DOC = = sí *, mucho calor? yes, very warm?
56. 02:10.0 COM mucho duele * y/ A lot of pain, and
57. 02:11.0 DOC = = ¿ha vomitado? Has she vomited?
58. 02:13.0 COM = =ha vomitado She has vomited
<2>
<2>
¿comprende español usted? ((He gazes at the Do you understand Spanish?
patient))
...
60. 02:19.0 COM = = el picho es muy raro, rojo me entiendes The brist is very weird, red, do you understand
... me?
= the breast looks very weird, reddish, do you
understand me? =
61. 02:22.0 DOC sí, sí, vamos a ver// Yes yes, we will see.
<1>
62. 02:23.0 COM tiene algo o? Does she have something?
.. = What is the problem?
63. 02:35.0 DOC = = se, señora, ma, madam,
..
¿me comprende? o no? do you understand me or not?
= nada. español? francés? Arabe? Nothing? Spanish? French? Arabic?
= [PAT groans]
((he addresses the companion:))
¿qué habla? árabe. Solo* árabe? What does she speak? Arabic? Only Arabic?
64. 02:36.0 COM = = Árabe Arabic
65. 02:36.5 DOC ¿de donde es? de marruecos? Where is she from? From Morocco?
66. 02:37.0 COM = =Marruecos, Marruecos// Morocco, Morocco
<3>
67. 02:44.0 DOC y cuando- And when
..
en Bélgica, ¿hace mucho tiempo ha estado? in Belgium? Since a lot of time has she been?
= Since when has she been living here?
68. 02:45.0 COM = = 2005 ((in French)) 2005
69. 02:49.0 DOC 2005 ((in Spanish)) ella, ah, oké oké 2005, she, ah, okay, okay
<1> = okay, so she has been here since 2005=
70. 02:50.0 DOC y usted vive aquí o de And you live here since?
= and what about you, since when have you
been living here?
71. 02:54.0 COM = = no* vive aquí, en 2005, que vendrá de No live here, in 2005, she will come from
España hasta aquí Spain to here
(( I think he is talking about the patient)
=?=
72. 02:57.0 DOC = = ah, oké, oké Ah, okay, okay
<1>
73. 02:58.0 COM se (casi) en nuevos, diez años She (almost/got married) in 9, 10 years
<9>
74. 03:08.0 DOC ((feeling PAT’s pulse)) Does she have a baby?
¿tiene un* bebé?
75. 03:08.5 COM …no tengo quatre ((French)), cuatro* No I have four, four
<2> =???=
76. 03:09.0 DOC …cuatro? Four?
<2>
77. 03:12.0 COM ((COM gazes at PAT))
78. 03:16.0 DOC oké, oké, on va voir ça// Okay, okay, let’s see
<2>
79. 03:16.5 DOC [Clears throat ]
80. 03:18.0 DOC ¿ha vomitado? Has she been vomiting?
81. 03:19.0 COM ..Claro Sure
= yes, she has=

82. 03:19.5 DOC ..¿sí? Yes?
84. 03:21.0 DOC ¿mucho? A lot.
<1>
85. 03:21.5 COM mucho A lot
<1>
86. 03:22.0 DOC ¿no está embarazada? She is not pregnant?
87. 03:24.0 COM = = no*, no está embarazada No she is not.
88. 03:25.0 DOC = = ¿seguro? Are you sure?
89. 03:28.0 COM = = solo* tengo bebé de un año I only have a one year old baby
==my baby is only one year old==
90. 03:29.0 DOC …sí pero todo es posible// Yes, but anything is possible
91. 03:31.0 COM No* No
<2>
92. 03:36.0 DOC ça c’est bon ((he looks at the blood pressure)) That’s good
[clears his throat]
((physician puts on his rubber gloves) (clears
his throat))
In turn 46, the doctor again asked three questions in a row, with a different but related meaning,
without leaving any time in between for the patient to reply (“And is she giving it the breast?
Has she been giving the baby the breast for one year? Is it the same baby?”). The companion’s
replies to these questions are rather confusing (turn 47): “No, no, no, and now it is (three/a lot);
and one year ago, she was in pain; and now she is having a lot of pain.” It is not clear which of
the doctor’s questions the “no, no, no” referred to.
As he had intertwined his Spanish utterances with French words previously, it was also unclear
whether the word “tres” he used referred to the Spanish word “tres” (three) or rather to the
French word “très” (“very”). In the former case, it could mean that the patient had three
children. In the latter case, the expression “a lot” would concur with the companion’s next
utterances, stressing how much pain the patient was in. To further underline the gravity of the
patient’s condition, the companion added she had lost her appetite and was unable to feed her
baby (turn 47); and he answered “Too much” (turn 49) to the doctor’s confirmation check “Is
she having a lot of pain?”.
The doctor then revisited the issue of whether the baby that was being breastfed was the same
baby that was being breastfed a year ago. He asked the companion how old the baby was, and
the latter replied that it was almost one year old. This seemed to contradict the physician’s
hypothesis that the baby being breastfed was different from the one which was breastfed one
year ago, based on the companion’s earlier statements (“No, no, no” to the question “Has she
been giving milk to the baby for one year? Is it the same baby?” in turn 46). Possibly, there
was a misalignment in frames, and the “No, no, no” was an answer to a different question. The
doctor temporarily suspended the topic and asked whether the patient had fever and/or had
vomited recently, as these are two typical symptoms associated with lactation mastitis (Section
6.17.1). The companion confirmed this by echoing the doctor. However, his echoing behaviour
also implied a certain risk of false fluency. This might explain why the doctor repeated and
rephrased his question regarding fever a second time as an accuracy check.
As the doctor at that moment still did not have an assessment of the reliability of the
companion’s answers, he decided to try to involve the patient more intensively in the
consultation. He addressed her with the words “Do you understand Spanish?” (turn 59) but
failed to get a reply from her. The companion tried to recapture the doctor’s attention by
highlighting the remarkable visual aspect of the breast (turn 60). The doctor tried to brush him
off (“yes yes, we will see”, turn 61) and to refocus attention on the patient (“Madam, do you
understand me, or not? Nothing? Spanish? French? Arabic?”). Only after this last word, the
patient groaned, which sounded like a confirmation to the doctor. He then addressed the
companion: “what does she speak? Arabic? Only Arabic?”. The latter confirmed that the
patient spoke Arabic. In what followed, the doctor tried to find out some background
information on the patient: where she came from, and how long she had been in Belgium.
Consequently, the doctor did another try at addressing his remaining doubts on the patient’s
babies by asking “Does she have a baby?” (turn 74). The companion replied “No, I have four;
four.” This response provided some possible relevant new information, as it suggested that the
companion was the father of the patient’s children, in other words that she was his spouse. This
took away some of the uncertainty regarding his reliability as a source of diagnostic
information on the patient. Still, some uncertainty remained, as the companion could as well
have erroneously used the first instead of the third person because of his poor Spanish skills.
In addition, his words seemed to contradict his earlier statements that the patient had three
children (turn 47), even more so as the companion again mixed up words in French and Spanish
(possibly due to word-finding problems). Again, this could be the result of a misalignment of
frames between the doctor and the companion. The doctor reacted in a surprised way (“four?”,
turn 77).
The physician switched back to the topic of associated symptoms, notably vomiting. As the
companion had previously indicated that the patient had indeed been vomiting lately, the doctor
wanted to know whether this could have been triggered by a new pregnancy. The way in which
the companion reacted (“my baby is only one year old”, turn 89) seemed to suggest that he
found the mere idea that his spouse could already be pregnant again rather unthinkable. This
again added some support to the hypothesis that the patient had been breastfeeding the same
baby for almost one year now. The doctor then dropped the topic and moved on to the next
stage of the consultation, the physical examination.
Via the interaction illustrated by these two excerpts, the doctor tried to find out information on
the context or setting of the patient’s symptoms. This communicative purpose was however
not fully achieved: even after repeated iterations, there remained confusion regarding the
number and the age of the patient’s children and on how long she had been breastfeeding at the
time of the consultation. The companion had answered the questions regarding the patient’s
other symptoms (vomiting and fever), but the reliability of this information is rather low as
well, as a result of the companion’s poor language skills and his generally low level of
engagement. The tight time constraints however urged the doctor to move on to the next
sequence of the consultation.
6.2.4 Location of the pain

After extracting information from the patient’s companion on the timing and the setting of her
pain, the doctor moved on to the physical examination phase, meant to confirm or refute the
conclusion from the verbal interview (Watt et al. 2008, 13). In a language discordant medical
interview, the physical examination can sometimes also contribute to mitigating information
gaps and ambiguities that remain after the verbal interview. In this sequence, the doctor
combined auscultation (listening with a stethoscope) and palpation (touching with the fingers)
with verbal communication to find out more details on the location of the pain, and whether
the patient had pain in only one breast, or in both. In order to perform physical examination on
her in a patient-centered approach, he also needed to ask her consent.
Excerpt 6.6: Location of the pain (Part 1)
93. 03:41.0 DOC oké, on va regarder, hein Okay, let’s have a look, (see, see)
((Moroccan Arabic:)) tchouf?... Tchouf? (see, see)
((DOC performs auscultation of the breasts))
94. 03:43.0 PAT = =[groans]
95. 03:44.0 DOC = =oké? Okay?
<1> = may I ? =
96. 03:44.5 DOC ¿dónde // le duele? Where does it hurt ?
97. 03:47.0 COM = = (ça va?) * ((addressing the researcher)) Everything okay?
<12>
98. 03:59.0 DOC ((DOC is feeling patient’s breasts))
99. 04:00.0 DOC ((Moroccan Arabic))… Pain?
100. 04:02.0 DOC ((Moroccan Arabic)) Are you having a lot of pain?
101. 04:03.0 PAT = [groans][breathes out]
102. 04:05.0 DOC c'= = est ça? ((DOC is palpating)) This is it?
103. 04:11.0 DOC <2,5>{incomp}
105. 04:14.0 PAT = [groans]
106. 04:20.0 PAT =[groans]
107. 04:21.0 DOC …et ici non?...((pointing at the other breast)) And here not?
108. 04:23.0 PAT …((Berber:)) {incomp}
109. 04:25.0 COM …Tambien le duele This is also hurting her =This one also hurts =
110. 04:28.0 PAT = = ((Berber:)) {incomp}
111. 04:28.3 COM ici (a veces) ou c’est ceci? Here (sometimes) or it is here
((pointing at the other breast))
112. 04:28.6 DOC …Como? What do you mean?

113. 04:30.0 COM = = Et- And…
114. 04:31.0 DOC = là c'est normal? ((talking about breasts while Is this one normal?
palpating))
115. 04:40.0 COM = =No*.. No, how many … it is okay that she changes it
cuanto ((incomprehensible)) está bien que la to the other one, now the other one is fine.
cambia para otro/.. ahora la otra está bien// = it is okay if she uses the other breast, which
<2,5> is fine now =
116. 04:45.0 COM no hay problema pero ici, pase There is no problem, but this one
<3> = with that breast there is no problem, but with
this one there is =
117. 04:46.0 DOC solo esto*? Only this one?
..
Aquí* le duele ((pointing at breast))? Is she having pain here?
118. 04:48.0 COM = = no, solo esto No, only this one
119. 04:51.0 DOC = =oké Okay
The doctor put on his rubber gloves, took the earpiece of his stethoscope from his neck, and
placed it on his ears. These non-verbal signs announced the start of the physical examination.
To get the patient’s consent for the examination, the doctor addressed her first in French (turn
93) and then by drawing on some basic Moroccan Arabic phrases he knew (“see, see”). As he
was going to touch an intimate part of her body, he wanted to check first whether she was
comfortable with that. He also asked, “okay?” in turn 95, which could be interpreted as “may
I?” The patient groaned in response. The doctor perceived this as a backchannel that indicated
the patient had understood what would happen and gave her implicit consent and started with
the auscultation of the breasts.
The examination process took place principally between the doctor and the patient. The
companion remained silent for a while. It was not always clear who the doctor was addressing
with his questions. For example, his question in turn 96 could mean both “Where do you have
pain?” (polite form) or “Where does she have pain?”. Eventually, the question remained
unanswered. When the researcher turned around to respect the patient’s privacy (turn 97), the
companion asked him whether everything was fine with the researcher.
After having auscultated the patient’s breasts for about 12 seconds, and still not having received
an answer to his question, the doctor again tried to foster convergence with the patient by using
some basic Arabic expressions. He pronounced the Moroccan Arabic word for pain with a
rising tone, as to formulate a question, while he touched different parts of her breast with the
chest piece of his stethoscope. He then rephrased his question in a more elaborate way (“are
you having a lot of pain?”, turn 100), still in Moroccan Arabic. The patient groaned, and the
grimace on her face suggested she was indeed in pain. When the doctor performed a
confirmation check in French (turn 102: “This is it?”), the patient did not react. She only
groaned in response to his utterances in Arabic; suggesting that she did not understand French
at all.
The doctor then moved to the other breast. He pointed at it and asked “And here not?” in
French, to which the patient replied something in Berber. Possibly as a translation of what the
patient had said, the companion latched on the patient’s utterance with the words “This one
also hurts” (turn 109). The patient said something in Berber again. The companion then verified
what she was telling her by pointing at one of her breasts and asking “here, or is it here?” (turn
111). The doctor got confused by what he was observing and reacted by the open-ended
question “what do you mean?” (turn 112). When the companion started to say something
(“and…”), he was interrupted by the doctor who moved into a simpler closed-ended question
(“is this one normal?”). The companion answered with a stressed “no”, which suggested that
the other breast was painful as well.
The companion then went on to produce an ungrammatical utterance in Spanish (turn 115)
which literally meant “how many … it is okay that she changes it to the other one, now the
other one is fine”. The first word was probably mispronounced: instead of “cuanto” (“how
many”) he might have intended to say “cuando” (“when”). The lack of prosody in the
companion’s utterance added to its ambiguity. The utterance could have meant that the
companion was saying that the patient had been able to use the other breast (that was not
affected) to feed her baby. It could also have meant he was seeking the doctor’s confirmation
that it was appropriate to do so. However, either interpretation raised confusion as they both
suggested that only one breast was affected, in contrast with the companion’s previous reply
(“No”) to the doctor’s question. This could again point at a misalignment in frames at some
point, where the companion was implying something else with his “No” than the implicature
the doctor had assumed.
The companion’s next utterance (turn 116: “There is no problem, but this one”) was equally
confusing, but also seemed to hint that only one breast was affected. As a self-accuracy check,
the doctor asked “Only this one?”. In the absence of an answer, he pointed at the patient’s
second breast and asked: “Is she having pain here?”. After the companion’s next reply (“no,
only this one”), it seemed that the doctor had finally achieved reasonable reassurance that only
one breast was hurting. However, as the challenged communication flow had generated
substantial uncertainty, the doctor returned to the same question again at a later stage in the
consultation, to ensure no misunderstanding would remain.
Between turns 120 and 167, the doctor continued with the palpation of the patient’s breasts,
performed a lung exam (auscultation) on her back, and asked additional questions relating to
what she had eaten that day; and whether she felt like vomiting. When the companion denied
that, the doctor touched again briefly upon the location of the patient’s pain, asking the patient
and/or the companion to show him where it hurted the most (Excerpt 6.7).
Excerpt 6.7: Location of the pain (Part 2)
167. 06:40.0 DOC muy bien Very good

(2)
y, ¿puede mostrar donde le duele más? Can you/she indicate where it hurts most?
168. 06:40.5 COM …((Berber)) He is telling you: where do you have (feel)
most pain.
169. 06:42.0 PAT = = ((Berber)) Here
((she touches her right-hand breast))
170. 06:46.0 DOC …mostrar, la, Show, there
oké,.. c'est tout// Okay, that’s all
171. 06:46.5 COM = = solo el pecho Only the breast
For the first time in this interaction, the companion took up the explicit role of interpreter, and
translated the question to the patient (turn 168). The patient replied “here” in Berber and
touched her right-hand breast to show the exact location of the pain. The doctor was satisfied
with this answer and concluded the topic with the interjection “okay, that’s all”. The companion
emphasised once more “only the breast”, to make sure the doctor had understood.
Subsequently, the doctor returned to asking questions on the patient’s possible fever, a topic
that he had previously already made some inquiries on. After that, he started to wrap up the
consultation (see Section 6.2.5).
During the physical examination phase, the doctor found out that the patient’s pain was mostly
limited to her right breast. In this sense, his communicative purpose was achieved. The doctor
also managed to get the patient’s personal (albeit non-verbal) consent to manually examine her
breasts, which contributed to the more general objective of patient-centeredness. However, due
to the problematic communication and the conflicting messages given by the companion,
considerable ambiguity remained. On the positive side, for the first time during this
consultation, the patient seemed to be more involved: the combination of verbal and non-verbal
communication appeared to somewhat mitigate the absolute language barrier between the
doctor and the patient.
6.2.5 Diagnosis, treatment, and closing

In the final stage of the consultation, the doctor typically decides upon the most plausible
diagnosis of the patient’s illness, and explains this to the patient. He is also expected to propose
a treatment plan to the patient, which can consist of a referral, a therapy, or hospitalisation, and
to seek the patient’s agreement. While patients usually do not interfere with the diagnostic
decision, the treatment decision ideally involves explicit or implicit negotiation between the
doctor and the patient in a patient-centred approach (Stivers 2006b). Patient participation, and
hence a more than basic understanding by the patient of the provided information is crucial.
Finally, at the end of the consultation, the doctor typically asks the patient whether he or she
has any questions left.
Through the examination sequence, the doctor had extracted as much information as possible
on the timing, the setting, and the location of the patient’s symptoms; despite the near
impossibility of direct communication with the patient and the significant language barrier
between the doctor and the patient’s companion. The doctor reasonably assumed that the
patient indeed suffered from mastitis and decided to refer the patient to a gynaecologist for
further check-ups. Excerpt 6.8 shows what happened next.
Excerpt 6.8: Diagnosis, treatment and closing
178. 07:02.0 DOC = voy a hablar con mi colega/ I am going to talk to my colleague
179. à7:02.5 COM = = huh?
180. 07:04.0 DOC Ginecólogo* …entiendes? Gynaecologist, do you understand?
181. 07:04.3 COM = = mmh?
182. 07:04.6 DOC = = Hm*?
183. 07:07.0 COM …¿cómo? What?
184. 07:08.0 DOC mi colega/ My colleague
185. 07:08.5 COM = = Mmh
186. 07:10.0 DOC = =mi compañero, *ginecólogo// My companion, gynaecologist
187. 07:10.5 COM = = Mmh
188. 07:11.0 DOC ¿comprende*? Do you understand?
189. 07:12.0 COM = = mmh, claro Sure
190. 07:15.0 DOC = = mi compañero, que se ocupa de las My companion, who takes care of women
mujeres,.. especialidad// Specialty
191. 07:18.0 COM = = Mmh
192. 07:18.3 DOC = = ¿ya*?... y… probablemente, es una Ok? It is probably an infection here
infección aqui
193. 07:18.6 COM = = Mmh
194. 07:26.0 DOC Ya* eum. < 1,5>
probablemente, hay que darle antibióticos/… She is probably to be given antibiotics
¿comprendes?
195. 07:27.0 COM = =(incomp)
196. 07:27.5 DOC = = huh?
197. 07:28.0 COM = = (incomp)(muito sabes?) A lot you know =?=
198. 07:29.0 DOC = = huh?
199. 07:31.0 COM = = sabes como va las cosas pero yo no sabes You know how things go but I know nothing
nada
200. 07:33.0 DOC = = sí*.., estoy explicando* Yes, I am explaining it
201. 07:34.0 COM = = Ay* claro oké Ah sure okay
202. 07:36.0 DOC …oké? Okay?
203. 07:36.5 COM …Claro Sure
204. 07:40.0 DOC …y vamos saber si hacemos euhm un And we will know whether we will do a lab
laboratorio// ==we will find out whether further medical
checks will be needed==
205. 07:41.0 COM …Mmm Mmm

206. 07:45.0 DOC …comprende? Do you understand?
<3>
207. 07:45.5 COM Merci Thank you
<1>
208. 07:49.0 DOC ¿ella tiene preguntas?... Does she have questions?
puedes preguntar// Can you ask?
209. 07:49.3 COM = = ((talks in Berber)) He is asking you, whether you want him to
give you one so you can take it.
210. 07:49.6 PAT …((talks in Berber)) why?
211. 07:51.0 COM ..((talks in Berber)) It has to do with the pain you are having.
212. 08:01.0 PAT …((talks in Berber)) Yes
213. 08:01.3 COM …{incomp} huah?
214. 08:01.6 DOC = =huah?
215. 08:02.0 COM {incomp}
216. 08:02.5 DOC = = ((talks in Moroccan Arabic)) Yes
217. 08:03.0 COM = = sí,
218. 08:04.0 DOC = =oké, ça va.
The doctor was struggling to explain to the companion that he was calling in his colleague, a
gynaecologist. He tried to make himself understood by repeating, rephrasing (“my colleague”,
“my companion”, turns 184 and 186; “gynaecologist”, “my companion, who takes care of
women, speciality”, turns 186 and 190) and through explicit comprehension checks (“do you
understand?”, turn 180 and turn 189). He explained that the patient probably had an infection
(turn 192); and that she might need to take antibiotics (turn 194). The companion’s engagement
in this final part of the conversation was relatively limited. He reacted passively to what the
doctor was explaining him, mostly by backchannelling (“mmh”, “sure”, …). When pushed
further by the doctor’s interjections (“huh?”), he replied “You know how things go, but I know
nothing” (turn 199). This statement could reflect low self-display (Section 4.4.6), indicate he
felt some frustration over his lack of control, and relate to how he perceived the hierarchy and
power relations in the consultation (Roberts and Sarangi 2005). It could as well be an implicit
hint that he was not very enthusiastic about going for further tests, but the companion did not
openly protest the proposed way further.
Just before closing the consultation, the doctor asked the companion whether the patient had
any questions left (“Does she have questions?”, turn 208). Later discussions with a language
expert showed that the companion did not translate this question to the patient. Instead, he told
her in Berber “He is asking you, whether you want him to give you one so you can take it”
(turn 209). This is an extreme case of a non-rendition (Wadensjö 1998). The reason he did so
is not clear. On the one hand, he might have done this strategically, to pursue his own agenda
of obtaining pills. Possibly, he saw the doctor’s question as a final chance to get the pills they
seemed to have come for – rather than having to go through further tests. In this case, the
misunderstanding would be caused by a misalignment of goals (see as well Section 6.2.2). On

the other hand, it could have resulted from the companion’s lack of textual comprehension of
what the doctor was asking (Li 2013) or, in a broader sense, of pragmatic failure, notably a
misalignment in frames.
The question seemed to confuse the patient, who asked, “Why?” in Berber. After some
additional explanations by the companion, the patient responded “yes”, suggesting that she
would indeed like to have some medication. However, the companion encountered difficulties
conveying this to the doctor. The only word he managed to convey in a comprehensible way
was “yes”. Again, it remains unclear whether this was because of his weak language skills; or
rather because he was not sure it would be appropriate to reiterate his request for pills. The
“yes” seemed to be interpreted by the doctor as a signal that the patient did not have any further
questions, and hence, that the consultation could be closed.
As the doctor did not understand Berber, he was not aware that the companion had not
translated his question to the patient. It must be noted, however, that even if the companion had
interpreted the question correctly, the patient would have been unable to express her concerns
or questions adequately, given that the interpreter had not shared with her the information he
received on the diagnosis and treatment proposal.
The communicative purpose of negotiating the treatment or way forward in a patient-centred

approach, allowing the patient some negotiation space, was not achieved. The patient did not
get a proper account of the doctor’s proposal. The companion did not translate the proposal
into Berber, and the doctor also did not ask him to do so. Due to his low self-display, it is not
clear whether the companion had himself understood the diagnosis and treatment proposal. The
doctor also failed in asking the patient whether she had any remaining questions. This
misunderstanding remained unnoticed.
6.3 Discussion
This section describes the different levels of miscommunication that could be identified
throughout the selected excerpts. It then goes on to review the main sources of these instances
of miscommunication, and the strategies participants in the concerned interaction were using
to prevent, accommodate or repair it.
6.3.1 Levels of miscommunication

In the case under study, several types of miscommunication were observed. Some of these had
only very localized implications, and were repaired almost immediately; others had broader
implications that surfaced at different points throughout the consultation. Coupland et al.
(1991)’s framework allows to classify the different observed instances of miscommunication
into different levels, accounting for the nature of their origins and implications (see Section
4.3).
First, the most basic level of miscommunication (“Level I miscommunication”) refers to the
inherent ambiguity and incompleteness of messages, which gives rise to misunderstandings
that remain unnoticed. By definition, these micro-misunderstandings are not even detectable
by researchers.
Level II miscommunication refers to minor misunderstandings which are recognisable, but are
in general not problematic. As both the doctor and the companion were speaking in a language
that was not their mother tongue (notably Spanish), they frequently encountered word finding
difficulties, or made errors against grammar, vocabulary, pronunciation or prosody that did not
necessarily change the meaning of what they wanted to say or give rise to misunderstandings.
Listeners could guess, based on the context and the frame of the conversation, which message
speakers wished to convey. In such cases, correct language use can be considered redundant
(Pinker 1995; Seife 2007) in the sense that it is not indispensable to achieving the
communicative purpose of a conversation.
A particularly high incidence of level III miscommunication was observed, which can be
attributed to a lack of specific skills (be it language skills or gaps in communication skills) or
ill will, and can potentially be remediated through training. First, there was a complete language
barrier between the doctor and the patient, who did not share any common language. The only
way direct communication was possible was through non-verbal communication. Thus, the
only moment when the doctor could get across to the patient (albeit still to a very limited
extent), was at the time of the physical examination, when the doctor used basic expressions in
Moroccan Arabic complemented by physical gestures (palpation), and the patient reacted to
this by groaning (or not groaning), which in this way also had a semantic value.
Communication difficulties were further aggravated by the fact that the patient displayed low
engagement. She did not undertake any significant effort to establish a direct connection with
the doctor, for instance by asking the companion to involve her and translate the doctor’s
questions to her. The communication flow between the doctor and the patient was so weak that
it could be considered a communication breakdown, as the patient was mostly excluded from
the interaction.
Verbal communication relied fully on intermediation by the companion. However, the

language barrier between the doctor and the companion was also substantial and made the
communication troublesome. In fact, it was a double language barrier, as Spanish, the main
language of communication, was a second language for each of them. Especially the
companion’s limited Spanish language skills and repertoire for participation in a medical
consultation hampered the interaction. The language barrier was so strong that it was often
impossible to get messages across to each other; and often the doctor felt that different
statements by the companion were contradicting each other. To some extent, this could be
triggered by undetected misalignments in frames, e.g. in the case discrepancies occurred
between the question the doctor was expecting the companion to answer, and the question the
companion was intending to answer. It is not surprising that because of the high incidence of
level III miscommunication, the doctor failed to achieve several of his communicative goals.
This brings us to level IV miscommunication, which considers the strategic goals of

communication. Section 6.2 discussed the communicative goals of each sequence, and whether
these had been achieved or not during the consultation. Those instances where the
communicative goal was not achieved, can be considered as level IV miscommunication.
Sometimes, the doctor was even not aware that his goal was not achieved, as the language
barrier hampered the detection of misunderstandings.
While the doctor obtained some information on most of the relevant attributes of the patient’s
pain, the poor quality of the communication flow led to substantial diagnostic insecurity. While
some issues seem to be reasonably clear when studying these transcripts at the time of the
consultation, in the hectic environment of the ED, the doctor had the impression that many
more information gaps remained, and that the reliability of a large part of the information he
had received, was questionable.
Another, more general, communicative goal of the consultation is to build rapport with the
patient and his/her possible companion(s). The language barrier strongly complicated this
process. Nevertheless, the doctor did converge to the companion by addressing him in
Moroccan Arabic as to create rapport. Rampton (2005) calls this process of addressing people
in their own language to create rapport “crossing over”. The fact that he did this in a humorous
way, triggering a laugh from the companion, helped.
There was also a misalignment between the doctor’s communicative goal and the companion’s
objectives, which I earlier referred to as a discrepancy between the doctor’s “medical world”
and the patient’s “lifeworld” (Mishler 1984). Such discrepancies commonly occur in the ED
(Lachance 2016). The doctor tried to pursue his medical agenda and to gather information on
the different attributes of the patient’s pain, while the companion tried to convince the doctor
that the patient was suffering heavily and should be given a tablet. His own agenda distracted
his attention and reduced his engagement in responding diligently to the doctor’s questions,
often reinforcing confusion.
The line between different sources of miscommunication is difficult to draw, as these factors
interact. Because of the language barrier, the doctor did not notice that his communicative goal
(e.g. of asking whether the patient had any questions remaining) was not achieved. Moreover,
because of the language barriers, the doctor did not realise the misalignment of goals and of
the communication frame between him and the companion. From a broader perspective, based
on the member-checking exercise it turned out that the doctor ascribed most of the
miscommunication to the companion’s linguistic skills, while in fact there were many other
factors at play.
6.3.2 Sources of miscommunication

This section follows the framework presented in Section 4.4 to identify the main origins of the
instances of miscommunication identified in the considered interaction.
Lack of background information or pretextuality

Most of the information the doctor could extract from this consultation came directly from the
patient’s companion, not from the patient. As the doctor did not have any information on the
ties between the patient and her companion (aside from a slight hint during the consultation
that the companion was the patient’s husband), it was very difficult to assess the reliability of
the companion’s replies.
The doctor could have avoided this by starting the consultation by enquiring more extensively
about the patient and the companion’s background and their language skills. The doctor only
asked about the patient’s language skills later in the consultation (Section 6.2.3) in an attempt
to speak with her directly. Instead, the doctor proceeded immediately to the problem
presentation. As the ED context is commonly characterised by time constraints, doctors are
often observed to drive forward the medical agenda as soon as possible. During the member
checking of this analysis, the doctor in question mentioned that he usually does enquire about
the relationship between the patient and other people present at the onset of the consultation.
In this nightly consultation, however, he had overlooked to do so, as he was very tired.
Lack of engagement
The patient did not engage verbally in the consultation. She only showed her presence and pain
through non-verbal behaviour, notably her body position, her facial expression, and by
groaning and moving slightly up and down as she sat on the examination table. This was a
result of the language barrier between her and the doctor, aggravated by the language
discordance between the companion and the doctor, the role taken up by the companion in the
interaction, and the patient’s own low self-display.
The companion did not show very high engagement in the conversation either. His answers
were often confusing, he did not always make clear whether he had actually understood the
doctor’s questions, and his replies were often limited to interjections such as “hmm” or to
echoing the doctor’s questions. For instance, when the doctor asked question about the setting
of the patient’ pain (8.2.3), he responded affirmatively to the doctor’s question, without really
giving more details (e.g. “does she have fever?” “She has fever, yes.”; “has she vomited?” “she
has vomited”; “a lot?” “a lot”). This passive behaviour could be a result of the companion’s
limited language skills. During member checking, the doctor explained that this happened so
often that he eventually started to call into question the reliability of the companion’s answers.
In the last excerpt, the companion also showed low self-display by telling the doctor “You
know how things go, but I know nothing”.
Language barrier effects

There was no common language between the patient and the doctor, implying that she could
not directly communicate verbally with the doctor. The doctor did some efforts to address her
in French and Spanish, but this was in vain. The only response he was able to get from her was
a groan when he uttered some basic expressions in Moroccan Arabic.
There was also a substantial language barrier between the doctor and the companion. The only
language they shared was Spanish, of which neither of them was a native speaker. The
companion’s Spanish skills were particularly limited and implied that he had considerable
difficulties to construct a proper sentence in Spanish. The companion’s utterances often
presented little syntactic glue; and his pronunciation of relevant terms and mixing of French
and Spanish terms was often misleading. During member checking, the physician mentioned
he had the impression that the companion did not understand what he was saying. From his
own side, the doctor also felt uncertain about what the companion exactly implied by his
utterances. This feeling is common: language barriers in healthcare have been reported to
increase ED physicians’ diagnostic insecurity (Garra et al. 2010; Hampers et al. 1999). As he
was speaking in a second language, the doctor also frequently encountered word-finding
difficulties. The language barrier also contributed to his failure to detect the potentially relevant
reference to a specific medical term (CAT) made by the companion.
Interpreting errors
When the doctor requested the companion to take up the role of interpreter and ask the patient
whether she had any questions remaining (Excerpt 6.8), the companion told the patient “He is
asking you, whether you want him to give you one so you can take it.” This is an extreme case
of a non-rendition. It was not clear whether this incorrect translation was due to the
companion’s textual incomprehension, or rather a strategic move, with a view to re-establishing
the agenda he was pursuing earlier, notably obtaining medication for his spouse. When the
patient replied affirmatively, the companion seemed again unable to convey her message back
to the doctor due to his weak language proficiency and potential other concerns.
Violation of interactional rules

At times, overlapping in turn-taking was observed. In Excerpt 6.1 the doctor interrupted the
companion while the latter had not finished talking yet, as evidenced by the word “but…” at
the end of his utterance. This signals that the physician was not allowing the companion
processing time in the interaction. Understandably, given the time pressure, the doctor wanted
to maintain a steady communication flow. However, the ensuing risk was that the companion
was not able to express his main concerns. In this particular case, the overlap in turn-taking
related to a broader misalignment of agendas between the doctor and the patient’s companion.
The structure of the questions formulated by the doctor also may have caused
miscommunication. At several instances, the physician asked different questions in a row
without leaving the companion or the patient processing time or time to answer. Such behaviour
violates the basic interactional rules of questions and answers (Section 4.4.2). Especially in a
context of language barriers, this way of asking questions makes it more difficult to come to a
shared understanding. While the doctor’s intention clearly was to promote the companion’s
understanding by repeating and rephrasing his questions, this practice more likely has added to
the confusion, rather than mitigating it. In a language discordant encounter, it is particularly
important to leave additional processing time for the listener to understand the questions and
to formulate an answer across a language barrier. Understandably, however, this is not always
possible under the tight time constraints of the ED environment.
The doctor also mentioned during the member checking session that he was very tired and
becoming slightly impatient over the course of the interaction, as he had already spent
considerable time with the patient and her companion without having been able to gather a lot
of information. He admitted this could have contributed to some of the observed violations of
interactional rules.
Misalignment of frames
There were a few instances of miscommunication which seemed to arise from a misalignment
of frames between the doctor and the companion. In particular, the companion often gave
conflicting answers to similar questions asked by the doctor. This happened for example when
the onset and duration of the patient’s pain was discussed; when the doctor asked whether the
baby that was being breastfed at the time of the consultation was the same as the one being
breastfed a year earlier; when the doctor asked how many children the patient had; and when
he inquired whether the patient’s left breast was affected as well. The language barrier and the
relatively weak engagement of the companion in the interaction complicated the alignment of
frames considerably.
Misalignment of goals
In Excerpt 6.2, a divergence in agendas between the doctor and the companion was revealed.
On the one hand, the doctor tried to pursue his medical agenda, which implied the collection
of a specific type of information on the patient’s symptoms, as to establish a diagnosis. On the
other hand, the companion gave the impression that he would prefer to skip the interview and
proceed fast to his main objective, which seemed to be obtaining pills to ease the patient’s pain.
Studies have found that the main expectation of patients entering the ED (and their possible
companions) is to be relieved from their pain (Fosnocht, Swanson, and Bossart 2001) and often
preferably through medication for pain relief (Toma, Triner, and McNutt 2009). The
appearance of a discrepancy between what Mishler (1984) calls the patient’s “lifeworld” and
the doctor’s “medical world” is relatively common in the ED (see Section 4.4.8). Because of
this goal misalignment, the companion gave a slightly impatient impression, and he did not
seem fully interested in responding correctly to all the physician’s questions. He was not fully
engaged in the interaction.
Another explanation for the companion’s behaviour may be that he tried to help the doctor, as
he believed he already knew the diagnosis, since the patient had had a similar problem about a
year ago. Studies have found that it is common for patients who suspect to have a recurrent
problem, to state this in the initial stages of the medical interaction (Heritage and Robinson
2006, 52).
In addition, while the doctor wanted to find out under which conditions the pain arose, the
companion seemed focused on trying to justify his visit to the ED by emphasizing the suffering
of his presumed spouse. He did not waste any opportunity to stress her.(“it hurts her”, turn 29;
“and one year ago she had been in pain and now she was having a lot of pain”, turn 47; “she is
unable to eat, unable to the breast breastfeed the baby”, turn 47; “too much”, turn 49; “a lot of
pain”, turn 55; “the breast looks very weird, reddish”, turn 59). In the meantime, he paid little
attention to the specific questions of the doctor.
It is not clear whether the physician was fully aware of this misalignment of goals at the time
of the interaction. In general, language barriers (level III miscommunication) make it more
difficult to recognise and remediate other sources of miscommunication such as the
misalignment of goals (level IV miscommunication). Therefore, it is often observed that
physicians attribute miscommunication to language barriers, rather than recognizing potential
broader sources.
Problems of role dynamics

The patient was verbally excluded from the history taking. The doctor and the companion were
the key speakers in the conversation, and the patient remained an unaddressed participant. The
companion did not try to act as an interpreter nor as an animator (a “sound-box”) of the patient’s
immediate ideas. He did not consult the patient and refrained from co-constructing the
responses to the doctor’s questions with her. Instead, he acted as an author of his own ideas
and experiences. This gave the doctor only an indirect account of the patient’s feelings, which
made the communication process more fragile. From his side, the doctor allowed the
companion to do this, as he did not enter a more explicit role negotiation with the companion
for most of the consultation, for instance by asking him to consult the patient before responding
to questions. The patient also contributed to this role dynamics as she did not do any significant
efforts to be more involved in the conversation.
6.3.3 Prevention, accommodation, and repair
Generic communication strategies in medical interviews (Q&A)

The way questions are asked is a crucial feature in the negotiation of meaning, which can either
facilitate or hamper understanding. Medical practice handbooks prescribe the use of open
questions to invite patients to voice all their concerns, and the use of closed questions to address
more specific issues, such as technical descriptions of symptoms. In language discordant
environments, each of these question types come with their own challenges. Open-ended
questions are more difficult to reply to for patients (or their companions) who need to rely on
a limited linguistic repertoire. Conversely, closed-ended questions imply the risk of false
recognition as they allow the respondent to give a “passive” response without really having
understood the question.
The studied interaction provides a salient illustration of these issues. When the doctor asked
open-ended questions (for instance, “What is the problem?”; “Does she have questions?”), the
companion clearly had problems (including word-finding difficulties) to answer these. In
response, the doctor frequently turned to closed-ended questions. However, these generated a
risk of false recognition, given the companion’s relatively passive behaviour. The doctor
explicitly mentioned his doubts about the reliability of the companion’s replies as a result of
the latter’s habit of responding to every question by echoing the doctor.
As the successfulness of a question strategy was shown to strongly depend on the local context,
the best approach in a language discordant (or any other communicatively challenged) context
seems to be to vary one’s question and answer strategies and perform various confirmation and
accuracy checks. The doctor in the studied case applied this strategy frequently. For instance,
during the physical examination, the doctor rephrased his question repeatedly, varying the
Q&A strategy. He started with an open-ended question (“where does it hurt?”), then rephrased
his question in a closed-ended way (“Pain?”, “Are you having a lot of pain?”, “This is it?”,
“Are you having a lot of pain?”, “and here not?”, “is this one normal?”, “only this one?”, “is
she having pain here?”). As a final check, he again rephrased his question in an open-ended
way (“can you/she indicate where it hurts most?”).
Communication strategies in a second language

At several instances, the doctor and the companion used communication strategies that are
commonly applied by non-native speakers to make their messages clearer. For example, as he
suspected that the companion did not understand him well, the doctor tried to repeat and
rephrase his question in order to promote understanding (turn 11: “and how long?”; turn 13:
“since when? how long? how many days?”; turn 15: “one year?”; turn 35: “already since one
year?”; turn 37: “now it is new?”; turn 39: “for how many days?”; turn 41: “did it start today?”).
This repeating and rephrasing has been found to be a typical accommodation technique by
physicians who speak to foreign language-speaking patients (Valero-Garcés 2005, 19).
During the member checking sessions, the doctor told us that he had the impression the
companion hardly understood him based on his deficient pronunciation and the lack of
syntactic glue in many of his utterances. Therefore, the doctor performed multiple confirmation
and accuracy checks. He did not only repeat and rephrase himself (e.g. today? Today it started?
does it start today?”, turn 41), but he also repeated and rephrases the companion’s responses
(e.g. “one year?” in turn 15). Confirmation checks can be seen both as a signal of potential
miscommunication, and as an attempt to prevent or repair a possible misunderstanding (Pica
1994; Long 1996; Long 1983).
Both the doctor and the companion at times used explicit comprehension checks (“Do you
understand me?”, turn 31 and turn 59; “Do you understand me, or not?”, turn 62) to see if their
counterpart was still on board. They also used message reduction (focusing utterances on a set
of key words) and circumlocution or paraphrasing. For instance, as the doctor seemed
concerned that the companion would not understand technical terminology such as
“gynaecologist”, he rephrased the concept (“my companion, who takes care of women,
specialty”).
Non-verbal communication
Non-verbal communication often makes an important part of the communication process. In
language-discordant medical consultations, non-verbal communication becomes particularly
important given the constraints to verbal communication imposed by the participants’ lack of
skills in each other’s language. In the studies excerpts, the patient used non-verbal
communication (body position, grimace) to express her pain. Non-verbal communication was
also instrumental in the identification of the exact location of the pain. Complementing verbal
communication with visual illustration is a common strategy speakers use in L2 contexts
(Dörnyei and Scott 1997).
Meta-communication: role assignment

As was extensively discussed above, the patient remained excluded from the interaction for
most part of the consultation. This was a result of the companion taking up the role of the
patient’s spokesperson rather than of interpreter; and the general acceptance of this role by the
doctor and the patient. At the onset of the consultation, the doctor asked some clarification on
the role distribution (“Are you going to explain it to me?”), but he did not go into negotiating
an alternative division before the final stage of the consultation.
For a patient-centred medical interaction, it is of critical importance that the patient is informed
of the diagnosis and consulted on the proposed treatment or way forward. This requires patient
involvement in this last stage of the consultation. However, based on what was observed in this
interaction, the patient was not involved at all in the delivery of the diagnosis and the treatment
decision: during this part of the conversation, the doctor only addressed the companion. Only
at the very end of the consultation (the closing), when the doctor wanted to know whether the
patient had any questions, he asked the companion very explicitly to take up the role of
interpreter (“does she have questions? can you ask?”). The companion then asked a question
to the patient which was actually very different from what the doctor had intended to ask.
Ideally, the companion should have consulted the patient much more often during the
consultation, to reassure the doctor that his agenda was aligned with the patient’s, and that she
supported everything the companion said. However, the interaction clearly shows that even if
patients bring companions with better language skills than themselves, it is not evident that
these will spontaneously act as ad hoc interpreters. They may need to be convinced by the
doctor through explicit role negotiation. Furthermore, even if companions agree to act as
interpreters, they may not possess sufficient language skills to do so.
Linguistic accommodation
Now and then, the doctor drew on his (albeit limited) linguistic resources in Moroccan Arabic
(which he believes is the patient’s mother tongue) to foster convergence or strengthen rapport
with the patient. In combination with non-verbal communication, this appeared to be successful
at thinning the language barrier at least during the physical examination sequence. The doctor
also used humour (through his facial expression) as a tool for convergence.
6.4 Conclusion
In the case under study, the communication process was very troublesome. Miscommunication
was present at different levels of Coupland et al. (1991)’s framework: level II, where
miscommunication is mostly harmless and less problematic, but more often at level III, where
miscommunication arises mostly from gaps in language or communication skills or from ill
will or a lack of engagement. There were also some instances of miscommunication at level
IV, which mainly results from the non-attainment of strategic communicative goals.
The case study illustrates how miscommunication in language discordant consultations tends
to have more origins than just language barriers. Notably, I have identified miscommunication
originating from a lack of background information (which generated uncertainty), lack of
engagement (which significantly reduced the quality of the communicative process), violations
of interactional rules (often driven by time constraints and tiredness), interpreting errors, and
misalignments in frames and goals. These causes are not always easily to disentangle from the
pure language barrier effects, as they generally reinforce and exacerbate each other. Due to the
lack of a sufficiently solid linguistic repertoire, it was much more difficult to negotiate
meaning, detect misalignments and repair them.
In this challenging context, the main interactants used several communication strategies to
prevent and/or overcome miscommunication. While the quality of the information exchange
remained relatively low, the participants could achieve some of their communicative goals, and
interactionally achieved mutual understanding at times, attested by the second turn proof
procedure. Hence, their communicative efforts were certainly not in vain. However, the
ethnographic after-action interviews with the doctor show that this achieved understanding did
not always correlate with the doctor’s perceived quality of the understanding and that
consequently diagnostic uncertainty surrounded the interactional understanding. The language
barrier thinned out significantly during the physical examination, when the doctor could
combine non-verbal resources with a few basic expressions in Moroccan Arabic, and the
patient provided feedback by groaning.
Some parts of the communication flow could have been improved through applying more
diligently specific communication strategies, such as establishing a shared background by
gathering more complete information from the onset of the consultation. Communication could
as well have been improved by establishing a more adequate role distribution from the start,
with the companion acting more diligently as an ad hoc interpreter instead of voicing his own
beliefs. Finally, some confusion could as well have been avoided by sticking more closely to
basic interactional rules, by leaving sufficient processing time between questions.
Nevertheless, it is unlikely that these strategies would have sufficed to bring the quality of the
communication flow up to a reasonable level. The only way that communication in this
consultation could have been drastically improved is by invoking an interpreter fluent in Berber
and in French and familiar with the medical environment and jargon.
Case study: The patient with a kidney stone 101
7 Case study: The patient with a kidney stone
This chapter presents the analysis of the second case study, in which a patient from Pakistan
and his companion visit a doctor in the Emergency Department to find out what causes the
patient’s pain in his lower back region. Section 7.1 describes the broader context in which the
interaction took place. Section 7.2 presents the analysis of the interaction excerpts, including
details on their local context, and a discussion of the most relevant observations from a
communicative perspective. Section 7.3 summarizes the insights gained, focusing on the
different levels of miscommunication identified in the analysed excerpts; the sources of
miscommunication, and the strategies used by the participants to overcome miscommunication.
Section 7.4 concludes.

The interaction took place on a Friday late afternoon, towards the end of a shift (16h-18h)
between three men: a physician (DOC), a patient (PAT), and the latter’s companion (COM).
The physician was a 30-year-old native Dutch speaker, specialised in emergency medicine.
The patient was a Pakistani man of 25–30 years’ old. As he appeared not to speak Dutch,
French or English, his companion did part of the talking for him and acted now and then as an
interpreter. The latter was a Pakistani man in his forties who spoke English, Urdu and Pakistani
Punjabi.
During the interaction, different languages were spoken. The doctor spoke with the companion
in English, as this seemed to be their only “lingua franca” or shared language. The companion
mostly talked in Urdu to the patient, but sometimes he used Pakistani Punjabi.23 The doctor did
not understand any of these two languages. His major strategy to overcome the language barrier
between him and the patient was to use the patient’s companion as an ad hoc interpreter.
DOC L2 English
COM L2 English, Urdu, Pakistan Punjabi
PAT Pakistani Punjabi, Urdu
The doctor met the patient for the first time in the examination booth, where the patient was
sitting in an inclined position on the examination table with his hand on the left-hand side of
the lower back region. This gave a non-verbal signal about the location of his pain. His
companion stood next to him.
23
The language specialist that was consulted for the analysis of the transcripts of this case, suggested that there
might as well have been a certain language barrier between the patient and the companion, as frequent language
switches between Urdu and Pakistani Punjabi could be observed in their bilateral conversations.
Figure 7.1: Typical body position for patients with a kidney stone
Source: Wikivisual (2016)
The patient’s first indications about his health complaints led the doctor to think that he suffered
from a renal colic, triggered by a kidney stone. A kidney stone causes pain when it moves
around within the patient’s kidney or passes into the patient’s tube or ureter connecting the
kidney and the bladder. Typical symptoms are: (i) severe pain in the side and back, below the
ribs; (ii) pain that spreads to the lower abdomen and groin; (iii) pain that comes in waves and
fluctuates in intensity; (iv) pain on urination; (v) pink, red or brown urine; (vi) cloudy or foul-
smelling urine; (vii) nausea and vomiting; (viii) persistent need to urinate; (ix) urinating more
often than usual; (x) fever and chills if an infection is present; and (xi) urinating small amounts
of urine (Mayo Clinic 2016a). 24
7.2 Analysis
For the analysis in this chapter, five excerpts were selected from a longer transcript which is
provided in full in Annex 7. The excerpts present “apt illustrations” of the medical consultation
under consideration (see Chapter 5). As has been explained in Chapter 4, a medical interview
can be broadly structured into six sequences (see Figure 7.2). The selected illustrative excerpts
are taken from Sequence 2 (presenting complaint: Excerpt 7.1), Sequence 3 (examination:
Excerpt 7.2, Excerpt 7.3, and Excerpt 7.4) and Sequences 5 and 6 (treatment negotiation and
closing: Excerpt 7.5).
In the context of the examination, medical communication guidelines prescribe that the doctor
interviews the patient to obtain relevant information with regard to seven major “attributes” of
the patient’s symptoms, notably the location, the quality (e.g. is the pain continuous or
intermittent); the quantity or severity (often on a scale of 1-10); the timing (including aspects
such as the onset, the duration and the frequency); the setting in which the symptoms occur;
remitting or exacerbating factors (in other words circumstances or actions which relieve the
symptoms or make them worse) and potential associated manifestations (see Bickley 2013, 70
and Section 4.2). In the excerpts considered in this paper, the doctor asks for information on
24
Mayo Clinic is a non-profit medical practice and research group based in Minnesota. On its website, it provides
extensive patient health information, including descriptions of diseases and conditions, and symptom checkers.
the quality (Excerpt 7.2) and onset (Excerpt 7.3) of the patient’s pain, and on possible remitting
or exacerbating factors (Excerpt 7.4).
Source: Based on Maynard and Heritage (2006, 14) (see Section 5.2)
In what follows, the selected excerpts are presented in chronological order. For each of the
excerpts, I specify its communicative goal and describe the most relevant observations from a
communicative perspective, with specific attention to instances of miscommunication, their
origins, and the strategies used by the participants to the interaction to prevent or repair such
instances. At the end of each excerpt’s discussion, I also verify whether the premised
communicative goal has been achieved.
7.2.1 Presenting complaint

In this sequence, the physician tried to gather information on the symptoms or concerns that
urged the patient to seek in order to get some first indications on the patient’s complaint as a
basis for the detailed examination which follows next (Bickley 2013, 7). Prior to the start of
this fragment, the doctor had greeted the patient and introduced himself and the observing
researcher. The patient’s consent was asked for recording the medical interaction.
Excerpt 7.1: Presenting complaint
1. DOC What happened?

<3.5>
2. COM He has got a problem there* with the kidney** problem. ((he points at the patient’s
lower back))
<2>
3. COM He has pain* here.
<2>
4. DOC Since when?
<1>
5. COM Euh...
…
6. COM Last four days//.
<2>
7. 00:00:17.3 DOC Is it the right* side?
..
8. 00:00:18.7 COM Euh… "rait na lepht ?” ((mix Urdu-English: right or left?))
<2>
9. 00:00:21.8 PAT “lepht" ((mix Urdu-English: left))
10. 00:00:23.8 COM = ((incomprehensible)) ((Urdu)
11. 00:00:28.8 COM = Left side//

<3>
12. 00:00:25.7 DOC Since four days?
13. 00:00:27.1 COM = = Yes//

<3>
14. 00:00:31.3 DOC Is it continuously* or once a lot and then afterwards <1> almost no pain.
In turn 1, the doctor asked “what happened?”, inviting the patient to explain what his main
complaint or reason for the visit was. It is usual for this sequence of the medical consultation
to be characterised by open questions, which offer the patient the opportunity to answer in his
own way and shape the content of his response (Watt 2008, 3). The companion took the floor
and told the doctor, while pointing at the patient’s back, that the patient was in pain. He
immediately added his own candidate diagnosis (turn 2), namely that the patient had a problem
with his kidney (Stivers 2002). As was confirmed through member-checking with the physician
afterwards, the patient’s non-verbal signals and the companion’s suggestions indeed led the
doctor to think of a kidney stone as a candidate diagnosis. Kidney stones are commonly
associated with “severe pain in the side and back, below the ribs” (Mayo Clinic 2016a).
At first, the patient’s companion took up the role as a carer or advocate of the patient, rather
than strictly as an interpreter. Now and then he asked the patient a question for clarification.
This was helpful as far as the information conveyed by the companion reflected adequately the
patient’s thoughts and feelings. In this initial sequence of the interaction, the doctor accepted
the roles taken up spontaneously by the patient and the companion and went on with a set of
broad introductory questions (turn 4: “Since when?”).
The companion took a few seconds to process the doctor’s question (turn 5: “Euh…”) and then
replied in turn 6: “last four days.” The hesitation may either imply that the companion was not
sure what the doctor was asking, or that he was not sure about the answer and tried to count
back the days since the patient had started complaining of pain in his back. After a short silence,
the doctor proceeded to the next topic on the location of the pain. He did so by performing a
closed ended question, “is it the right side?” while pointing at the patient’s lower back. As such,
he performed a confirmation check of what was said by the companion and suggested by the
patient’s non-verbal behaviour.
At this point, the companion consulted the patient before responding. One potential reason for
this is that he was confused by the doctor’s question, which referred to the patient’s right-hand
side, while the patient was in fact pointing at his left-hand side. The companion put the doctor’s
question to the patient as a twofold question in a mix of Urdu and English, which is quite
common in Pakistan: “right or left?” (turn 8). His translation was not exact, and could be
classified as an “expansion error” under Flores et al.’s (2012) classification of interpreting
errors. However, while the semantic content may have been different, the pragmatic content
seemed to be maintained. It is possible that the interpreter intentionally changed the utterances
to facilitate understanding by the patient, and avoid that the patient would be confused in the
same way as he was confused himself by the doctor’s question. The patient replied “left”; and
the companion conveyed this message to the doctor in turn 11.
After four seconds of silence, the doctor echoed the companion’s earlier answer regarding the
onset of the pain as to double-check it (turn 12: “Since four days?”). The confirmation check
may have been a sign that the doctor suspected a possible misunderstanding. The companion
confirmed subsequently (turn 13: “yes”). After a brief silence, the doctor then shifted to the
next sequence of the medical interview, in which he carried out a differential diagnosis (turn
14).
One can conclude from these observations that, thanks to a combination of verbal and non-
verbal communication, the doctor succeeded in achieving a broad idea of the patient’s main
complaint, notably the candidate diagnosis of a kidney stone, which sufficed as a basis for
further examination. He did not know exactly, however, how reliable the information was that
he gained, as the patient’s companion mostly provided it, and was not confirmed by the patient
himself. As the doctor did not have any details on the relationship between the patient and his
companion, he did not know how reliable the companion was as a source of history. Quality of
the pain: is it continuous or intermittent?
The second excerpt, which immediately follows the previous one, is part of the examination
sequence. The doctor started interviewing the patient and his companion to obtain a description
of specific attributes of the patient’s symptoms. This is necessary for a differential diagnosis,
which distinguishes the patient’s candidate diagnosis (notably a kidney stone) from other
potential conditions with similar clinical features. The doctor’s goal was to collect information
on the quality of the patient’s pain, more particularly on whether it was continuous or rather
intermittent, occurring at (irregular) intervals (Bickley 2013). In the case of a kidney stone, the
pain is typically intermittent (see Section 7.1).
Excerpt 7.2: Quality of the pain: is it continuous or intermittent?
15. 00:00:36.9 COM No/
…
Continuous//
<1>
16. 00:00:38.6 DOC Ask him
…
17. 00:00:39.6 COM Yes ((incomprehensible))
18. 00:00:40.6 DOC = Is it continuously* .. also at night or is it ((snaps fingers)) sometimes heavy
pain and afterwards no pain.
<1.5>
19. 00:00:49.2 COM Is it continuous or intervals ((in Urdu))
…
20. 00:00:52.6 PAT "thora thora chakkar aya tha" ((little little pain to start with, then it intensifies, I
am seeing stars == I feel dizzy))
21. 00:00:55.2 COM = ahh

<1>
Euh now he has no pain/
<1>
After some time/
…
He has a big* pain//.
<1>
22. 00:01:01.6 DOC Yes
(1)
My question is, the past four* days/
23. 00:01:04.9 COM = Yes
24. 00:01:05.7 DOC = Was the pain continuously*?
25. 00:01:07.6 COM = Continuously
26. DOC = Or was it sometimes big* pain after which no* pain?
27. 00:01:11.7 COM = Sometimes big* pain sometimes, sometimes small* pain.
(4)
28. 00:01:15.9 COM Sometimes big* pain, but sometimes big* pain.
(1.5)
29. 00:01:20.0 DOC hmhm
30. 00:01:20.4 COM = Ahhhh
(5)
Euhh
(.)
((incomprehensible))
31. PAT = ((incomprehensible))
32. 00:01:28.1 DOC did he have fever*?
(1)
33. 00:01:30.0 COM no fever//
(2)
34. 00:01:30.0 PAT ((incomprehensible)) ((in Urdu))
In turn 14, the physician asked about the quality of the patient’s pain by formulating a twofold
question: “is it continuously, or once a lot and afterwards almost no pain?”. Rather than using
the word “intermittent”, which could have been difficult to understand for patients with limited
English resources, he described the meaning of the term in more simple English language,
using a combination of lexical items to replace a relatively technical term. Such circumlocution
or paraphrasing is a strategy which may be used by non-native speakers in case they encounter
word finding problems for a particular concept (Dörnyei and Scott 1997). It may also be used,
as is the case here, as an instrument of linguistic accommodation, to facilitate the patient’s
understanding.
The companion did not translate the doctor’s question to the patient but instead answered
himself that it was continuous (turn 15). Unsatisfied with the fact that the companion had not
consulted the patient prior to responding, the doctor insisted (turn 16: “Ask him”) that the
companion translated the question to the patient. This can be considered as explicit role
negotiation. To facilitate proper translation by the companion, the doctor repeated his question
(turn 18). In doing so, he slightly rephrased it and added a new element (“also at night”),
although this could generate confusion. In response, the companion changed footing and put
the question to the patient in Urdu (turn 19). However, in doing so, he reduced the doctor’s
utterance, omitting potentially important concepts such as “also at night” and “sometimes
heavy pain and afterwards no pain.” This qualifies as an interpreter error.
The patient answered in Urdu that the pain was first “little little” but that it afterwards
intensified to the extent that he was “seeing stars”, in other words, that he was feeling dizzy
(turn 20). The doctor did not understand what the patient had said and had to rely on the
companion’s English translation which suggested that the patient was not having pain at that
moment, but that he suffered from “big pain” at irregular time intervals. This implies a strong
editorialization (interpreter error) of the patient’s initial response, as well as a reduction, as the
interpreter failed to convey the patient’s message that the pain got so strong that it made him
dizzy.
The companion’s utterances were ungrammatical, but their implicature was understandable by
the doctor. Still, he was confused by what the companion had said, as the notion that the patient
did not feel any pain right now was at odds with the patient’s inclined position and his contorted
facial expressions. For the doctor, it was very challenging to assess the quality of the
communication process, as he did not have a view on the quality of the translation.
Because of his uncertainty regarding the validity of the companion’s answer, he performed
another confirmation check (turns 22, 24, 26). This time he tried to split the two-fold question
into two parts (turns 24 and 26), leaving some time in between the two parts to allow the
companion more time to process the question and translate the respective parts to the patient.
However, the companion added further to the confusion by merely echoing or repeating the
doctor’s questions in English, without translating them to the patient. Moreover, he repeated
the doctor’s utterances in a confirming tone, suggesting that the patient’s pain was at the same
time continuous and intermittent, which was a contradiction in terms.
Note that, in a conversation without a language barrier, the doctor often infers recognition from
a non-response by the patient. In contrast, in a conversation that is mediated by a companion,
it is difficult for the doctor to find out whether a non-response reflects recognition by the patient
(yes, indeed, this happened over the past four days) or rather zero rendition by the companion
(the reference to the time frame was not translated to the patient). As the doctor could not
recognise the companion’s interpreting errors, he could not repair potential misunderstandings
stemming from these errors.
Urged by his own time constraints, the doctor then shifted topics, even though he was still not
sure of having received a correct answer to his question. He asked whether the patient had
fever. The companion replied for the patient, saying that the patient did not have any fever.
Without asking the companion to double-check his answer with the patient, the doctor went on
to the next topic (see Excerpt 7.3).
Based on this excerpt I find that, despite repeated efforts, the doctor did not get a clear answer
on whether the patient’s pain was continuous or rather intermittent. Because of the language
barrier, he also missed a potentially important symptom (dizziness) which was reported by the
patient but not translated by the companion.
7.2.2 Onset of the pain: Is it sudden or gradual?

The following excerpt, which follows immediately upon the previous one, is also a part of the
examination sequence, in which the doctor seeks a full description of the patient’s symptoms.
In this excerpt, he explored the onset of the patient’s pain, more particularly whether the pain
came up suddenly or gradually. This is a key element in the process of differential diagnosis.
In the case of a kidney stone, the onset is usually sudden.
Excerpt 7.3: Onset of the pain: Is it sudden or gradual?
35. 00:01:33.5 DOC Did it came ((snaps fingers)) suddenly?

..
Or did it came little by little//
…
36. 00:01:38.7 COM Little by little//
<2>
37. 00:01:41.7 DOC Ask* him//
…
38. 00:01:41.7 COM Yes eeehhhhe, he live with me, I know// ..
39. 00:01:45.0 COM He live* with me*
40. 00:01:46.1 DOC = = Yes yes yes
…
Ok
…
But he can have other feelings than you think*.. So you must*
…
Translate//
41. 00:01:53.3 COM = =Hmhm//
42. 00:01:53.3 DOC = =Did the pain came ((snaps fingers)) suddenly or.. did it came
…
first a little pain and afterwards more:: and afterwards still* more::.
43. 00:02:02.0 COM = "kabhi kabhi hota hai ya ahista ahista chalta hai” ((is it now and then or does it
come slowly slowly?))
…
44. 00:02:05.1 PAT ahista ahista kr kay chalta hai … ((it comes slowly slowly))
45. 00:02:04.7 COM = aha
46. PAT … ((incomprehensible))
47. 00:02:10.5 COM = solosolosolo ((slowly slowly slowly)) and then fast* pain
48. 00:02:12.6 DOC = Huh?
49. 00:02:15.9 COM his pain // solosolosolo
50. 00:02:18.3 DOC = solosolosolo?
= What does it?
51. 00:02:20.0 COM = Little little little//

52. DOC = = OK
<4>
53. 00:02:26.1 COM ((incomprehensible))((in Urdu))
<1,5>
54. 00:02:30.2 PAT ((incomprehensible))((in Urdu))
55. 00:02:30.5 COM Euh
In turn 35, the doctor formulated a twofold question to find out whether the pain started
suddenly or gradually. To facilitate the companion’s comprehension, the doctor paraphrased
the semi-technical term “gradual” as “did it come little by little?” (turn 35). Instead of putting
the question to the patient, the companion replied by repetition, “little by little”. As the
requested information was quite specific and therefore the doctor would like the companion to
consult the patient and act as an interpreter, the doctor again engaged in explicit role assignment
(“Ask him”, turn 37). However, the companion met his request with protest, arguing that he
already had all the necessary information as he was a cohabitant of the patient (turn 38). The
implicature of his utterance seemed to be that he did not want to translate back and forth
between the doctor and the patient.
This reveals a clash of frames between the doctor and the companion: both have a different
idea of how the roles should be assigned in the conversation (Goffman 1986). The doctor
wanted the companion to take the role of interpreter, transferring the patient’s thoughts and
feelings as an animator. The companion defended his role of author, claiming that he had all
the necessary information about the patient’s problems. To repair the role distribution, in turn
40, the physician embarked on a session of instant interpreter education. He started out by
acknowledging the value of the companion’s knowledge on the patient (turn 40: “Yes, yes,
yes”). Subsequently, he refuted the companion’s argument in a friendly way and explained why
he wanted the companion to pass the question explicitly to the patient (turn 40: “But he can
have other feelings than you think…”). This can be considered as meta-communication, a
strategy to facilitate communication.
On the one hand, the doctor’s approach is friendly and patient, on the other hand, it is assertive
and persistent. This approach seemed successful, as the companion stopped his protest. His
confirming interjection or back channel (turn 41: “Hmhm”) encouraged the doctor to repeat his
question, expecting that it would now be translated (turn 42). This time, the doctor made even
greater effort to facilitate comprehension, by using non-verbal behaviour (finger snapping) to
emphasise the aspect of swiftness and abruptness embedded in the concept “sudden”; but also
by further paraphrasing the concept of a gradual onset (“first a little pain and afterwards more
and afterwards still more”).
The companion translated the question to the patient (turn 43) but made several interpreting
errors in the process. Instead of the concept “suddenly”, the companion used the concept “now
and then”, which was close to the concept “intermittent” which had been discussed in the
previous excerpt and referred to the quality of the pain rather than to its onset. Such an
interpreter error is referred to as a substitution error by Flores et al. (2012). Instead of providing
a literal translation of the doctor’s utterance “first a little pain and afterwards more and
afterwards still more”, the companion translated something like “does it come slowly”, which
could be understood as conferring a similar notion in the sense of a pain that slowly builds up.
Still, it constituted an editorialization error and could add confusion.
The companion translated the patient’s response as “slowly slowly slowly and then fast pain”.
However, the way he pronounced “slowly slowly slowly” was misunderstood by the doctor as
“solo solo solo”, a response which he could not identify as relevant (linguistic failure). The
doctor asked for clarification twice, once through a back channel (turn 48: “huh?”), and a
second time more explicitly, by repeating the companion’s utterance and asking him to explain
it (turn 50: “what does it?”). The companion eventually provided clarification by rephrasing
his initial utterance (turn 51: “little little little”). However, he only repeated and rephrased the
first part of the initial utterance, omitting the part “and then fast pain”, even though this part is
potentially relevant from a clinical perspective. The doctor then confirmed he had understood
(turn 52), and the companion and the patient mumbled and exchanged some further words in
the background. Even though the doctor had understood the semantic content of the patient’s
response (first slowly and then fast), the pragmatic content could still be confusing, as it
confounded the notions of a “gradual” and a “sudden” onset. Nevertheless, given the time
constraints typical of the Emergency Department environment, the doctor decided to move on
to the next topic.
These observations show that after several attempts to receive information on the onset of the
pain, the doctor had received only a partial answer. While the companion insisted that the pain
came up gradually, the patient’s addition of “fast pain” went almost unnoticed. Thus, it is not
clear whether the answer provided by the companion was valid or could be relied on for the
differential diagnosis.
This situation reflects an “illusion of understanding” (Roberts et al. 2005, 468). From what the
doctor could observe, the companion translated the question to the patient, the patient had
produced a response, and the translation of this response by the companion eventually made
sense. However, upon deeper exploration, the companion had made several subsequent
important mistakes (most notably the substitution in turn 42 and the subsequent omission of
the peculiar utterance “and then fast pain” in turn 46) which generated an undetected
misunderstanding of potential clinical consequence.
7.2.3 Remitting or exacerbating factors

The next excerpt again immediately follows the previous one and is still part of the examination
sequence. In this excerpt, the doctor tried to collect information on remitting or exacerbating
factors of the pain, in other words, actions (e.g. urinating) or body positions (such as sitting,
standing) which alleviated or aggravated the patient’s pain. These factors again constitute
important elements to account for in the differential diagnosis.
Excerpt 7.4: Remitting or exacerbating factors
56. 00:02:32.1 DOC If he is in pain*.

…
Is
…
Is it better if he stays calm* or is it better that he moves* all the time.
<1.5>
57. COM ((starts in Urdu))-
58. DOC = ask him
59. 00:02:41.5 COM = if you sit down, is it okay then? ((In Urdu)) …
60. 00:02:44.0 PAT While seating, the pain becomes less. ((In Urdu))
…
61. 00:02:45.3 COM If you do like this, do you feel less pain? ((movement)) ((In Urdu))
62. 00:02:46.5 PAT = Yes ((in Urdu))
63. 00:02:46.0 COM ((Urdu))
(.)
64. 00:02:48.9 PAT Sometimes it stops and sometimes it starts. ((in Urdu))
65. 00:02:52.2 COM = OK
66. 00:02:55.1 PAT = ((incomprehensible)) ((Urdu))
67. 00:02:55.2 COM = He says he sit down <1> no pain// … But he moving* he still have pain
<7>
68. 00:03:06.7 DOC Is… is he in pain when he urinates*?
…
When he makes pipi*.
The doctor shifted to the new topic by asking whether the patient had less pain when he sat
quietly or rather when he moved. The way he put this question (turn 56: “it is better if he stays
calm or is it better that he moves all the time?”) was more explicit than the way he would
normally have addressed a native speaker. Such “overexplicitness” is a typical strategy to
promote understanding in a linguistically diverse context (Dörnyei and Scott 1997).
To ensure that the companion translated his question to the patient, and as such reconfirming
the earlier role assignment, the doctor repeated, “Ask him”. His utterance coincided or
overlapped with the companion who had already started to translate (turn 57). He did so by
splitting the doctor’s binary choice question into two leading questions. He first asked whether
the patient felt less pain when sitting down. Then, after leaving some processing and response
time to the patient, he asked whether the patient felt less pain when moving. The concept of
movement was expressed nonverbally through body movement. The patient answered
affirmatively to both questions (turns 60 and 62) and added in turn 64 that the pain tended to
come and go (“sometimes it stops and sometimes it starts”).
When the companion translated the patient’s message to the doctor, he summarised the
patient’s message in his own way, telling the doctor that when the patient was seated, he felt
no pain, but when he moved, he still felt the pain. As the conversation between the patient and
the companion was not fully comprehensible, it was not clear whether the statement that the
patient still felt pain when moving was accurate. The patient said that he felt less pain when he
moved (turn 62). Moreover, the companion did not translate the statement of the patient that
the pain “sometimes stops and sometimes starts”.
Believing that the patient’s pain was worse when moving around, the doctor then shifted to the
next topic. Notably, he explored a list of other typical symptoms, such as painful urination,
blood in the urine, nausea and vomiting, diarrhoea and eating behaviour, and coughing. During
0the remainder of the examination sequence he used strategies such as self-repetition and
paraphrasing as well. For instance, to make a semi-technical term (urinating) clearer for the
companion, he used the expression “when he makes pipi” (turn 68). The physician went into
more detail regarding the location of pain and the patient’s medication and health history, and
then moved on to the physical examination, which included a blood sample and a scan.
Hence, based on these observations I find that, thanks to the companion’s support as an ad hoc
interpreter, the doctor succeeded in receiving an answer to the question whether the pain was
stronger while the patient was seated or rather while he was moving around. However, it
remains unclear whether the information received was accurate, or whether it implied an
undetected misunderstanding. Moreover, the doctor missed crucial information which related
to the quality of the patient’s pain (continuous versus intermittent), which, if translated by the
companion, could have mitigated the uncertainty created at an earlier stage of the interview.
7.2.4 Treatment negotiation and closing

Based on the verbal and physical examination, the doctor concluded that the patient suffered
from a kidney stone in one of his ureters, which needed to be removed. He explained this to
the companion and asked him to transmit this message to the patient. The doctor asked the
patient’s consent to put him to sleep (under narcosis) and perform the surgery on him.
The thought of surgery made the patient very anxious, and he replied he would prefer not to
have it. When the doctor explained to me that the kidney stone was too big to remove in any
other way, the patient finally agreed, but he indicated he wanted to call his parents first. The
doctor tried to reassure the patient by explaining that it is a minor operation of around 10
minutes, with very few risks, except for some minor risks related to the narcosis. The doctor
and the researcher then left the room to make the necessary calls to find out whether it was still
possible to perform the surgery the same day.
The following excerpt starts upon the doctor’s and the researcher’s return. They had gone back
to the patient and his companion, who were waiting on a bench in the hospital corridor. The
doctor brought the news that as it was 5 pm already, it was not possible for the surgery to take
place the same day and it would instead be deferred to the early morning of the next day. He
also proposed the patient to stay overnight in the hospital, to allow close monitoring of his
situation.
As the patient was anxious about staying over in the hospital, the doctor needed to negotiate
with the patient (through the companion) as to convince him it was the best option. The process
of this negotiation is presented in the following excerpt.
Excerpt 7.5: Treatment negotiation and closing
277. 00:00:19.1 DOC We will do the operation.. tomorrow*.. in the morning*/…

So, he* stays in the hospital/.. So we can do operation tomorrow morning/.. Ok?
278. 00:00:19.6 COM .. Tomorrow morning//
279. 00:00:20.5 DOC = =Yes…
280. 00:00:21.6 COM Euh:: what time?
281. 00:00:23.7 DOC = = I don't know, in the morning//
282. 00:00:24.6 COM = = In the morning//
<1>
283. 00:00:26.0 DOC Everything fine?
284. 00:00:26.6 COM … ((in Urdu)) Euh. Is it okay ?
285. 00:00:28.8 COM ((in Pakistani Punjabi)) So, what should we say to him?
286. 00:00:30.7 COM ((in Urdu)) Qamar has to stay here.
287. 00:00:35.6 PAT ((Patient says something incomprehensible in Urdu))
288. 00:00:38.2 COM He said euuh "tomorrow I come home?"
<1>
289. 00:00:40.6 DOC = = No no no, he stays* in the hospital.*
<2>
290. 00:00:50.1 COM ((in Urdu)) (4) that he (doctor) is saying that, no, today you have to stay here but
there is no risk. No risk.
291. 00:00:53.0 COM No risk? <1> Euh minus, he is afraid//
292. 00:00:57.6 DOC If it was me, ((snaps fingers)) they can do it now// No* problem//
<1>
293. 00:01:00.3 PAT (in Pakistani Punjabi)) Yes, yes, I will tell in 10–15 minutes.
294. 00:01:00.1 COM = = Eh ((in Urdu: incomprehensible))
295. 00:01:01.6 PAT ((in Urdu: incomprehensible))
296. 00:01:04.1 COM okay, he said "after ten* minutes I tell - tell you".
297. 00:01:04.6 DOC = = Pardon?
298. 00:01:10.4 COM …After ten minutes <1> he is. he is. is he (phoning?) Pakistan
((incomprehensible)) mobile
299. DOC = = Yes/ but-
300. 00:01:14.3 COM =After ten* minutes he tell* you//
<2>
301. 00:01:21.9 DOC If he decides to go home/ ... this night* he will come back*. Because it will be too
painful/. Ok?
302. 00:01:22.9 COM …Yes yes yes yes
<1>
303. 00:01:38.5 DOC It's useless to return home... and if he would return home and he looks for another
hospital afterwards, they will redo all the exams and they will also say that he
must stay in the hospital/
304. 00:01:39.0 COM = = Yes//
305. 00:01:40.8 DOC = = So// it's useless to refuse/
…
306. 00:01:40.8 COM aaah
307. 00:01:42.3 DOC ..you can discuss/
308. 00:01:53.6 COM = =((in Urdu)) He (doctor) is saying if you go home you cannot come back here. If
you will go to another hospital, they will examine again and you have to wait for
the reports and it will take more time.
309. 00:01:58.7 PAT ((in Urdu/Pakistani Punjabi)) Give me half an hour and I will call while sitting
here.
310. 00:02:05.3 COM If you. He said: "if you give me half an hour."//
<2>
311. 00:02:06.4 DOC Ok/
…
312. 00:02:07.2 COM He?
313. 00:02:13.4 DOC = = Yes, but I don't understand he is hesitating so much because it's, it's the most*
easy operation that exists.-
314. 00:02:14.4 COM = no no no
315. 00:02:15.4 DOC = = Heh?
316. 00:02:19.5 COM = Sir, listen to me. First* time he stays. He is afraid/
317. 00:02:21.0 DOC = = Ok, ok. No problem.
318. 00:02:22.3 COM = = He is afraid*
<1>
319. DOC Tell him it is the most* easy* operation that exists/
320. 00:02:29.2 DOC Ok? He shouldn't worry about it//.
321. 00:02:34.4 COM …((in Urdu)) is very easy operation, very short operation, no problem at all.
322. 00:02:35.7 PAT ((in Urdu)) here or here? ((pointing at his left and right kidney respectively))
323. 00:02:36.2 COM ((in Urdu)): Here and here
324. 00:02:39.5 DOC = = No* no*. .. First he sleeps/
325. 00:02:39.9 COM = = Uhuh
326. 00:02:44.0 DOC = = Afterwards they put a little wire in the penis/
327. 00:02:44.0 COM = = Uh
328. 00:02:48.4 DOC = = And they go upward to seek for the little stone//.
329. 00:02:48.4 COM = = Aaah
330. 00:02:49.9 DOC Okay?
<1>
331. 00:02:58.1 COM ((in Urdu)) First they will make unconscious then, from the urine way, they will
operate and through the camera they will watch where the problem is and
afterwards they take out.
332. COM Ok I call you
The doctor started the conversation by informing the patient and his companion that the surgery
would take place in the morning of the next day (turn 277). In doing so, he spoke slowly and
distinctly. Moreover, he used more words than he would have used in a native speaker context
(over-explicitness). In the next turn, the companion echoed the doctor’s words (“tomorrow
morning”) as a confirmation check (turn 278). The doctor confirmed. When the companion
asked for more details (turn 280), the doctor replied that he did not know, but that it would be
in the morning. Again, the companion repeated the doctor’s words (turn 282: “in the morning”).
The way in which the companion was repeating the doctor’s words, rather than expressing
agreement, or consulting the patient, suggests hesitation.
In response, the doctor expressed his concern: “everything fine?” (turn 283). Rather than
conveying everything the doctor had said to the patient, the companion only translated this last
utterance (turn 284). It is not clear whether the patient understood himself what the doctor had
said. He may have understood the words “tomorrow morning”. The companion then started
negotiating with the patient what they should tell the doctor about the surgical intervention. In
this negotiation process, the companion and the patient switched forth and back between Urdu
and Pakistani Punjabi. It is not clear whether this “code switching” had an implicature.
The companion summarised the doctor’s initial message as “Qamar has to stay here” (turn
286). This is a substantial reduction of the initial message in turn 277. This lack of information
might have contributed to the patient’s anxiety. In response, the patient asked the companion a
question, which the latter translated to the doctor as “tomorrow I come home?” (turn 288). This
suggests that the patient wanted to know whether he would be able to go home the next day.
However, the doctor seemed to understand the question differently, namely as referring to
whether the patient could go home today instead of staying overnight in the hospital, as he
replied “no no no, he stays in the hospital” (turn 289).
This misunderstanding may have several origins. It could be a language barrier effect, as the
doctor did not have direct access to the utterances in Urdu and as the doctor and the companion
were both speaking in English as a second language. Alternatively, it may have resulted from
a difference in frames between the doctor, on the one hand, and the companion and the patient,
on the other hand. The latter referred to what would happen tomorrow after the surgery, while
the doctor referred to the option of staying overnight. Such a misunderstanding could also occur
in a native speaker context, but the likelihood of its occurrence is higher in a more fragile
communicative context. As such, language barriers can exacerbate instances of
miscommunication that are not directly caused by language barriers.
The companion then translated the doctor’s reply to the patient (turn 290), but he added some
information (“but there is no risk. No risk.”). On the one hand, this can be considered as an
interpreter error (addition). On the other hand, it is possible that he added this information
(which the doctor told him at an earlier point in time) as to reassure the patient and support the
negotiation process by trying to align the doctor’s and the patient’s views. The companion was
balancing between different roles in this excerpt: sometimes he took up the role of carer of the
patient, trying to reassure and convince the patient, sometimes he took up the role of an
interpreter, and sometimes he took up the role of spokesman or advocate of the patient. Each
of these roles required different communication strategies, and might have led the companion
to deviate from what a professional interpreter would be expected to do according to the
textbook model interpreter behaviour. The companion also wanted to be reassured himself. As
a self-accuracy check, he asked the doctor, “no risk?” (turn 291) further motivating his question
by adding that the patient was afraid. To reduce the patient’s anxiety, the doctor said: “if it was
me, they can do it now. No problem.” He added emphasis to his words by snapping his fingers.
The patient reacted to the companion’s statement by requesting 10–15 minutes of respite. Some
additional words were exchanged between the patient and the companion, but these were not
translated to the doctor by the companion (omission). The doctor asked for clarification (turn
297), either because he had not understood the semantic context of the companion’s utterance,
or because he was surprised by what he had heard, and wanted to check whether it was accurate.
The companion tried to explain the doctor in broken English that the patient wanted to do a
phone call before giving his final consent (turn 298). He might have drawn this information
from the words he exchanged with the patient at the time of the interaction (turn 295), or from
an earlier conversation with the patient. When the doctor said “yes”, the companion repeated
his own words to ensure the doctor had correctly understood him (turn 300).
The doctor was still in a different frame, thinking that the patient was hesitating whether to stay
overnight in the hospital. Based on this frame, he tried to convince them that it would be useless
for the patient to return home as it would be too painful (turn 301), implying that if the patient
remained in the hospital, the pain could be mitigated through painkillers. The companion took
on a waiting attitude by uttering back channels (turn 302: “yes yes yes yes”; turn 304: “yes”;
turn 306: “aaah”) instead of talking to the patient. This encouraged the doctor to continue his
plea, adding some arguments in turn 303 and emphasising the word “useless” in turn 305. As
the companion still did not react, the doctor again engaged in explicit role negotiation (turn
307: “you can discuss”).
The companion then translated the doctor’s utterances to the patient, to explain to him why he
should stay in the hospital overnight (turn 308). The rendition was not fully accurate, however:
the companion modified the doctor’s words (substitution) and added his own interpretations
(expanded rendition or editorialization). His substitution changed the content of the doctor’s
message substantially (turn 308: “if you go home you cannot come back here”). The added
interpretation (turn 308: “you have to wait for the reports and it will take more time”) on the
other hand might have been in line with the doctor’s original implicature. The patient insisted
that he wanted some respite to make a phone call. In the rendition to English, the companion
omitted the notion of calling; he only mentioned the patient would like half an hour of respite.
The doctor nevertheless understood from the context of the utterance why the patient wanted
respite, as it got clear previously that the patient wanted to call his parents.
The doctor agreed (turn 311: “Ok”), but at the same time he expressed his disbelief (turn 313:
“I don’t understand he is hesitating so much…”). The companion then acted as the patient’s
advocate, trying to convince the doctor to show understanding for the patient’s anxiety. He first
explicitly requested the doctor’s listenership (turn 316: “Sir, listen to me”) and then explained
that the patient was afraid because it was the first time that he had to stay overnight in a hospital
(turn 316: “First time he stays. He is afraid.”).
The potential conflict arose because the treatment negotiation between the patient, the doctor
and the companion were hampered by a discrepancy in agendas between the patient’s “life
world” and the doctor’s “medical world” (Mishler 1984). As is common in the ED, the doctor
mainly focused on the patient’s medical issues (Scheeres et al. 2008; Slade et al. 2008;
Lachance 2016). On the other hand, the patient was afraid of surgery and wished to speak to
his family in Pakistan before taking a final decision. Rather than leaving the patient alone to
make a phone call, the doctor undertook further efforts to convince the patient he should not
be afraid. To address the patient’s anxiety, he told the companion to explain to the patient that
it was a very easy operation (turns 319-320). The companion passed the message. The patient
asked for more details on the operation (turn 322: “here or here?”).
In response, the doctor provided more details (turns 324-326-328) and then performed a
comprehension check through the interjection “okay?” (turn 330). During the doctor’s speech,
the companion reacted through back channels (turn 325: “Uhuh”, turn 327: “Uh”, turn 329:
“Aaah”), but when the doctor finished, he translated the explanation to the patient. The
rendition was not semantically accurate (e.g. “first they will make unconscious” instead of
“first he sleeps”; “from the urine way, they will operate” instead of “they put a little wire in the
penis”; 25 “through a camera they will watch where the problem is” instead of “and they go
upward to seek for the little stone”) and he also added elements not mentioned by the doctor
(“afterwards they take it out”). Still, the pragmatic content was like the original utterance, and
the main message was passed to the patient. At last, the companion indicated he wanted to
bring an end to the conversation and allow the patient to call his parents (turn 240: “Ok I call
you”). Subsequently, the doctor and the researcher left.
The details added by the companion suggested that the companion was familiar with the kidney
stone removal procedure. As per the Urdu-language interpreter who supported the transcription
of these excerpts, the procedure is standard in Pakistan as well. The fact that the patient seemed
so anxious in view of a minor surgery could relate to the stress of being in an unfamiliar
environment, exacerbated by the language barrier between the patient and the doctor. 26 This
shows how language barriers can reinforce more general challenges to the medical interview,
which do not directly relate to the language barrier itself; and that medical interviews involving
language barriers are more vulnerable to different types of miscommunication than those
involving native speakers only. The patient’s anxiety could as well be reinforced by his
background. For instance, it is possible that due to the poor state of public healthcare in
Pakistan, small interventions such as the removal of a kidney stone sometimes have fatal
consequences and therefore have high stakes and induce anxiety.
25
The fact that the interpreter used the word “urine way” instead of “penis” could relate to the taboo surrounding
the latter word in Urdu, per the Urdu-language expert supporting the transcription of this audio-record.
26
As argued by Meuter et al. (2015), language barriers may result in increased psychological stress for already
anxious patients.
In sum, the doctor’s goal in this excerpt was to convince the patient to stay overnight in the
hospital and undergo a minor surgery in the morning of the next day. This proposal was met
with resistance, mostly due to the patient’s anxiety, but it seems that the doctor’s substantial
communicative efforts have successfully contributed to the treatment negotiation, conditional
on the reassurance the patient hoped to receive from his parents. Later informal communication
with the involved doctor showed that after the phone call with his parents, the patient agreed
to the surgery as well as to staying overnight in the hospital.
7.3 Discussion
This section reviews the different levels of miscommunication which can be identified in the
considered excerpts, drawing on Coupland et al. (1991)’s framework (see Chapter 5). It also
discusses the origins of the identified instances of miscommunication. Finally, it reviews the
different strategies applied by the participants to the interaction in order to prevent,
accommodate and repair instances of miscommunication, if and when they are noticed.

In the case under study, instances of miscommunication were present at different levels. First,
in any conversation there are Level I micro-misunderstandings which relate to the inherent
ambiguity and incompleteness of messages, but remain undetected, even for researchers.
It is more straightforward to identify instances of Level II miscommunication¸ involving minor

misunderstandings which are recognisable, but in general not problematic. Most of the
consultation took place in English as a lingua franca between the companion and the doctor
who spoke English as a foreign language (L2 level). This led to errors against grammar,
vocabulary, and pronunciation but without necessarily changing the meaning of what they
wanted to say. In these cases, correct language use was redundant (Pinker 1995; Seife 2007)
for achieving the communicative purpose of the conversation. At most of the instances where
the companion acted as an interpreter, there were significant errors in the renditions (omission,
substitution, expansion, editorialization). Many of these were not problematic, as they mostly
maintained the implicature of the original message. Some were, however, more problematic,
and triggered miscommunication at higher levels (III and IV).
Level III miscommunication is defined in my framework as miscommunication which hampers

the communication process and can be attributed to a lack of specific skills, such as language
skills or communication skills, or to ill will. One major source of level III miscommunication
was the fact that the doctor could not directly communicate with the patient, which implied that
all information exchanged had to be channelled through the patient’s companion. The latter did
not spontaneously take up the role that would have been most supportive for patient-centred
care (namely as an ad hoc interpreter, focusing on conveying the doctor’s questions to the
patient, and the patient’s thoughts to the doctor). Instead, he often spoke on the patient’s behalf,
without consulting the patient. To encourage the companion to take up a more patient-centred
role, the doctor had to engage explicitly in role negotiation with explicit meta-communication,
which eventually was successful – but only after several repeated attempts which included
using explicit meta-communication to motivate the role assignment. These multiple attempts
prolonged the duration of the consultation considerably.
Furthermore, the companion’s limited linguistics resources did not allow him to convey the
doctor’s questions accurately to the patient. To some extent, he may as well have lacked the
willingness or the patience to do so in a diligent way. When interpreting, the companion often
omitted, added, and substituted information, which led to errors of potential clinical
significance. When confronted with the translated transcripts of the interview during the
member checking session, the doctor in question expressed his surprise over the amount of
information loss that had occurred during the translation by the interpreter. This concurs with
earlier research by Meyer et al. (2010) on family interpreters in a medical setting, which reveals
that many misunderstandings pass unnoticed as ad hoc interpreters may give a relatively fluent
impression, hiding underlying difficulties to cope with the specific repertoire of doctor-patient
communication. In some cases, problems in textual comprehension were mitigated through
non-verbal clues conferring crucial information (Gerwing and Landmark Dalby 2014). For
example, the patient’s inclined position with his hand on his side was of great semantic value
to the doctor, as this position is typical of patients suffering from kidney stones.
The last level of miscommunication considered here is level IV. This level considers the
strategic goals of communication. The doctor’s inability to achieve his communicative goal,
notably to get a solid answer to his questions about the quality (continuous or intermittent) and
the onset (sudden or gradual) of pain, even after repeated efforts, is a salient illustration of level
IV miscommunication. Sometimes, the doctor was even not aware that his goal was not
achieved (false fluency).
There was also an instance where the patient was visibly unable to achieve his goal. When he
asked whether he would be able to leave the hospital the next day (after the surgery), the doctor
misunderstood the question and replied to another question. This misunderstanding resulted
from a misalignment in frames between the doctor and the patient, potentially partly triggered
by the errors against grammar of the companion translation of the question. In a native language
context, it would be easily repaired. However, due to the challenged communication process,
the doctor never noticed this misunderstanding, and it was never repaired. Hence, the patient
did not receive an answer to his question, and this may have contributed to his anxiety. The
companion’s inability and possible resistance to act as the patient’s “sound-box” (Goffman
1981b) did not only prevent the doctor to get across to the patient in the way he would have
preferred; they may also have discouraged the patient from initiating his own questions and
thoughts, or from insisting on them.
From a broader perspective, as achieving “patient-centred care” is also a major communicative

purpose of a medical consultation, the fact that the doctor could not build rapport with the
patient can also be considered an instance of level IV miscommunication. According to Roberts
(2010), language barriers often hamper the process of shared decision-making between the
doctor and the patient, possibly resulting in patient resistance, as it is more difficult to build
rapport with patients (e.g. through humour and/or sharing information).
In the case under consideration, there is very little information exchange between the doctor
and the patient regarding details of the surgery. Part of this may be due to the companion only
taking up his role as an interpreter partially, which meant that the doctor had to spend a lot of
time on role negotiation.

This section follows the framework presented in Section 4.4 to identify the factors and
processes which gave rise to miscommunication at micro-level in the studied excerpts.
Lack of background information or pretextuality

As the doctor did not have any prior knowledge on whom the companion was and how he
related to the patient, he did not know to what extent the companion was aware of the patient’s
pain experience, and how reliable his answers were. It was therefore not clear whether the
companion could serve as a “valuable source of history” on the patient (Bickley 2013, 8). The
doctor also did not have any prior information on the patient’s and the companion’s language
skills, which implied that it was difficult to prepare adequate language support services for the
consultation.
Lack of engagement
The patient did not verbally engage in the conversation. He left the floor to the companion and
limited his involvement to non-verbal communication. The uncertainties generated by the
fragility of the communication context might have added to the patient’s anxiety and stress
experience. The patient’s involvement improved once the doctor urged the companion to
interpret for, rather than speaking on behalf of, the patient. On the other hand, flaws in the
companion’s interpretation led to the patient not always being accurately represented.

There was a language barrier between the doctor and the patient, as a result of which the doctor
needed to rely on the companion to get information on the patient. However, there was also a
language barrier between the doctor and the companion, as English was a second language to
both. At the level of verbal language use, these barriers gave rise to difficulties in the
communication process and prolonged the consultation as extra negotiation of meaning was
needed.
The doctor’s and the companion’s utterances showed at times grammatical errors and little
syntactic glue. This could relate partly to their efforts of using simple language as a strategy
for accommodation. In most instances, correct grammar use was redundant, in the sense that
the doctor and the companion both succeeded in bringing across their main messages. From a
broader perspective, there were several indications in the conversation between the doctor and
the patient’s companion (such as hesitation, confirmation checks) that may point at the
existence of interactional uncertainties (Roberts, Sarangi, and Moss 2004). There were some
instances where an apparent lack of textual comprehension by the interpreter seemed to give
rise to misunderstandings, which often remained unnoticed (see below).
I also found, in line with earlier research by Roberts et al. (2005) that differences in
pronunciation between the companion and the doctor gave rise to miscommunication, for
example when the companion’s unfamiliar pronunciation of the words “slowly slowly slowly”
raised confusion.
Interpreting errors
As the companion was translating the doctor’s utterances into Urdu or Punjabi and the patient’s
utterances into English, information often got lost (reductions, e.g. Excerpt 7.4) or editorialised
(e.g. Excerpt 7.2) in translation. There were also instances of expansion errors (e.g. Excerpt
7.1; Excerpt 7.5) and of substitution errors (e.g. Excerpt 7.3) (Wadensjö 1998; Flores et al.
2012). In some cases, these errors led to a significant deviation in content between the conveyed
rendition and the original utterance. These cases occurred especially when the doctor was
seeking the more technical details on the patient’s pain experience. For instance, in Excerpt
7.3, the interpreter substituted the concept “now and then” for “suddenly” and used “slowly
slowly” instead of the doctor’s longer construction “first a little pain and afterwards more and
afterwards still more”. The former substitution was the most problematic one, as it conveyed a
different message and even removed the logic from the question (notably the notion of a
“sudden” versus a “gradual” onset). Also in Excerpt 7.4 the interpreter seemed to give his own
interpretation of whether the patient felt more pain while seated or while moving, rather than
reflecting the patient’s own thoughts. Another significant deviation in implicature was
observed when the companion told the patient that if he would leave the hospital the same
night, he would not be able to return (Excerpt 7.5).
Important pieces of information went lost because the interpreter did not translate them. For
example, in Excerpt 7.2, the companion did not convey to the doctor the patient’s message that
the pain sometimes made him feel dizzy. Furthermore, the companion did not translate the
patient’s notion that the pain tended to “come and go” to the doctor (Excerpt 7.4). It is possible
that the companion did not do so because he did not consider it as relevant for the specific
question that was asked. However, this shows that significant information can be lost by
working through an ad hoc interpreter.
Often, these interpreting errors remained unnoticed during the interaction. From the doctor’s
perspective, the answers produced by the companion often appeared to be valid and relevant,
creating the false impression of understanding, while leaving space for an unnoticed
misunderstanding of potential clinical consequence. When looking at the transcripts of the
interaction, complemented with the translations of the conversation between the patient and his
companion, the doctor who participated in the interaction was surprised by the significant
amount of information which had been lost in translation.
Frequently, these interpreting errors stemmed from a lack of textual comprehension by the
companion, rather than an intentional attempt by the companion to provide his own view. This
is in line with Li (2013)’s argument that when interpreters are non-native English speakers,
their lack of textual comprehension frequently leads to troubled translations. Sometimes,
however, interpreting errors contributed positively to the communication process, notably
when the interpreter was adding more information (e.g. expansion errors in Excerpt 7.1 and
Excerpt 7.5), which could have facilitated understanding and/or the alignment between the
doctor and the patient.
Pragmatic failure – misalignment of frames

At some point during the interaction (Excerpt 7.5), the companion translated the patient’s
question to the doctor as “tomorrow I come home?”. The doctor understood this as a request
to spend the night at home, while the patient probably meant to ask whether he would be able
to return home the same day of the surgery. This can be considered as a misalignment of frames,
given that the doctor did not understand what the patient meant pragmatically. The frame
misalignment may have been exacerbated to some extent by the grammatical errors in the
interpreter’s utterance. Frame misalignments can as well occur in native language contexts, but
interactions involving language barriers are typically more vulnerable to misunderstandings,
including of those types that are not directly related to the language barrier.
Violation of interactional rules

At times, the communicative flow was disturbed when the doctor and the companion
overlapped and interrupted each other’s speech or did not provide enough processing time for
their counterpart to answer their questions. This occurred particularly at times of misalignment
of goals or of frames between the doctor on the one hand, and the patient and his companion
on the other hand. Also the way in which questions were asked might have had a significant
impact on how (much) information was revealed by the patient (Watt 2008, 3). In Excerpt 7.2,
for instance, the companion’s confusing replies (turns 25 and 27) could have been encouraged
to some extent by the doctor’s way of asking questions. Notably, each of the two parts of the
doctor’s question separately could be considered as a “leading” question, which is known to
prompt respondents to provide “desirable answers” (or false positives) as argued by Bickley
(2013). Also Hull (2010) warns that in the presence of communication barriers, there is an
inherent risk that respondents misunderstand the use of multiple-choice questions and answers,
and therefore two-part questions should be used cautiously.
Misalignment of goals
A salient example of misalignment between the doctor’s “medical world” and the companion
and/or the patient’s “life world” (Mishler 1984) was observed during the treatment negotiation.
The doctor wanted the patient to agree to undergo the surgery, but the patient first wanted to
call to his family in Pakistan before agreeing. The doctor continued to negotiate with the
patient, including by providing more technical details on the surgery, with a view to obtaining
the patient’s agreement. One could also refer to this instance of miscommunication as a
“misalignment of frames” as the patient did not necessarily need to be convinced by the doctor
anymore; he just wanted to call his parents first. As it was not possible to break the patient’s
resistance before he called his parents, the companion interrupted the conversation at some
point, indicating that the patient would now call home. In both cases, meta-communication was
used (by the doctor and the companion respectively) to facilitate frame convergence. In either
case, the misalignment in objectives did not alter the outcome of the consultation, but resulted
in protracted consultation time.

As there was no pre-established role distribution between the doctor, the patient and the
companion, the companion took up a variety of roles during the interaction. From the outset of
the interaction, the patient’s companion spoke on the patient’s behalf instead of acting as an ad
hoc interpreter. It was not clear whether he was adequately reflecting the patient’s thoughts,
acting as an animator (Goffman 1981b), or whether he rather was explaining to the doctor what
he himself thought the patient’s problem was, as an author (Goffman 1981b). The doctor also
did not know how reliable the companion’s replies were (see above). The fact that someone
else speaks for the patient may result in misunderstandings. In a textbook patient-centred
approach, the doctor should be able to address the patient directly, or indirectly through
someone who takes up the role of interpreter by restricting his interventions to the transmission
of the doctor’s questions to the patient, and of the patient’s thoughts to the doctor. However,
the companion seemed to have very different expectations with respect to his own role.
In the initial stage of the consultation, the doctor passively accepted the roles taken up by the
patient and his companion. After some time, however, he became less sure about the replies
given by the companion. To renegotiate the roles in the interaction and encourage the
companion to take up the role of interpreter, the doctor then engaged in explicit role assignment
and instant interpreter education. The companion resisted this at first, as he felt he had sufficient
knowledge of the patient to be a reliable source of history, given that they lived in the same
house. This could also be considered, to some extent, as a misalignment in goals between the
doctor and the companion. Through repeated efforts of the doctor in a friendly, patient, but
assertive way, he achieved frame alignment and the companion acted as an interpreter, albeit
with some interpreting errors.
During the treatment negotiation and closing (Excerpt 7.5), the companion took up the role of
interpreter much more diligently than in the preceding fragments. One reason could be that this
part of the consultation (and in particular the treatment negotiation) inherently needed even
more active input from the patient than the examination phase, as the patient needed to agree
to a treatment proposal.
7.3.3 Prevention, accommodation and repair

This section discusses the different strategies and contextual elements that contributed to
mitigating miscommunication, building on the framework described in Section 4.5.
Generic communication strategies in medical interviews

Specific types of questions can be used to facilitate the patient’s replies in a medical interview.
In some contexts, open-ended questions work best; in other contexts, closed-ended questions
work better. Multiple-choice questions or closed ended questions offer the advantage that they
allow the doctor to get a more precise idea of essential symptoms, facilitate replies and
minimise bias. On the other hand, they might unintentionally trigger false positive answers
and/or create an illusion of understanding and should therefore be used with due attention for
the patient’s comprehension. For example, in the case of multiple choice questions, sufficient
processing time needs to be provided for the hearer to understand the questions. Too much
processing time, on the other hand, can as well be confusing, as the hearer may no longer grasp
the multiple-choice structure anymore.
An example of a successful application of a multiple-choice question was observed in Excerpt

7.2, when the doctor wanted to know whether the patient’s pain was continuous
(“continuously”) or intermittent (“once a lot and then afterwards almost no pain”), and
formulated his question as a multiple-choice (binary) question (Bickley 2013, 60). In between,
he left a 4-second silence for the companion to understand what was being said, and probably
also to encourage him to translate the questions to the patient.
At other instances, however, the use of similar multiple choice questions turned out to be more
confusing. For instance in Excerpt 7.3, the companion did not seem to grasp well the semantic
difference between the concepts “sudden” and “gradual”, and the answer received by the doctor
seemed to imply that the onset of the pain was gradual and sudden at the same time (“slowly
slowly slowly and then fast pain.”) In Excerpt 7.4, possibly because of too much processing
time left in between both questions, the patient does not seem to grasp the fact that the question
is meant as a binary choice question. He answered affirmatively to both parts. The companion
then provides his own interpretation of the patient’s reply to the doctor, indicating that the pain
was worse when moving.

The doctor extensively applied self-accuracy checks to prevent misunderstandings, rather than
accepting every piece of information at face value (Dörnyei and Scott 1997, 188–94). When
he did not understand what the companion said, the doctor used implicit and explicit
clarification requests. For example in Excerpt 7.1, he checked verbally whether the patient was
effectively pointing at the side of his back that hurts. In Excerpt 7.3., he used a back channel
(“huh?”), followed by other repetitions (“solosolosolo?” and an explicit clarification request
(“what does it?”) to encourage the companion to explain the semantic content of his utterance.
Both the doctor and the companion extensively used repetition or “echoing” and rephrasing of
their own words, to ensure that their counterpart had correctly understood what they meant.
For example, in Excerpt 7.2, when the doctor wanted to find out whether the pain was
continuous or intermittent, he repeated his question three times (“Is it continuously or once a
lot and then afterwards almost no pain?”; “Is it continuously, also at night, or is it sometimes
heavy pain and afterwards no pain?”; “My question is, the past four days, was the pain
continuously (…) or was it sometimes big pain after which no pain?”), each time slightly
rephrasing it. The addition of new elements such as a time reference (“also at night” / “the past
four days”) could, however, have generated confusion. Repeating and rephrasing one’s
questions is a common communication strategy to check comprehension and accuracy; or,
alternatively, to reveal potential misunderstanding, as is the case here, given the companion’s
contradictory replies. Uncovering misunderstandings is important, as it gives the doctor an
indication regarding the reliability of the information received.
In Excerpt 7.5, the companion used confirmation checks (“tomorrow morning?”, “no risk”) to
check that he had accurately understood the doctor’s words. The doctor also used self-repetition
to emphasise his own words (“useless”) as to ensure the companion had understood him. Using
confirmation, clarification, and comprehension checks can prevent misunderstandings, but they
can also be a signal of the potential fragility of an interaction, and of the existence of
interactional uncertainties.
There were instances where the doctor very clearly adjusted his language to the companion’s
relatively limited English proficiency level. This strategy is also referred to as
“overexplicitness”, in the sense that a speaker tends to use more words to achieve a
communicative goal in the presence of a language barrier than in a native speaker context. For
example in Excerpt 7.5, the doctor pronounced his words more distinctly and used more words
than he would use in a native setting (e.g. “We will do the operation tomorrow in the morning.
So, he stays in the hospital. So we can do operation tomorrow morning. Ok?”).
A related strategy is paraphrasing, in which (semi-) technical terms are replaced by terms which
are presumably easier to understand for non-native speakers. For example, the specific
formulations as “once a lot and then afterwards almost no pain” (for intermittent) in Excerpt
7.2 and “little by little” (for gradual) in Excerpt 7.3 may be considered as paraphrasing. I
learned from the feedback interview that the doctor intentionally used non-technical easy
English to facilitate the companion’s understanding of his question. Alternatively, it may point
at the doctor’s own word-finding difficulties when speaking English. Another example is
observed in Excerpt 7.4 where the doctor paraphrased a semi-technical term (“urinating”) in
more simple words (“making pipi”). Some caution is due in using this strategy as there is a
slight risk of “overaccommodation”, namely using excessively simplified language that some
patients may experience as slightly downgrading (Street 1991).
The patient’s non-verbal clues were non-redundant in that they were an integral part of the
answers offered to the doctor (Holtgraves et al. 2014). Notably, the patient’s inclined position
with his hand on the left-hand side of his lower back region and his contorted facial expressions
were meaningfully related to the companion’s utterance “he is having pain here” in
combination with a pointing gesture (Gerwing and Allison 2009, 312). This attenuated the
patient's lack of verbal engagement. In a similar way, the pointing gestures by the patient, the
companion as well as by the doctor were co-expressive with their respective utterances
(Gerwing and Allison 2009) and counteracted the potential confusion arising from the
seemingly contradictory verbal references to the patient’s left and right-hand side.
There were other instances too when non-verbal behaviour contributed to the communication
process. For instance, when the companion asked the patient whether he felt less pain while
moving, he simultaneously demonstrated body movement to clarify to the patient what he
meant. This could mean that there was also some language barrier between the companion and
the patient, and that the companion was encountering word-finding difficulties for the word “to
move”.
During the treatment negotiation (Excerpt 7.5), non-verbal communication contributed to

understanding across the language barrier as well. The doctor snapped his fingers to emphasise
the simplicity and instantaneity of the surgical intervention . The patient pointed at his kidneys
to ask which one would be subject to surgery. The doctor understood this non-verbal cue and
responded to it by giving more technical details about the surgery, clarifying that the surgery
would pass through his penis, rather than target any kidney directly.
Meta-communication
One salient instance where meta-communication proved crucial, was when the doctor tried to
reach alignment with the companion with regard to the role assignment in the interaction
(Excerpt 7.3). To this purpose, the doctor had to motivate extensively why it was so important
for the companion to consult the patient before responding to the doctor’s questions. The way
in which the companion explained to the doctor the reason for the patient’s anxiety during the
treatment negotiation (Excerpt 7.5) can be considered as meta-communication as well. The
companion explained why the patient was acting the way he was, in order to convince the
doctor to give the patient some respite before agreeing to the surgery and the hospital stay.
7.4 Conclusion
In this case miscommunication was present at different levels of Coupland et al. (1991)’s
framework: level II, where miscommunication is mostly harmless and less problematic, level
III, where miscommunication arises mostly from gaps in language or communication skills or
from a lack of engagement, and level IV, where miscommunication results from the non-
attainment of strategic communicative goals.
Many of the identified misunderstandings stemmed from the different language barriers present
in the interaction: first and foremost, the one between the doctor and the patient, which was
almost absolute and impeded any direct verbal communication between them; secondly, the
language barrier between the doctor and the companion, which was also strong, but still
allowed the transfer of information between both, albeit through a fragile communication
process. Finally, there might even have been a language barrier between the companion and
the patient, which was however more difficult to trace.
Several factors exacerbated the impact of these language barriers on communication. The
patient exhibited particularly low engagement in the interaction. Lack of background
knowledge on the patient and his companion made it difficult for the doctor to assess the
reliability of the companion as a source on the patient’s history. Problems at the level of role
distribution and flaws in textual comprehension by the companion resulted in weaknesses in
his performance as an ad hoc interpreter. These challenges were compounded by other factors
such as the hectic environment of the Emergency Department, with strict time constraints, as
well as the high stress and anxiety level of the patient. In other words, there were many other
issues that challenged the communication process than just the language barrier. Still, the
language barrier clearly increased the fragility of the communicative environment and made
the communication process more prone to misunderstandings based on these other factors. In
native language contexts, simple misunderstandings are more likely to be repaired swiftly. This
is more difficult in conversations across a language barrier.
A key characteristic of this case study was the absence of the patient’s voice. The main reason
was probably the absolute language barrier between him and the doctor. However, the patient
also took a relatively passive stance during the interaction. He did not attempt to ask the
companion to involve him more closely in the interaction with the doctor. After the doctor
engaged in role negotiation with the interpreter and requested him explicitly to involve the
patient more closely, the companion started to translate and from then on, although not void of
inaccuracies, the patient’s voice was more present in the interaction.
In related work, Roberts et al. (2004) argue that patients with limited communicative
competence in English often do not have the linguistic resources that enable the doctor to apply
textbook models of patient-centred medical interviews. This generates some uncertainty, both
for the doctor (e.g. on how to proceed with the consultation) and for the patient (e.g. with regard
to whether their messages are understood as intended), and underscores the need to train
doctors on how to deal with such “interactional uncertainty” (Roberts, Sarangi, and Moss
2004).
The use of the companion as an ad hoc interpreter led to several instances of

miscommunication. First of all, there was a problem in role distribution where the companion
preferred to speak on behalf of the patient, rather than consulting him on every single question.
This role conflict was particularly salient in this case study. In this context, it is important to
note that the role taken up by the companion was much broader than that of an ad hoc
interpreter. The companion took up different roles throughout the conversation, going from
carer and advocate to interpreter. This implies that the companion at times also had his own
agenda. However, as the doctor did not have sufficient background knowledge on the
companion, he was uncertain about the reliability of the companion as a source of history and
therefore he wanted the companion to act as an interpreter instead.
Even when the companion acted as an interpreter, he made several interpreting errors. Most of
these were due to a lack of textual comprehension by the interpreter. Some may have been due
to his own agenda (e.g. pushing the conversation forward; facilitating the patient’s
comprehension by simplifying questions, …). Some of these errors were recognized by the
doctor and gave rise to interactional uncertainty. Others were however not recognized, and
generated a false illusion of understanding (false fluency). Hence, the accuracy of the
information flow between the patient and the doctor would most likely have been significantly
with a professional interpreter.
On the other hand, without the companion and in the absence of other alternatives for language
support (such as a professional interpreter), the doctor would have been fully restricted to non-
verbal communication with the patient and this would have reduced the quality of care.
Moreover, despite the observed communication problems, the companion promoted comfort
and support for the patient. This aspect is difficult to analyse and to measure, but it is a crucial
element of patient-centred care, which should not be overlooked.
The interpreting errors in this case study did not seem to have significant clinical consequences.
A crucial factor here was the significant presence of specific non-verbal signals, more
specifically the position and facial expressions of the patient, which clearly pointed towards
kidney stone problems. In a different context in which there is no clear non-verbal clue on the
pathology, the communication problems encountered in this chapter could have had more
dramatic clinical consequences. The communicative challenges gave rise to considerable
uncertainty, confusion, and prolongation of the consultation. It also required effort and patience
from all participants to the interaction to try to prevent and repair misunderstandings. Finally,
the patient hardly had an opportunity to talk, which leaves us in the dark on whether his
concerns and questions were adequately addressed during the consultation.
Case study: Tuberculosis and technology 129
8 Case study: Tuberculosis and technology
This chapter presents the analysis of the third case study, in which a patient from Poland visited
the Emergency Department (ED) with a companion. She was feeling tired, had difficulties
breathing, and she was also coughing up blood from time to time. Section 8.1 describes the
broader context in which the interaction took place. Next, Section 8.2 analyses a set of selected
excerpts from the interaction, focusing on those aspects most relevant from a communicative
perspective. Section 8.3 reviews the insights gained from this analysis, summarizing them in
terms of the levels of miscommunication that can be identified; the sources; and the strategies
used to overcome them. Finally, Section 8.4 concludes.

The interaction took place during a Friday night shift with three participants. The physician
(DOC) was a native French speaking male internist in his early thirties. He had started work at
8 a.m. that day and had been working non-stop since then. The patient (PAT) was a Polish
woman between 50 and 60 years old who spoke with a higher pitched voice. She was
accompanied by another Polish woman (COM) of around the same age who spoke with a lower
pitched voice. Both women spoke some French, but their repertoire of French medical
terminology was weak. Over the course of the medical interaction, the doctor also interacted
with a nurse and with a pneumologist (PNE). The latter was a native Dutch-speaking man in
his thirties who came down to the ED to escort the patient to the Department of Infectious
Diseases.
DOC French
PNE L2 French
COM L2 French, Polish
PAT L2 French, Polish
The candidate diagnosis was tuberculosis (TB). Therefore, the doctor focused his review of
systems on the respiratory system and on symptoms typically associated with tuberculosis (see
Table 8.2 for an overview).
At some point, the doctor started using a multilingual consultation software application to
mitigate the language barrier between him and the patient and/or her companion. The
application is called Universal Doctor Speaker Web (UDR) and was operated (in this case)
from the doctor’s tablet. UDR is a relatively novel tool that is not yet widely used in hospital
settings. Prior to the consultation, the researcher had informed the doctor about the existence
of this application and they had discussed the possibility of piloting it sometime in the near
future. When he encountered serious difficulties explaining medical terms to the patient and
her companion, the doctor decided that this was a good opportunity to test the application. As
he was using it for the first time, however, he still needed to explore how it worked.
Table 8.2: Symptoms of tuberculosis

According to Mayo Clinic (2016): According to Bickley (2013, 11):
- Cough - Coughing that lasts three or more weeks
- Sputum (colour, quantity) - Coughing up blood
- Shortness of breath - Chest pain, or pain with breathing or coughing
- Wheezing - Unintentional weight loss
- Pleurisy - Fatigue
- Last chest X-ray - Fever
- Night sweats
- Chills
- Loss of appetite
The application offers a structured menu of pre-formulated medical questions, potential

answers, and statements that were translated into different languages and could be read aloud.
For instance, the respiratory system section features a screen with respiratory system-related
questions and multiple choice answer options in French, and the corresponding translations in
Polish (see Figure 8.1). Upon clicking a question, the application reads it aloud in Polish.
Essentially, the app is a one-directional tool: it provides the doctor with translations of usual
medical questions; but it cannot provide translations of the patient’s answers, as these are
typically less structured or predictable. The use of closed-ended multiple-choice questions can
accommodate this problem to some extent by allowing the patient to convey information
through simple yes/no answers, but not fully.
Figure 8.1: Screenshot of Universal Doctor module on respiratory system
Source: Universal Doctor ©

Note: This screenshot shows an illustration of the English interface of UDR. During the consultation under study,
the doctor was using the French interface.
This case study is of interest as it offers a specific setting with (i) a partial language barrier
between the doctor and the patient; (ii) mediation of verbal communication between the doctor
and the patient by a companion; (iii) a partial language barrier between the doctor and the
patient’s companion; (iv) medical language support provided by a multilingual app. The
language barrier is only partial, as the interaction takes place mostly in French, and the doctor
is a native speaker of French. The patient and her companion both have some notions of French
and encounter some difficulties in expressing themselves. To my knowledge, there is no
empirical study on the use of translation technology in the Emergency Department.
8.2 Analysis
The analysis in this chapter focuses on a set of five excerpts from the transcript of the medical
interaction under consideration. To prepare the transcript for analysis, a Polish healthcare
specialist was consulted for the translation of the utterances in Polish between the patient and
her companion. The selected excerpts are taken from stages 3 (Examination) and 5 (Treatment
and referral) of the medical consultation (see Figure 8.2 and Section 5.2). The full transcript of
the medical interaction is provided in Annex 8.
During the examination stage, the doctor performed a review of the respiratory system in
combination with a differential diagnosis of TB. In the selected excerpts, he mostly tried to
characterise the manifestations associated with the patient’s main complaint (coughing),
notably sputum and breathlessness. I first discuss an excerpt that looks into whether the
patient was producing sputum, based on a verbal interview (Excerpt 8.1) and a physical
examination (Excerpt 8.2). In these two excerpts, the doctor relied on the patient’s own and
her companion’s French-language skills, and on non-verbal communication. When this
approach turned out unsuccessful, the doctor repeated the verbal interview with the
application (Excerpt 8.3).
Excerpt 8.4 provides further illustrations on how the application helped the doctor to receive
responses to very technical questions that are characteristic of the examination sequence, and
in Excerpt 8.5 the doctor used the application to inform the patient on his referral decision.
Source: Based on Heritage and Maynard (2006, 14) (see Section 5.2).
For each of the excerpts, I specify its communicative goal and describe the most relevant
observations from a communicative perspective. I discuss at which level potential instances of
miscommunication occur, with specific attention for the origins of these instances, and the
strategies the participants to the interaction use to prevent or repair such instances. At the end
of each string of conversation excerpts pursuing a particular medical communicative purpose,
I verify whether the communicative goal has been achieved.
8.2.1 Associated manifestations: sputum (without UDR)

In this excerpt, the doctor wanted to find out whether the patient was coughing a lot (quantity),
and whether the cough was dry or produced sputum or phlegm (Bickley 2013, 302). He also
wanted to find out whether the patient was coughing up blood (haemoptysis, see Mayo Clinic
2016b); and, if this was the case, whether the blood was pure or mixed with saliva (NHS 2015).
As the patient had entered the ED after 8 pm, outside triage hours, she had not gone through
triage before seeing the doctor. She had only seen a nurse who had taken her vital signs. Based
on this contact, the nurse suspected that the patient had TB, a condition which was not rare for
patients visiting the considered ED. This hypothesis was the only knowledge the doctor had
before entering the examination booth. He had no prior information on the patient’s social,
linguistic, or medical history; nor did he know who was the woman accompanying the patient,
and whether she was a valid source of information on the patient’s medical history.
Upon the doctor’s entry into the examination booth, he asked the patient what her problem was.
The patient responded in loose phrases that it did not go well: she had problems with walking,
she was very tired, she had difficulty breathing; and she was coughing up blood (“C’est pas
bienne (normally pronounced without the ‘nne’). C'est quand on marcher, c'est beaucoup
fatiguée, c'est pas bon respirer. […] c'est tousser avec sang. Ça c'est tout rouge.”). 27 Although
her utterances showed little syntactic glue, the doctor could extract relevant information from
the patient’s initial problem presentation, as it already revealed two important symptoms: pain
while breathing and spitting blood. These symptoms seemed to confirm the nurse’s hypothesis
that the patient had TB.
27
Translation: “It is not good. It is when walking, a lot of tired, it is not good breathing. […] ”
With this candidate diagnosis in mind, the doctor started the history-taking process. He asked
the patient for the duration of her condition, and she replied it had been lasting for two weeks.
He asked whether she felt mostly while breathing, but she said she felt while not breathing as
well. After the doctor was briefly interrupted by an internal phone call on another patient, the
patient added that she suffered from back pain. The doctor then asked for more details on the
attributes of the patient’s cough. His questions followed the pattern of a typical structured
review of the respiratory system. The selected excerpt describes the conversation that takes
place subsequently.
Excerpt 8.1: Associated manifestations: sputum (without UDR)
33. 00:01:38 DOC et donc, vous toussez aussi beaucoup? = And so you are also coughing a lot? Is
= = C’est ça? that it ?
<1>
34. 00:01:40 PAT non pas beaucoup tousser/ No, not much coughing.
35. 00:01:42 DOC = = d'accord/ Okay
36. 00:01:42 PAT = = et tousser tout sec* And coughing all dry
37. 00:01:44 DOC = = sec Dry
38. 00:01:45 PAT = = Sec* ((confirming tone)) Dry
39. 00:01:46 DOC = = quand vous toussez c'est sec ? When you cough, it is dry?
40. 00:01:46 PAT = = ouais Yes
41. 00:01:48 DOC = = d'accord/… Donc vous crachez* Ok. So you do not spit? There is no
pas? [il] n'y a pas de crachats? Cracher? sputum? To spit? Do you understand “to
[Vous] comprenez “cracher”? spit”?
42. 00:01:52 PAT Cacher*? To hide ?
43. 00:01:53 DOC CR**acher (with emphasis) To spit ?
<1>
44. 00:01:54 PAT …cracher To spit
45. DOC = = comme ça ((demonstrates how to Like this
spit))
46. PAT = =non No
47. 00:01:55 DOC Non, No
48. 00:01:56 PAT = = eh oui un peu, un peu des petits* Euh yes, a little, some small red pieces.
morceaux .. rouges/
49. COM = ((Polish :))((incomprehensible))
50. 00:02:01 PAT = ah c'est aujourd'hui aussi c'est ave[c]- Ah it is today as well it is with
51. COM = avec sang/ with blood
<1>
52. 00:02:05 DOC c'est que du sang?... ou c'est salive/ avec Is it only blood ? or is with blood with
du sang? saliva?
53. 00:02:08 COM = avec… =avec des .. With… with…
54. 00:02:08 PAT = avec des petits morceaux ((shows a With little pieces
napkin in which she has coughed up))
55. 00:02:10 DOC …d'accord/ Okay
56. 00:02:11 PAT = = ça c'est/ That’s…
57. DOC = o-
58. COM = très très rouge Very very red
59. 00:02:13 DOC ok. Est-ce que vous avez perdu du poids Ok. Have you lost weight, madam?
madame?
The patient answered swiftly and effectively to the doctor’s first question regarding whether
she was coughing a lot (turn 33). Her utterance “no, not much coughing” presented weak
syntactic glue in French. Still, the doctor understood her, as can be derived from his
backchannel “okay” (turn 35). They were both aware that they were talking about the patient’s
cough. Because of this “frame alignment”, the use of a proper pronoun (I) and/or of a correct
verb conjugation was not required for mutual understanding: it was redundant (Tannen and
Wallat 1987). Nevertheless, the weak syntactic glue of her utterances gave a first indication on
the weakness of her French language skills and revealed the presence of a significant language
barrier (Roberts et al. 2005). After the doctor’s backchannel, which indicated he wanted to
move on to the next question, the patient expanded on her previous answer by adding that her
cough was dry (turn 36), underscoring her pro-active behaviour in the interaction.
The doctor subsequently echoed the patient’s answer (“dry?”, turn 37). This other-repetition
can be considered as a confirmation or self-accuracy check, to ensure he had understood the
patient correctly. The patient repeated herself once more (“dry”, turn 38) with a confirmatory
intonation. The doctor still seemed uncertain of his correct understanding and asked once more
for confirmation: “When you cough, it is dry?” (turn 39). The patient confirmed this with a
stressed “yes” in turn 40. The reason why the doctor used multiple confirmation checks
suggests he was confused by the patient’s answers, seemingly at odds with her earlier assertion
that she was coughing up blood, and pointing at a potential misunderstanding.
It was important for the doctor to clarify this issue, as sputum (spitting while coughing) is a
key symptom of TB. He therefore continued exploring whether or not the patient was spitting
as she coughed, this time through two subsequent closed-ended yes/no questions: “okay. So
you do not spit? There is no sputum?” (turn 41). His questions were formulated as negative
declarative statements. The negative polarity form of these questions displayed the doctor’s
expectation of a negative answer (Boyd and Heritage 2006, 161–62). The second question was
an example of self-rephrasing, as it had the same communicative intent as the first one, but
used a different wording. Self-rephrasing is a common technique to avoid misunderstandings
in language-discordant communication.
Then, the doctor seemed to become conscious of the possibility that the patient might not
understand the term “sputum” (“crachats”) he was using. He used a comprehension check
involving self-repetition (“To spit? Do you understand “to spit”?”, turn 41) to verify whether
she was familiar with this word (Dörnyei and Scott 1997). The patient’s reply reflected her
non-understanding, as she tried to repeat him by guessing/uttering the word “cacher” which
means “to hide”. Such a guessing confirmation check is common in L2 conversation and is an
indicator of possible miscommunication. The patient probably suspected a misunderstanding
as she considered it surprising that the doctor seemed to ask her whether she was hiding.
The doctor repeated the word “cracher”, with emphasis on the first two letters. The patient then
repeated after him, but with her questioning intonation she expressed non-understanding and
asked implicitly for clarification. The doctor tried to explain her what spitting meant by making
a “spitting” gesture (turn 45). Practically at the same time, the patient said “no”. This caused
some confusion as the doctor then seemed to understand that her “no” was a response to his
question of whether she was spitting while coughing; while it was more likely an attempt to
underline her non-understanding of the word “spitting” before she realised he was explaining
it to her.
The patient then seemed to realize what the doctor meant with his question. In the next turn,
she tried to repair the misunderstanding that had arisen by saying “Euh yes, a little, some small
red pieces” (turn 48). She then exchanged some words in Polish with her companion. Possibly,
she was looking for help to overcome her word-finding difficulties: in turn 50, the patient
started a sentence “ah, today it is also with …”, and her companion finished it off by adding
“blood”. In response, the doctor asked her whether she was coughing up pure blood, or rather
a mix of blood and saliva (turn 51). This is an important distinction as a combination of blood
and saliva is an indication of a chest problem and thus possibly TB, whereas vomiting blood is
more likely to reflect problems originating in the digestive system (NHS 2015). Of both
options, the latter would have been a reason for less concern than the former, as the doctor
explained after the consultation in a follow-up interview.
The companion tried to support the patient, but in doing so she also encountered word-finding
difficulties (“with with…”, turn 52). The patient then decided to resort to non-verbal
communication, and took out a napkin in which she had coughed up and on which blood stains
were visible, while she was repeating her earlier words “with small pieces”. The doctor’s
subsequent “okay” (turn 55) signalled his understanding and satisfaction with her answer, as
well as his willingness to move on to the next topic. Latching on the doctor’s interjection, the
patient said “It is”, and the companion added “very very red” (turn 57). The doctor took note
by means of a backchannel (“oh”, turn 55). He then shifted to the next topic, by asking whether
the patient had experienced any weight loss lately.
My analysis shows that the three interactants managed to achieve some degree of
conversational alignment based on the mutual willingness and commitment to communicate.
This helped the doctor to identify that the patient suffered from a cough, and that her sputum
contained at least some blood. After the interaction, during the member-checking session, the
doctor indicated that he had been able to collect enough information from the patient during
history-taking to consider pulmonary TB as a highly probable diagnosis, and to decide to isolate
the patient.
However, some important unclarities remained. For example, the assertion by the companion
that the sputum was “very, very red” could mean that it was not mixed with saliva. However,
getting clarity on these technical details was severely complicated by the patient’s weak
language skills. Later in the consultation, the doctor would come back to this issue once he had
found a strategy to address these linguistic challenges (see Section 8.2.4). an additional
problem, that was more related to the role dynamics in the interaction, was that it remained
unclear from the conversation whether the companion’s interventions adequately reflected the
patient’s thoughts (with the companion acting as an “animator” of the patient’s ideas) or rather
reflected the companion’s own thoughts (with the latter acting as the “author” of her
utterances). As the doctor did not have any information on the relationship between the patient
and her companion, he could not know whether the companion was reliable as a source of the
patient’s health history.
8.2.2 Physical examination: sputum sample (without UDR)

After having taken the patient’s history through a verbal interview, the doctor moved on to the
physical examination. Given the working diagnosis of TB, this included examining a sample
of the patient’s sputum. In the excerpt considered in this section, the doctor asked the patient’s
permission to carry out further diagnostic tests, and tried to instruct her to spit into a sterile
specimen cup.
In between the previous and the current excerpt, the doctor inquired about other symptoms
commonly associated with TB, such as weight loss and night sweats. Misunderstandings
stemming from the language barrier were overcome through rephrasing and non-verbal
communication. For instance, when the patient did not seem to understand the words “perdu”
(lost) and “poids” (weight), the doctor used the synonymous word “maigrir” (to slim down)
and depicted with his hands the act of slimming down. The patient understood and answered
the doctor’s questions. When asking about night sweats, the doctor added a sweating gesture
to the word “transpirez” (sweat) and checked the patient’s comprehension explicitly before
moving on (“vous comprenez ça?”: do you understand this?). As such, misunderstandings were
prevented and repaired. The patient communicated proactively, for example by adding the fact
that she needed to change her bed sheets frequently. These (unsolicited) additions reassured
the doctor that the patient had understood what he was asking, and that their frames were
aligned. The doctor also enquired about the patient’s possible loss of appetite, smoking habits,
and other relevant health problems. The patient mentioned she had had a kidney operation one
year before.
Once the doctor considered he had sufficient grounds to assume that the patient might have
TB, he announced everyone was to put on a mask. The doctor then left the examination booth
to look for masks. In the meantime, he discussed the case with the nurse who had seen the
patient before. She agreed with him there was a high probability the patient had TB. They put
on masks and re-entered the examination booth, where they requested the patient and her
companion to put on masks as well. They did not explain to the patient and the companion why
this was necessary. It is possible that this lack of meta-communication engendered additional
anxiety or stress among the patient and/or her companion.
The doctor then continued the consultation, asking where the patient lived and which type of
work she was doing (identifying data, see Section 5.2), whether she smoked (as a confirmation
check of what was said before) or used alcohol or any type of medication (health history, see
Section 5.2), and whether she suffered from any other symptoms which could relate to digestive
problems such as diarrhoea or vomiting.
When the doctor asked the patient for her source of referral (“Have you seen a doctor at home?
Was it the doctor who sent you here?”, turn 216), the patient’s reaction was confused (“No […]
the doctor not at home. I don’t […] doctor.”). It was not clear whether she had actually
understood that the word “médecin” meant “doctor”, and not “medicine” or medication. From
the translation of the companion’s utterances to the patient in Polish, it seemed that she had
mistakenly interpreted the doctor’s question as “Are you taking medication?” and had
translated this into Polish to the patient. Before the patient had the opportunity to respond, the
companion again addressed the doctor: “every day, taking something … euh… to breathe”
(turn 220). This illustrates again the patient and the companion’s difficulties to communicate
in semi-technical terms in French.
Similar problems surfaced when the doctor explored whether the patient had ever undergone
an X-ray exam. First, the patient said she had never undergone a scan; and confirmed this again
after the doctor’s first confirmation check. After the second confirmation check, however she
contradicted her first reply, by explaining it was long ago. Some confusion arose as to when
exactly that exam had taken place. The patient then mentioned something about an appointment
four days later with a doctor, but although she tried to give some cue by pointing at her chest,
it remained unclear what kind of doctor she would see or for which type of complaint. In all,
the doctor was hardly able to extract any useful information from this part of the consultation.
He then moved on to the physical examination, and found out the patient’s throat was red. The
following excerpt describes the conversation that took place subsequently.
Excerpt 8.2: Physical examination: sputum sample (without UDR)
264. 00:06:21 DOC ça va… Ok?. ça va… On va aller voir Right… okay? Right… We are going
la radio et on va peut-être faire une to look into your scan and we may take
prise de sang aussi eh// a blood sample as well.
265. 00:06:26 PAT = = oui Yes
266. DOC = = d'accord ? ok?
267. PAT = = d'accord. Ok
268. 00:06:28 DOC = =ça va ? …Donc vous allez cracher Are you okay? So you are going to spit
dans un petit pot* aussi eh//... ça va ? in a small cup too. Ok?
269. 00:06:31 PAT = =oui* ça va.// Yes, okay.
((DOC takes a specimen cup out of the
closet))
270. 00:06:32 DOC je vais vous donner un petit pot/ I’m going to give you a small cup.
((Rustle of paper + chair))
<2>
271. 00:06:36 DOC vous allez cracher* ? ((PAT says You are going to spit?
something incomprehensible in the
background, DOC mumbles
something))
272. 00:06:38 DOC Vous comprenez cracher* ? To spit?
273. 00:06:39 PAT …((Polish:)) Kaszel? Cough?
Non. Eh.
274. 00:06:41 DOC = =eh bien. Well
275. PAT = = pipi? To pee?

276. DOC = = Non/ No
277. COM = =Non/ No
278. 00:06:43 PAT = =kaka? Poo?

279. COM = = non No
280. 00:06:44 DOC = = on peut utiliser Universal Doctor ? Can we use Universal doctor? Euh…
Euh…
281. 00:06:49 COM = ((Polish:)) I think you have to spit.
282. 00:06:50 PAT = Ah ! C'est ... ((PAT makes a spiting Ah, it is…
gesture))
283. 00:06:52 COM = = Oui. Yes
284. 00:06:53 (.)((DOC and RES type out search
terms in Universal doctor))
<16>
285. 00:07:11 RES Cracher ? To spit?
286. 00:07:11 DOC Cracher. Oui. (.) ( mumbling op To spit. Yes.
achtergrond)
<10>
287. 00:07:29 DOC Tuberculose, ((gazes at tablet screen))
288. 00:07:31 DOC = = ah non (.) ( mumbling) Non
<24>
289. 00:07:57 DOC ((incomprehensible)) par système ?
290. 00:07:59 UDR = = ((Polish:)) tuberculosis
291. 00:08:00 PAT = = oh!*
292. 00:08:01 DOC = = non non No no
293. 00:08:03 PAT/COM = = (( anxious laughter ))
294. 00:08:05 DOC/ …Euh ça ne peut pas Euh, this cannot
RES ((incomprehensible)) … l'appareil …
Vas-y, vas-y... Go ahead, go ahead
295. 00:08:14 DOC ((DOC is mumbling))
296. 00:08:40 DOC Ah… oui… oui… ((incomprehensible)) Ah …yes….yes
Non… Ici… ça… Ici… Peut-être No, here, that one, this one maybe
que…
The doctor announced he would have a look at the patient’s scans and take a blood sample.
The patient expressed understanding and/or agreement, first through a simple “Yes” (turn 265),
then by echoing the doctor’s confirmation check (“okay?” “okay.”, turn 266–267). The doctor
continued to check whether the patient was at ease (“are you okay?”, turn 268). Doing so helped
ensure frame alignment between the doctor and the patient and prevent miscommunication, as
it provided the patient with additional opportunities, not only to disagree with the medical
procedure, but also to signal any potential misunderstanding.
The doctor then explained to the patient that she was to spit in a small container. Again, he
asked “are you okay?” (turn 268), verifying simultaneously her comprehension and her
consent. He explained very carefully that she would be given a small cup (turn 270) which she
needed to spit in (turn 271). Again, however, as before during history taking (Section 8.2.1),
the verb “cracher” (to spit) caused confusion. When the doctor asked the patient “are you going
to spit?” (turn 271), she seemed puzzled. To verify whether the patient had understood him,
and as a first repair strategy, the doctor repeated himself: “to spit?” (turn 272).
As a strategy to repair the misunderstanding, the patient guessed aloud what the verb “to spit”
could mean. She guessed three times. First, she guessed “kaszel?” (“to cough” in Polish). The
doctor did not understand her utterance and tried to take the next turn (“Well…”) but was
interrupted by the patient, who guessed a second time: “To pee?”. Both the companion and the
doctor dismissed this reply (“No”; ”No”, turns 275–276). The patient guessed a third time:
“Poo?”, but again, the companion dismissed her conjecture (“No”, turn 279). The doctor started
to feel it would be difficult to explain the patient what he meant, and started considering
alternative strategies to achieve frame alignment. He asked the researcher to use Universal
Doctor (UDR), a multilingual medical software application (turn 280).
While the doctor and the researcher were busy initialising the application on a tablet computer,
the patient’s companion told the patient in Polish “I think you have to spit” (turn 281). The
patient tried to feed back to the doctor she had understood what she needed to do by combining
verbal and non-verbal communication (“Ah, it is …” and a spitting gesture, turn 282). The
companion expressed support for the patient’s statement (“Yes”, turn 283). The doctor,
however, did not hear what the patient had said and did not see the gesture she made. The
reason was that a disruption in the participation framework between the doctor, the companion,
and the patient had taken place, as the doctor had changed footing by addressing the researcher
(see Chapter 3.1.3 and Goffman 1981a). As the doctor was focussed on the tablet screen, he he
did not notice he was being spoken to. This can be considered as miscommunication due a lack
of engagement in the conversation (Gass and Varonis 1991, 127–28). The resulting
misunderstanding implied that the doctor was still not aware that the patient had understood
him, and therefore he continued to look for a solution.
Together with the researcher, the doctor was searching UDR for the Polish translation of the
verb “to spit” (turns 285–286), but failed to find it immediately. As the application is organised
by medical systems, the doctor hoped that he could to find the corresponding test instructions
under the heading of tuberculosis, and therefore he typed out the word “tuberculosis” in the
search window. As the doctor clicked on the word “tuberculosis”, the application read to word
aloud in Polish.
This Polish translation of the word tuberculosis (turn 290) was overheard by the patient and
her companion. The patient reacted startled (“oh!”, turn 291). The utterance might have struck
her as a premature diagnosis announcement. To indicate that this was not the message he
intended to deliver, the doctor swiftly replied “no, no” to the patient. The latter and the
companion reacted by laughing anxiously. Additional meta-communication might have helped
to reassure the patient in this case. Instead, the doctor continued searching the application until
he finally found the module relating to the attributes of the symptom “cough”.
My analysis of this excerpt shows that obtaining the patient’s consent to carry out further
diagnostic tests was hardly problematic in this interaction. However, achieving mutual
understanding with regard to semi-technical terms and/or instructions was severely hampered
by the language barrier. Initially, the patient did not understand what the doctor wanted her to
do. By the time she had understood, he was not paying attention to what she said anymore. As
he was not aware that she had understood him, I consider that his communicative goal was not
achieved. The relational goals of the communication process in this excerpt were not achieved
either. Rather than making the patient feel at ease, the use of masks without explanation and
the application’s unintentional utterance of the Polish word for “tuberculosis” seemed to
contribute to the patient’s stress and anxiety.
8.2.3 Associated manifestations: sputum (with UDR)

In the next excerpt, which immediately followed the previous one, the doctor retook the
patient’s history, this time with the linguistic support of the Universal Doctor (UDR)
application. His main objective was to re-assess the quality of the symptom “cough”, more
specifically whether the cough was dry or produced sputum or phlegm (Bickley 2013, 302);
whether the patient was coughing up blood (Mayo Clinic 2016b), and if so, whether it was pure
blood or blood mixed with saliva (NHS 2015). All these questions had been asked before
(Excerpt 8.1) but considerable uncertainty had been left over the validity of the provided
answers due to linguistic difficulties.
Through the introduction of UDR, the researcher also became partly involved in the
conversation. He was holding the tablet and helped the doctor to find the correct phrases. The
doctor was now standing next to the patient who sat straight on the examination table.
Excerpt 8.3: Associated manifestations: sputum (with UDR)
297. 00:08:45 UDR ((Polish:)) is your cough a dry cough?

<1,5>
298. 00:08:48 PAT Euh... Oui*. Euh… yes
299. 00:08:48 DOC = == oui ?.. Et alors… le suivant là/ Yes ? And so… the next one there
((PAT in the background)
incomprehensible))
300. 00:08:51 PAT = sec*… Tout sec Sec// Dry. It is all dry. Dry.
301. 00:08:53 DOC = = no,c elui là, celui-là*, celui-là*.. No, that one there, that one there, that
ok… one there
(( points at questions) on the tablet’s
screen while talking to RES while
pointing at screen ))
<5>
302. 00:09:00 DOC Ça ne va pas? ((talking to RES while That doesn’t work?
<2>
303. 00:09:03 UDR ((Polish:)) do you cough up phlegm
when you cough?
304. 00:09:07 PAT … euh. Oui./ Euh… yes

305. DOC = = ((Polish:)) Tak ? Yes ?
306. PAT = =Oui/.. Oui/. Yes, yes
307. 00:09:10 DOC = =Oui** ? Yes?
308. PAT = = oui/. Yes?
309. DOC Ah*! Ah!
310. 00:09:11 COM = = et avec sang// And with blood
311. 00:09:13 PAT = = Avec* sang* With blood
312. DOC = = Ouais, c'est ça// . Yes, that’s it
<17> ((talks to RES while looking at

tablet – incomp))
313. 00:09:33 UDR ((Polish:)) What is the phlegm like?
<1>
314. 00:09:37 PAT euh. C'est <1> comment,/ Euh… it is like…
<2>
315. 00:09:43 PAT ((Polish:)) it is solid It is solid
316. COM = ((Polish:)) which colour ? Which colour?
317. DOC = ((talks to RES while looking at tablet
– incomp))
318. 00:09:48 PAT ((Polish:)) it is solid. It is solid
319. 00:09:51 PAT = liquide avec... avec sang. Liquid with… with blood
320. 00:09:53 DOC = = avec sang, ok. With blood, okay.
321. 00:09:55 PAT = = liquid, liquid// Liquid, liquid
322. DOC = = oui, c'est ça. Ok.((talks to RES – Yes, that’s it. Ok.
incomp))
323. PAT Liquid, liquid// <3> Liquid, liquid
324. 00:10:01 UDR ((Polish:)) blood-stained?
325. 00:10:02 PAT = =oui* Yes
326. COM = oui* Yes
<1>
<2>
328. 00:10:07 UDR ((Polish:)) whitish?
<1>
329. 00:10:09 PAT Non*, c'est rose// No, it is pink
330. 00:10:11 DOC = = rosé Pink
331. 00:10:11 PAT = = rosé Pink
332. 00:10:13 DOC = = ok Ok
The excerpt starts when the doctor had located the symptom “cough” in the UDR module (see
Figure 8.3 for a visual illustration). He clicked the first question proposed by the module: “Is
your cough a dry cough?”, triggering the application to speak the same question aloud in Polish
(turn 297). The patient answered hesitantly: “Euh… Yes”. The doctor then changed footing
again as he talked to the researcher, switching back to the backstage frame. The patient tried to
regain his attention, by pro-actively adding unsolicited qualitative information, notably that her
cough was dry (turn 300). She repeated herself twice, emphasizing her own words, as to draw
the doctor’s attention. It worked to some extent, as the doctor expressed his comprehension
through a back channel (“okay”, turn 301). Nevertheless, he still seemed distracted, producing
incomplete utterances (“Because… okayk… okayk...”,”Euheuh” turn 301), indicating that he
was still hovering between the consultation frame and the application operator frame.
The doctor then clicked the next question “Do you cough up phlegm when you cough?”, which
was read aloud in Polish by UDR (turn 303). The patient again answered hesitantly: “Euh…
Yes”. As a confirmation check, the doctor uttered the Polish word, “tak?’ (yes) in a questioning
tone. This illustrates how the doctor used linguistic accommodation to build rapport and make
the patient feel more at ease. In response, the patient expressed confirmation (“yes, yes”) in a
more confident tone. The doctor performed one more confirmation check by echoing her
answer (“yes?”, turn 307). When the patient confirmed her answer, he expressed relief (“ah!”).
Figure 8.3: Screenshot of Universal Doctor module on attributes of symptom “cough”
The patient’s companion then added some unsolicited information (“and with blood”, turn
310). The patient repeated her words as to confirm what she had just said (“with blood”). The
doctor indicated, by means of a backchannel, that he had understood and took note of what
they had said (turn 312). He then clicked the question, “what is the phlegm like?”. The open
question formulation encouraged the patient and the companion to describe different aspects
of the patient’s phlegm (its consistency, its colour and so on). However, while the patient and
the companion seemed to understand the question correctly, and started to describe the phlegm
between themselves in Polish, they struggled to translate these characteristics into French.
Strikingly, while the patient told her companion in Polish that her phlegm was solid (turn 315,
318), she told the doctor in French that it was liquid (turn 319) and even repeated the same
word five times as to emphasize it.
The doctor, switching forth and back from the frontstage to the backstage conversation, then
had the UDR application read the question, “Blood-stained?” aloud in Polish. Both the patient
and the companion confirmed this (“yes”, “yes”, turns 325–326). Possibly by mistake, the
doctor had UDR repeat the same question again; but without leaving space for the patient to
repeat her answer, he launched the next question: “whitish?” (turn 328). The patient answered
dismissively, “no, it is pink.” The doctor echoed her answer as to confirm it: “pink” (turn 330).
The patient repeated again “pink”, eliciting a backchannel from the doctor (“okay”) with which
he indicated that he was ready to move on to the next topic.
In this excerpt, the doctor obtained additional information on the features of the patient’s
phlegm. Technical questions were facilitated by the use of UDR, especially by using closed-
ended questions that were easy to respond. However, important unclarities remained. First, it
was unclear whether the patient’s phlegm was solid or rather liquid, as her statements in Polish
seemed to be at odds with those in French. However, as the doctor did not understand the
utterances in Polish, he did not notice this potential misunderstanding. Second, the assertion
that her phlegm had a pink colour, seemed to confirm that she was coughing up saliva mixed
with blood. 28 However, this reply was somewhat confusing because previously, the patient’s
companion had indicated the phlegm was “very, very red”.
The overarching goal of establishing rapport with the patient was enhanced by UDR. As
compared to the initial stages of the consultation in which UDR was not used, the patient and
the companion now seemed to be more at ease and to speak in a more relaxed way. The
linguistic accommodation seemed to have achieved its goal, notably to strengthen rapport
between the doctor and the patient and her companion.
8.2.4 Associated manifestations: breathlessness (with UDR)

In the next excerpt, the doctor tried to assess the attributes of the symptom “breathlessness”,
which is a prominent symptom of tuberculosis. The patient had already mentioned previously
that when walking, she got very tired and had difficulties breathing (see Section 8.2.1). In a
medical context, the severity of breathlessness is commonly assessed based on the patient’s
self-reported difficulties encountered during daily activities (Bickley 2013, 301), using the
Medical Research Council (MRC) Dyspnoea Scale (see Table 8.3). In addition to assessing the
severity of the patient’s shortness of breath, the doctor also enquired whether it had worsened
over time.
28
Pinkish phlegm can also be a symptom of cardiac decompensation (heart failure), hence this symptom needs to
be interpreted in combination with all other indications (including the sputum particles in the napkin).
Table 8.3: MRC breathlessness scale used to assess the severity of breathlessness
Grade Degree of breathlessness related to activities

1 Not troubled by breathlessness except on strenuous exercise
2 Short of breath when hurrying or walking up a slight hill
3 Walks slower than contemporaries on the level because of breathlessness or has to stop for
breath when walking at own pace
4 Stops for breath after walking about 100m or after a few minutes on the level
5 Too breathless to leave the house, or breathless when dressing or undressing
Source: Stenton (2008)
The following excerpt took place as the doctor was still standing in front of the patient and the
companion. The researcher was standing next to him, holding up the tablet computer with the
UDR application for him.
Excerpt 8.4: Associated manifestations: breathlessness (with UDR)
348. 00:11:10 DOC Eh, c’est bien//. Euh, it’s okay.

<2>
349. 00:11:13 UDR ((Polish:)) Do you get short of breath
when you exert yourself?
350. 00:11:16 PAT ((Polish:)) Tak/ beaucoup// Yes, very
351. DOC = = beaucoup// Very
352. COM = = beaucoup// Very

<3>
354. 00:11:25 PAT = = Euh. Pas chaque fois. C'est… Euh, not always. It’s… also… euh…
aussi… euh /
355. 00:11:27 RES = Si tu veux… je veux dire… eh… If you wish… I want to say…euh…
356. 00:11:28 DOC = = ok… Non, ça je vois.... J'ai pas Ok… No, I can see that… I do not need
besoin. this.
357. 00:11:30 PAT = = Ça, ça ne va pas It does not go well
((incomprehensible)) ((coughs))
<2>
when you climb a few stairs or walk 100
metres?
359. 00:11:39 PAT = Ah oui/ Oh yes
360. 00:11:41 PAT = = 50 mètres// 50 metres
361. 00:11:42 DOC = = 50 mètres? 50 metres ?
362. 00:11:43 PAT = = Oui… Pas 200/((laughter)) Yes. Not 200.
363. 00:11:45 DOC 200* ? ça c'est 100. Il a dit 100 en fait// 200? it is 100. It actually said 100.
364. 00:11:48 PAT = = Ah 100 Ah 100
365. 00:11:49 DOC = = Oui Yes
366. PAT = = 100 mètres… 100 metres

367. 00:11:50 UDR ((Polish:)) Has your shortness of breath
become worse in the last few days??
368. 00:11:52 PAT = = Oui//. Chaque jour plus. Yes, every day more
Between this excerpt and the previous one, the doctor asked some additional questions related
to the patient’s phlegm using UDR. Then, deciding he had sufficient information, he closed the
topic (“Euh, it’s okay.”, turn 348) and instructed UDR to initiate the next topic: “Do you get
short of breath when you exert yourself?” (turn 349). The patient replied “yes” in Polish and
then added “very” in French. This last word was echoed first by the doctor (turn 351) and then
by the companion (turn 352). Then, seemingly by mistake, the doctor clicked the same question
again. This time, the patient gave a different answer: “Euh. Not always. It’s… also… euh…”.
Again, she seemed to encounter word-finding difficulties because of her inadequate French
language skills. In the meantime, however, the doctor had shifted his attention to the backstage
conversation again the doctor seemed to have left again their joint participation frame, being
distracted by UDR and interacting with the researcher (turns 355-356). The patient tried again
to regain his attention and achieve alignment by showing or explaining to him that there was a
specific activity that did not go well (turn 357), but it was not clear what she referred to.
The doctor did not react explicitly to the patient’s struggle to express her thoughts. However,
he used UDR to come up with more specific questions, to help her to give more precise
answers. The next question was “Do you get short of breath when you climb a few stairs or
walk 100 metres?” (turn 358) (see Figure 8.4 for an illustration of the concerned UDR module).
This strategy was indeed more successful, as the patient answered with a convincing “oh yes”.
She further expanded: “50 metres”. The doctor echoed her response as a confirmation check:
“50 metres?” (turn 361). The patient confirmed, expanded again “Yes. Not 200” and laughed.
The doctor got confused by the patient’s reference to a distance of 200 metres, as the UDR’s
question had used a reference distance of 100 metres. He expressed his confusion (turn 363),
perhaps to ascertain that UDR had given the correct translation. The patient acknowledged his
reaction by first echoing him in a questioning tone (turn 364) and then repeating herself in a
confirming tone (turn 366). The doctor then asked whether the shortness of breath had become
worse over time (turn 367). The patient confirmed this, and added that it was getting worse by
the day (turn 368).
Figure 8.4: Screenshot of Universal Doctor module on shortness of breath

In this excerpt, the communicative goal was achieved in that the doctor had obtained the
necessary information regarding the patient’s shortness of breath. At the level of relational
goals, UDR might have reduced rapport by distracting the doctor and reducing his engagement
in the conversation with the patient at least temporarily. On the other hand, UDR may have
strengthened rapport by mitigating the language barrier between the doctor and the patient.
After this part of the consultation, the doctor asked further questions to characterise the
symptom of breathlessness, such as its onset (“did it come on suddenly?”), exacerbating factors
(“does it get worse when exerting yourself?”, “is it worse when lying down than when standing
or sitting?”), and associated manifestations (“do you wheeze?”). 29 He then went on to review
the attributes of the patient’s chest pain, using again a set of closed-ended yes/no questions to
facilitate her replies. He asked about the number of pillows the patient used, whether her legs
were swelling up more than usual, about fever, fatigue, and weight loss. Finally, considering
he had collected sufficient information to have reasonable certainty over the diagnosis of the
patient’s condition, he rounded up (“ok… Super. C’est bien.”, turn 471).
Then, a meta-conversation developed, in which the doctor talked about the communication
process with the patient and her companion. He asked whether they had understood everything,
and the patient answered “yes” and smiled. The doctor expressed satisfaction with the UDR’s
assistance (“c’est bien”, turn 475). The companion attested to this: “super” (turn 476). Then,
they laughed all three about their initial difficulties with the word “to spit”. The atmosphere
had become very relaxed. Several factors contributed to this: the use of linguistic
accommodation strategies, which facilitated the communication process and reduced the
patient’s stress and anxiety, but also the doctor’s deliberate use of meta-communication and of
humour. Finally, the doctor and the researcher left the examination booth.
8.2.5 Referral (with UDR)

When the doctor returned to the examination booth, he was accompanied not only by the
researcher and by a pneumologist (PNE), who had been called in to seek his opinion on the
candidate diagnosis of TB. The ensuing conversation was meant to inform the patient that
additional diagnostic tests needed to be carried out to confirm her diagnosis and that her
informed consent was needed for this. The pneumologist also wanted to find out whether there
had been a TB precedent in the patient’s family.
So far, the doctor had not explicitly mentioned the candidate diagnosis to the patient yet,
possibly to avoid upsetting her. He had told her that something was going on with her lungs,
but that it remained unclear what it really was, and that she had to stay in the hospital to see a
specialist doctor to clarify this. The following excerpts show the conversation that developed
next.
29
Wheezing means making a whistling sound while breathing and is another key symptom of TB.
Excerpt 8.5: Referral (with UDR)
548. 00:00:19 DOC Vous avez compris? Have you understood?

549. 00:00:20 PAT = = Compris// Understood.
<1>
550. 00:00:21 DOC D'accord. .. Je veux dire… euh/ Ok (.) I want to say… euh… (.)
<7>
551. 00:00:30 UDR ((Polish:)) Do you understand what illness
you have?
552. 00:00:33 PAT = = Ah* Ah.
553. 00:00:36 UDR ((Polish:)) We need to carry out different
diagnostic tests
<1>
554. 00:00:40 DOC Demain* Tomorrow
555. PAT = = D'accord. Okayk
556. DOC = = D'accord? Ok?
557. PAT = = Oui*. Yes
558. 00:00:42 DOC = = Demain on fait l’echographie/ Tomorrow echography will be taken.
559. 00:00:43 PAT = = Oui. Yes
560. 00:00:44 DOC <16> ((discussing with pneumologist in
background))
561. 00:00:59 DOC Ça va? ((incomprehensible discussion Ok?
between DOC and PNE)) ((to PNE)) Tu as
des questions à demander avant qu’il y a la Do you have further questions before
nuit ? the night falls?
562. 00:01:10 PNE = = ((to DOC) Oui, donc … sur les Well, yes… About any precedents of
antécédents de tuberculose, des contacts TB, any contact with TB…
avec tuberculose/
563. DOC = = (to PAT:) Dans la famille quelqu'un a eu Is there anyone in your family who
la tuberculose ? has had TB?
564. 00:01:17 PAT = = Euh. Oui. Oui. Ici. Oui// Euh yes, yes, here, yes
565. 00:01:19 DOC = = Elle n'a pas compris/ She has not understood it
566. 00:01:20 PNE Est-ce que, est-ce que vous avez de famille* Have you, have you got family
qui a eu* la tuberculose ? Est-ce que vous members who have had TB? Do you
connaissez des gens qui ont eu la know people how have had TB?
tuberculose ?
<2>
567. 00:01:30 DOC Attends, elle n'a pas compris// Wait, she has not understood.
<4>
568. 00:01:35 RES ((incomprehensible)) si tu tapes famille Maybe if you
peut-être parce que - click “family” because…
569. 00:01:39 UDR = ((Polish:)) Tuberculosis
570. 00:01:41 PAT = = Eh? Eh…
571. 00:01:42 DOC = = Vous comprenez ça? Do you understand this ?
572. PAT = = Oui. Yes
573. 00:01:43 DOC = = La famille? Your family?
<1>
574. 00:01:44 PAT Eh oui avant mon mother mother/// Euh yes, before, my mother, mother
575. 00:01:47 DOC …C'est ça*? That’s it?
576. PAT = = Euh/ Euh…
577. 00:01:49 PNE = Votre mère tuberculose ? Your mother TB?
578. 00:01:51 PAT …Ouais. Yes
579. 00:01:51 PNE = = mère* tuberculose. Yes? TB?
580. 00:01:52 DOC *UXĨOLFD ((Polish: reads from tablet)) TB
581. 00:01:53 PAT *UXĨOLFD* ma* mère// TB my mother
<1>
582. 00:01:55 DOC Ta mère/ Your mother
583. 00:01:56 PNE = = Sa mère/ Her mother
584. PAT = = Maman. Mom
585. PNE = = Elle a eu tuberculose ? Has she had TB?
<1,5>
586. 00:02:01 PAT Euh c’est avant ma mère, c'est moi, c'est Euh it is before my mother, it’s me,
petit bébé… it’s little baby
587. 00:02:06 PNE Elle a eu la tuberculose quand vous étiez un She has had TB while you were a
petit bébé? little baby?
588. 00:02:08 PAT = = Oui* oui//. Yes yes
589. 00:02:09 PNE = = D'accord… OK… Et vous? Vous* Ok. Ok. And you? Have you never
n'avez jamais* eu la tuberculose? had TB?
590. PAT = = jamais* Never
591. 00:02:14 PNE = = D'accord… OK.<3> Vous avez stoppé Ok. Ok. Have you quit smoking?
de fumer?
The doctor asked the patient whether she had understood his instructions regarding her staying
in the hospital and seeing another doctor. She confirmed by echoing his utterance
(“Understood”, turn 550). The doctor wanted to say something more, but ran into word-finding
difficulties (“okayk (.) I want to say… euh… (.)”). He reverted to UDR, which read out the
question, “Do you understand what illness you have?” The patient’s reply was ambiguous
(“Ah”) and did not clarify whether she had an idea about the diagnosis or not. Neither was it
clear whether the doctor actually expected her to answer the question, as he moved on without
seeking further clarification.
The doctor then instructed UDR to read aloud the statement “We need to carry out different
diagnostic tests” in Polish (turn 554), and added to that, in French, “tomorrow”. The patient
expressed her agreement (“okay”). To make sure that it was okay for her, the doctor echoed
her words as a confirmation check (“okay?”). The patient confirmed (“Yes”). The doctor then
told her that she would be undergoing echography the next day, to which she again agreed in
the next turn (“Yes”).
Subsequently, the doctor switched again away from the frontstage towards the backstage and
started to discuss with his colleague, the pneumologist. They discussed about what more could
be done that same evening, and what could be left for the next day. They also discussed which
information should further be collected from the patient (turn 562). The pneumologist proposed
to investigate briefly the patient’s family history regarding TB. In response, the doctor asked
the patient in French whether someone in her family had had TB (turn 564). The patient‘s
response revealed her confusion (“Euh, yes, yes, here, yes”). The doctor notified this to his
colleague, who in response repeated and rephrased the question. Still, the patient did not seem
to understand the question.
Suspecting that the technical term “tuberculose” caused her comprehension difficulties, the
doctor appealed to UDR again to read out the term in Polish: “JUXĨOLFD´turn 570). Indeed,
while in many European languages (including in French), the term for tuberculosis is similar
to the English term, the Polish term sounds very different. This may explain why the patient
had such difficulties understanding the word. The patient showed hesitation in her answer
(“eh…”), as if she did not know what (or how) to answer at first. Without waiting until she had
finished her sentence, the doctor recaptured the turn for a comprehension check (“do you
understand this?”, turn 572).
In the subsequent turns, the doctor and the pneumologist reduced their speech to short and
simple utterances to ensure the patient understood them. The doctor started off: “your family?”.
The patient understood and replied: “eh yes, before, my mother, mother.” While the rest of her
words were in French, she seemed to use the English word for “mother”, albeit with an unusual
pronunciation. The conversation was continuously shifting between different languages:
French, Polish, and now even English, to overcome the language barriers between the doctor
and the patient. This continuous language switching within a conversation is also referred to as
code switching or language switching (Dörnyei and Scott 1997).
To ensure he had correctly understood the patient, the pneumologist performed a confirmation
check by summarising what he had understood: “your mother TB?” (turn 578), which the
patient confirmed. The pneumologist uttered another confirmation check: “Mère, TB?”, after
the doctor repeated the Polish term for TB to make sure the patient understood the question
correctly. Even though his pronunciation was not perfect, the patient recognised what he
wanted to say and repeated it in Polish after him. To underline she had correctly understood
the doctor, she added “my mother” in French (turn 582). Another round of repetitions followed:
the doctor and the pneumologist both repeated after her (“your mother”, “her mother”), after
which the patient herself rephrased her initial utterance (“mum”). Again, the pneumologist
asked for confirmation (“has she had TB?”).
The patient then explained that her mother had tuberculosis when she was a baby. She did so
in a series of ungrammatical utterances with little grammatical glue, but for the doctor and the
pneumologist it was relatively easy to guess her communicative intent, given the frame in
which they were communicating. To ensure that he had understood correctly, the pneumologist
performed a last confirmation check by repeating what he had understood (“she has had TB
while you were a little baby?”, turn 588). The patient confirmed this: “Yes, yes”. The
pneumologist also asked whether the patient herself had ever had TB before. The patient replied
by echoing his word “never”. Then the pneumologist induced a topic shift (“okay. Ok. …”)
and turned to her smoking history. After this excerpt, the pneumologist asked a few additional
questions on the patient’s smoking history and the smoking behaviour of her cohabitants and
then moved on to a second, more detailed, physical examination.
Our analysis of this excerpt shows that, in spite of the linguistic difficulties, the doctor and his
colleague were able to obtain the patient’s consent for further tests. They also found out that
her mother had TB when she was still an infant. While there remained uncertainty as to whether
the patient had understood that she might have TB, this information did not seem to be crucial
in view of the doctor’s medical agenda.
8.3 Discussion
This section reviews the different levels of miscommunication identified in the considered
excerpts, drawing on Coupland et al. (1991)’s framework (see Chapter 5). It also reviews the
sources of miscommunication revealed by these excerpts, and which strategies are used by the
interactants to prevent, accommodate and repair instances of miscommunication when these
are noticed.

As in the previous case studies, miscommunication was revealed at different levels. In addition
to Level I micro-misunderstandings which almost by definition remain unnoticed, there were
several instances of Level II miscommunication. These are usually recognisable, but are in
general not problematic. Most of the consultation took place in French. While the doctor was
a native speaker of French, the patient and her companion spoke French as a foreign language
or at L2 level. The patient frequently made errors against grammar, vocabulary, pronunciation,
or prosody, but without affecting the meaning of what she wanted to say (for instance “no, not
much coughing […] and coughing all dry” in Excerpt 8.1; or “TB my mother” and “it is before
my mother, it’s me, it’s little baby” in Excerpt 8.5); and the doctor still understood her message.
A similar issue arose when, after the introduction of UDR, part of the conversation was carried
out in Polish, and the doctor tried to use some Polish terms as well. He did not always
pronounce the words correctly, but the patient understood what he meant.
On other occasions, however, the patient’s inadequate language skills did lead to
misunderstandings which hampered the exchange of information and the achievement of
mutual understanding. These instances of misunderstanding are considered as level III
miscommunication in Coupland et al. (1991)‘s framework. The patient and her companion
encountered difficulties particularly with the more technical terms (“to spit”, “tuberculosis”)
that were not part of their usual repertoire in French; and they struggled to describe the
characteristics of the patient’s sputum. Most often these instances of miscommunication were
identified and repaired, thanks to the significant pro-active efforts by the interactants to provide
additional information, to cross-check replies, to express and verify mutual understanding, and
to use UDR. The translation of the Polish utterances in the transcript of the interaction reveals
another potential misunderstanding which remained unnoticed at the time of the interaction,
notably regarding the consistency of the patient’s phlegm (“solid” versus “liquid”).
Level IV miscommunication relates to the failure to achieve strategic goal of communication.

In these excerpts, there are some instances where the doctor does not manage to achieve his
communicative objectives because of the patient’s linguistic difficulties. Notably, in some
cases important unclarities remained on issues that were quite important for the differential
diagnosis (for instance, as to the colour of the patient’s phlegm or the reason why she already
had a scan of her chest scheduled) even after several attempts for clarification.
While the use of UDR mostly facilitated the communication exchange and promoted rapport
building, it also led to miscommunication at one instance when the doctor made a mistake and
the application unintentionally read aloud the word “tuberculosis”. Suddenly the application
became an “author”, in a way expressing its own ideas, rather than acting as an “animator” or
mere soundbox of the doctor’s ideas (Goffman 1981a). At the level of the relational goals, the
UDR’s unexpected utterance visibly caused discomfort to the patient and could have disrupted
the process of rapport building. The patient’s panicky reaction may have contributed to the
doctor’s hesitance to inform her later about her possible diagnosis. The minor misalignment of
frames which occurred subsequently (see below) may as well be considered as a instance of
level IV miscommunication.

This section follows the framework presented in Section 4.4 to characterise the major sources
of miscommunication at a micro-level in the excerpts under consideration.

The most salient source of misunderstandings in this interaction was the language barrier
between the doctor and the patient, as a result of the latter’s inadequate French language skills,
particularly when it came to more technical terms. While the companion tried to offer some
linguistic support, her repertoire of French medical terms was not significantly better than the
patient’s. As a result of the linguistic difficulties, the patient’s utterances often presented weak
syntactic glue, and improper use of pronouns, verb conjugations and rhythm (e.g. Excerpt 8.1).
Often, these language problems did not prevent the patient from getting across her message to
the doctor. The “poor grammar” and lack of rhythm did not necessarily change the contents of
the patient’s message, reflecting grammatical redundancy (Seife 2007).
At other instances of the interaction the lack of linguistic comprehension was more
problematic, such as in the case of the patient’s difficulties with understanding the word “to
spit” (see Excerpt 8.1 and Excerpt 8.2) or “tuberculosis” (see Excerpt 8.5) but also “weight”
and “doctor” elsewhere in the interaction; or when it generated uncertainty over the attributes
of the patient’s symptoms, e.g. whether her phlegm was made up entirely of blood or rather of
a mix between blood and saliva (Excerpt 8.1 and Excerpt 8.3) or whether it was solid or liquid
(Excerpt 8.3). These comprehension difficulties led to protracted consultation time, as the
interactants tried to get across to each other by guessing, relying on linguistic support from the
companion or from UDR, or by using repeated rounds of confirmation and comprehension
checks.
In some cases, the doctor could have made a more strategic choice of vocabulary in order to
prevent miscommunication. For example, during the member-checking session, the doctor
acknowledged that instead of asking, “C’est que du sang?” (“Is it only blood?”, turn 52), it
might have been better to use the slightly different formulation: “C’est uniquement du sang?”,
as the word “uniquement” is more explicit and potentially clearer for second language speakers
than the word “que”. The confusion surrounding the relevant distance reference (walking 50,
100 or 200 m) when the doctor was assessing the patient’s difficulties with breathing, illustrates
how talking about precise figures can be problematic for non-native speakers.
Reduced engagement in the conversation

During most of the interaction, the doctor and the patient both showed high engagement in the
conversation, and a strong commitment to get across to each other. However, the downside of
using the UDR application during the interaction was that it generated a “backstage” interactive
frame in which the doctor was frequently focusing on searching the application or talking to
the researcher, rather than paying attention to what the patient said. As a result, he sometimes
missed out on important information, for instance when the patient had finally understood she
needed to spit (Excerpt 8.2).
Not only the doctor switched forth and back from the frontstage conversation between the
doctor and the patient to the backstage conversation with the researcher and the application (all
excerpts except for Excerpt 8.1), or with the pneumologist (Excerpt 8.5). The patient also
switched forth and back between that same frontstage conversation and the backstage
conversation with her companion. These shifts in frames resemble the shifts in interactive
frames observed and described by Goffman (1959) and by Tannen and Wallat (1987) (see
Section 2.2).
At some instances, when the patient encountered word finding difficulties and had problems
expressing herself (e.g. in Excerpt 8.4 when she tries to respond to the question of whether she
has difficulties breathing when she exerts herself; or in Excerpt 8.5 when she reacts confused
to the doctor’s question whether she is aware of the candidate diagnosis), the doctor does not
acknowledge this for being distracted by UDR. This may have discouraged the patient from
expanding further on her answer. On the other hand, the doctor’s behaviour could as well be
interpreted as an attempt to help the patient overcome her word-finding problems by searching
UDR for more specific questions which were easier to answer. However, he did not signal this
to the patient, who might feel a bit lost or left out as a result.

In general, the role distribution in this interaction was fairly balanced. In contrast with the
previous two case studies, the main interaction took place between the doctor and the patient.
The companion assumed the role of offering the patient support, including linguistic support.
Only in some cases did the companion speak on her own initiative. At one instance this
contributed to miscommunication, or at least to diagnostic uncertainty, notably when the
companion added that the patient’s phlegm was “very, very red”. To the doctor it was unclear
whether this was the companion’s own view, or whether it was supported by the patient. It was
at odds with the later statement by the patient that her phlegm was pink. The difference was
nevertheless important as it was an indicator of whether the problem was rooted in the patient’s
lungs or rather in her digestive system.
Another problem in the role dynamics arose when UDR suddenly uttered the word
“tuberculosis” aloud in Polish. This “slip of the tongue” caused an abrupt change in footing
between the doctor, on the one hand, and the patient and her companion, on the other hand.
The fact that everyone had put on respiratory masks just before might have caused the patient
to interpret the utterance as a diagnosis, and she appeared startled as a result. In a way, the
application erroneously took on the role of “author” instead of “animator” of the doctor’s
intended message (Goffman 1981a), and therefore this can be considered a problem of role
dynamics. The patient and her companion were unratified hearers or overhearers.
Misalignment of frames
When the doctor tried to repair the misunderstanding by uttering “no, no” to indicate he did
not mean UDR to read out the word “tuberculosis”, his words seemed to be wrongly interpreted
by the patient and her companion as implying that she did not have TB, given her anxious
laughter in response, which seemed to indicate partial relief. This can be considered as a partial
misalignment in frames between the doctor and the patient, as the doctor wanted to correct a
technical mistake; while the patient expected to be reassured about the severity of her condition.
8.3.3 Prevention, accommodation and repair

This section discusses the different strategies used by the interactants in the considered excerpts
to prevent, repair and accommodate miscommunication, building on the framework described
in Section 4.5.

Both the doctor and the patient extensively used common L2 strategies to promote
understanding (Dörnyei and Scott 1997). They frequently repeated themselves to ensure the
other person had understood their words (self-repetition). The doctor often repeated (and/or
rephrased) the same questions multiple times to make sure the patient had understood
(confirmation check). The doctor and the patient were also frequently echoing each other, and
at times the doctor (or his colleague, the pneumologist) formulated a summary of what they
had understood from the patient’s utterances, as a self-accuracy check. Echoing and
summarising can also be considered as a more generic strategy to enhance communication,

even in conversations between native speakers (Bickley 2013).
The doctor often asked the patient explicitly whether she had understood his words through
explicit comprehension checks (for instance “Do you understand ‘to spit?’, in Excerpt 8.1).
Conversely, the patient frequently expressed her lack of understanding, for example by
repeating the doctor’s utterances in a questioning tone, by pro-actively starting to guess what
they mean, or by simply replying “no” to a comprehension check. Expressing lack of
understanding is important, as it brings misunderstanding to the surface and can be responded
to as if it were a direct appeal for assistance (Dörnyei and Scott 1997). On the other hand, there
were also frequent instances where the doctor and the patient pro-actively confirmed their
understanding, for example by using back channels, which facilitated the flow of the
conversation.
When the patient started to guess what “to spit” meant, she first used the Polish word “kaszel”
(coughs). According to Dörnyei and Scott (1997, 188–94), using a word from the own native
language in non-native speech is a common L2 strategy and can also be referred to as a code
switch or a language switch. Next, the patient used the words “pipi” instead of urine and “kaka”
instead of stool or faeces. This childlike vocabulary would normally not be used in the
participation framework of a medical consultation among adults. However, as her French
repertoire might not extend to the more technical terms for these concepts, she relied on simple
vocabulary which could be easily understood by the doctor. This can be considered too as an
approximation, another common L2 strategy (Dörnyei and Scott 1997, 188–94). Finally, the
patient also used non-verbal communication to complement her verbal strategies (turn 282),
another common L2 strategy.
To facilitate the patient’s understanding, the doctor also applied ‘message reduction’, another
strategy that is common in L2 contexts (Dörnyei and Scott 1997). He mostly communicated in
short sentences that were to the point, sticking to simple vocabulary, and choosing only those
verbs, words and adjectives that were essential to convey a message. At times these short
utterances were not even grammatical, as they were restricted to key words only (for example,
“Your mother TB?” in Excerpt 8.5). These reductions did not have an impact on the meaning
of the utterances, due to the redundancy of many aspects of language (Seife 2007).
Finally, the patient and her companion also frequently expanded pro-actively on their answers,
providing unsolicited details relating to their previous answers. This can also contribute to
achieving mutual alignment in a conversation.
At a few instances in this excerpt, the participants in the interaction made use of non-verbal
communication to get their message across. For example, when the patient expressed non-
understanding of the verb “to spit”, the doctor made a spitting gesture. This worked, as the
patient then provided the requested answer. The patient also used non-verbal communication
to explain to the doctor the characteristics of her cough, when she showed him her napkin with
sputum. Both the spitting gesture and the napkin can be considered as a tool to achieve
alignment of frames between the doctor and the patient (Tannen and Wallat 1987).
Meta-communication
The doctor also used metacommunication as to repair and prevent miscommunication. For
instance, when he took out a cup from the closet in which he wanted the patient to spit, he gave
her detailed instructions on what he expected her to do with it (Excerpt 8.2). Meta-
communication can help the patient feel more secure, at ease, and in control (see Section 4.5.5)
and as such support the communication process. Also in the case of the premature diagnosis
after the Universal Doctor said “tuberculosis” in Polish, the doctor used metacommunication
“no, no” (in the sense of: “no, no, this message was not meant for you”). In this case, however,
he could probably have done even more in terms of meta-communication to mitigate the
patient’s anxiety. The lack of meta-communication may as well have contributed to the
patient’s anxiety and stress when the doctor ordered her to wear a mask without explaining
why; or when the doctor and the pneumologist starting querying her about her family history
of tuberculosis without first explaining why. The strength of meta-communication was
illustrated again at the end of the consultation, when the three interactants started to talk and
laugh about the communication process, the misunderstandings and the use of UDR, and this
brought about relief and a more relaxed atmosphere.
Linguistic accommodation and language switch

Both the patient and the doctor tried to facilitate the communication process when it was
challenged by language barriers through linguistic accommodation. While the patient had
brought her companion, to whom she appealed regularly for linguistic help, the doctor tried to
converge linguistically to her first language or “cross over” (Rampton 2005) by bringing in
UDR, a multilingual medical application that translated his messages into Polish. Initially, the
introduction of the application in the consultation had unintended consequences, but later in
the consultation it turned out to have significant advantages.
UDR allowed one dimension of the language barrier to be overcome. Notably, the doctor
could ensure that the patient understood him. In a way UDR provided the doctor temporarily
and instantaneously with an additional limited repertoire in Polish, in particular there where
communication with the patient was most challenged, notably when questions became
specific and semi-technical (such as the ones needed to assess the severity of breathlessness
per the MRC scale in
Excerpt 8.4). UDR enhanced the flow of communication and contributed to the rapport building
and frame alignment between the doctor, the patient, and the companion.
On the other hand, the application was not able to contribute to the patient’s repertoire, which
meant the communication process remained flawed by language barriers. This caused problems
especially when the doctor used open questions, to which the patient was expected to reply
using relatively technical concepts. However, UDR did provide the option of using a set of
multiple closed ended questions which only required simple yes/no answers by the patient. By
allowing the patient sufficient space in between each question to process it and formulate an
answer, the exchange of information became relatively effective.
When the doctor noticed that the patient was still somewhat uneasy, given her hesitation in
replying to UDR’s questions (“euh… oui…”), he performed a comprehension check using the
Polish word for “yes” (“tak”). This is also a linguistic accommodation strategy which can help
the patient feel more at ease. His attempt seemed successful, as the patient answered back with
a confirming “oui, oui” in French, which encouraged the doctor to continue using UDR. In
turn, the patient also showed flexibility in language use, as she replied in French to questions
asked in Polish, and when she did not find the adequate words in French, she used English (e.g.
“mother” in Excerpt 8.5). This continuous language switching is considered to contribute to
the rapport building as well as to frame alignment.
8.4 Conclusion
In the case under study, the communication process was hampered by the fact that the patient
and the companion were not fluent in French, and the doctor did not speak Polish. Despite this
language barrier, the communication process proceeded reasonably well, especially when
helped by the multilingual medical application UDR. Nevertheless, the quality of the
communication process, and the degree of patient-centeredness of the consultation was still
significantly weaker than in a language concordant medical consultation. It could have been
improved by using professional medical interpreter services, as these would allow the patient
to express her own thoughts more freely to the doctor.
Miscommunication was present at different levels of Coupland et al. (1991)’s framework: level
II, where miscommunication is mostly harmless and less problematic, level III, where
miscommunication arises mostly from gaps in language or communication skills or from a lack
of engagement, and level IV, where miscommunication results from the non-attainment of
strategic communicative goals. While this is not necessarily the case in general, in the medical
consultation studied here, most misunderstandings were either directly or indirectly a result of
the language barrier between the doctor and his patient.
The interactants extensively used strategies commonly applied in L2 contexts, such as

confirmation checks (often by echoing or other repetitions) and comprehension checks, but
also self-repetition, expansion, guessing, approximation and message reduction. As this did not
suffice to overcome the language barriers between them, the doctor and the patient also relied
on meta-communication and on non-verbal strategies to complement their verbal strategies.
The patient often relied on her companion for language support. Nevertheless, the latter also
faced important gaps in her French language skills.
The presence of the companion in the consultation facilitated the communication process. She
participated in the interaction in that she tried to help the patient with understanding and
answering the doctor’s questions. Sometimes she spoke at the same time as the patient or on
behalf of the patient, which can be considered as a problem of role dynamics, but this did not
lead to a reduction of the patient’s voice in the consultation as the latter also actively took the
floor. The companion is most likely to have fulfilled another important function in the ED
consultation, which is providing moral support and comfort to the patient. In contrast with the
patients in the case studies in Chapters 6 and 7, the patient in this case study took on a more
proactive role in the consultation, exemplifying and elaborating on her symptoms, which
helped the doctor progress with his medical agenda.
Despite the strong willingness among all parties present in the consultation to communicate,
as evidenced by displays of attentive hearership (Goodwin 1981), in particular the extensive
use of different strategies to overcome misunderstandings, there were still some instances
where they did not manage to understand each other. In an attempt to accommodate this, the
doctor called in the multilingual medical communication application UDR. UDR helped the
doctor to cross the linguistic barrier by offering a Polish translation of specific and semi-
technical concepts.
Apart from addressing language barrier effects, the UDR application helped the doctor to build
rapport with the patient and his/her companion, by facilitating a “cross-over” (Rampton 2005)
or linguistic convergence (Zuengler 1991) to the patient’s native language in a moment of
distress. After UDR was introduced in the consultation and created a common discursive
ground, both the patient and her companion voices sounded less anxious, and they produced
occasional laughter. This is in line with my earlier findings about the use of a multilingual
informed consent system for data collection in the context of this project (see Chapter 5). When
being hospitalised in a foreign country, and using a foreign language, patients may feel
additional stress. This underscores the importance of accommodation, as hearing a message in
one’s mother tongue or in another familiar language may have a calming effect on patients, as
it appeals to their cultural identity (Gumperz 1982a). This hypothesis was substantiated at the
end of the consultation when both the patient and the companion thanked the doctor with
laughter, while gazing at the application, for his efforts to communicate with them. An
additional advantage of UDR is that it can contribute to the structure of the medical
conversation by offering questions structured by medical themes, and to remind the doctor of
relevant questions he may have forgotten.
On the other hand, however, UDR also has some constraints. First, it is a unidirectional
communication tool. This means that its use is limited to (albeit a wide set) of available pre-
formulated questions that serve the overarching medical purpose of the consultation, or what
Mishler (1984) calls the “medical world.” However, it does not support the patient’s repertoire
or help her to express her own thoughts (and her “lifeworld”) more easily in French, other than
by helping the doctor to ask a very specific closed-ended questions to which the patient only
needs to reply yes or no. If patients cannot express their ideas freely, it is difficult to achieve a
patient-centred medical consultation. To overcome this, it would be more helpful to rely on an
interpreter who is fluent in Polish and in French and familiar with the medical environment and
jargon than to rely on a unidirectional multilingual application.
In this case, the use of UDR added some sources of miscommunication. It reduced the doctor’s
engagement in the front stage conversation with the patient and affected the participation
framework, leading to problems of role dynamics. As the doctor was not yet very familiar with
UDR, he spent a considerable time exploring it, searching for the right modules, and discussing
with the researcher. While doing so, he changed footing, switching from the frontstage to the
backstage, and paid less attention to what the patient was saying and/or doing. Due to his lack
of engagement, he did not notice that she was telling him that she had finally understood the
word “to spit”; and he also violated interactional rules by not always giving the patient
sufficient opportunity to express her thoughts.
This attentional dilemma was previously noted by Swinglehurst and co-authors in their work
on the impact of the electronic patient record in general practitioner consultations
(Swinglehurst 2014; Swinglehurst, Roberts, and Greenhalgh 2011). Like the electronic record
in their study, the UDR application also introduced an interaction contingency in the medical
consultation. As the doctor’s gaze was directed at the screen of the tablet computer, the earlier
established mutual engagement was momentarily disrupted. As it was the first time that the
doctor was using UDR in a live consultation, he needed extra processing time to find the correct
questions and modules in the application. It can be expected that this problem will become
more moderate over time, with additional training and experience of working with the
application.
The application also caused a problem of role dynamics, when the doctor made a mistake and
the application unintentionally read aloud the word “tuberculosis”. At the level of the relational
goals, the UDR’s unexpected utterance visibly caused discomfort to the patient and may have
disrupted the process of rapport building.
A member checking session with the doctor showed that, on the one hand, he considered that
the use of the application had not been decisive for the outcome of the consultation from a
medical perspective, as he had already collected sufficient information during the first few
minutes of the consultation to decide to isolate the patient. Nevertheless, the application helped
him to reduce uncertainty regarding his candidate diagnosis of TB. The application made it
significantly easier for him to ask very specific questions, which he might otherwise have
skipped as he anticipated the patient would not understand them. Moreover, he felt that the
application allowed him to make himself understood in the patient’s language and as such to
strengthen rapport building and enhance patient centeredness.
Conclusion 159
9 Conclusion
This PhD had the aim to describe the dynamics of miscommunication in language discordant
multiparty ED consultations, focusing on the nature of miscommunication, its origins, and the
strategies used to prevent and overcome miscommunication. The three case studies that are
discussed, and their different characteristics, have shed light on the complex multilingual
dynamics that take place in the setting under study. They strengthen each other as evidence for
the claim that differences in the intensity of the language barrier, but also in other aspects of
the communicative context, and their interaction with the communicative purpose of speech
events, lead to different communicative outcomes. In this final chapter, I reflect on the main
findings of the analysis presented in this PhD, and discuss their implications for research and
practice.
9.1 Reflection on findings

In this section, I reflect on the findings of the presented analysis, with a focus on those findings
that seem more generalisable across case studies and that can be of potential broader interest
for research and practice.
9.1.1 The nature of a language barrier and the “communicative swing”

The findings suggest that language barriers have a strong and tangible impact on the quality of
healthcare provision. Depending on its nature, a language barrier may have different
repercussions for the information flow and rapport building. The strength of these
repercussions is not necessarily constant over the course of the interaction. Instead, a language
barrier is found to be a dynamic concept, becoming weaker or stronger depending on other
communicative resources available in the context of the interaction. For example, while correct
language use may be redundant for understanding at some instances in the consultation, this is
not the case at other instances, where the language barrier will have a more detrimental impact
on the information flow.
In the first case of the Moroccan patient with mastitis, there was an almost absolute language
barrier between the doctor and the patient, which did not allow for any form of rapport building.
The patient’s low engagement exacerbated this situation, which could be described as a full
communication breakdown. There was also a considerable but not absolute language barrier
between the companion and the doctor who spoke to each other in Spanish as a lingua franca.
The companion had noticeable difficulties in formulating sentences and did not seem to have
the repertoire as to attend a medical consultation. In addition, the companion did not make
much effort to involve the patient in the communication exchange.
As a result, the information exchange between the doctor and the patient was almost
impossible. Miscommunication was easily detected, but rather difficult to accommodate,
leading to high diagnostic insecurity. The communication process was further complicated by
the interactants’ tiredness, as the consultation took place between 11 pm and 2 pm in the
morning. There was only one instance in which the language barrier between the doctor and
Conclusion 160
the patient dissolved somewhat, notably when the doctor sought the patient’s consent to
physically examine her, using both verbal and non-verbal communication.
In the second case of the Pakistani patient with a kidney stone, there was also a quasi-absolute
language barrier between the patient and the doctor, and a significant language barrier between
the patient’s companion and the doctor, who spoke to each other in English as a lingua franca.
The Pakistani patient relied on his companion for the communication, although in some
instances he seemed to understand at least parts of what the doctor said. Both the doctor and
the companion were speaking in a second language; and the lack of syntactic glue and
differences in pronunciation hampered the communication flow now and then (especially when
it came to more technical elements of the consultation such as the nature of the patient’s pain),
but not to the same degree as for instance in the first case.
The conversation was highly vulnerable to false fluency, however, as in many cases the
companion gave the impression that he had understood what the doctor said, but eventually
delivered a significantly altered message to the patient, and vice versa. This lead to high
diagnostic insecurity; when the doctor was shown the translations of the Urdu/Punjabi
utterances, he was surprised by the amount of information that got lost in translation. A key
element that helped mitigate that insecurity was the patient’s non-verbal behaviour, notably his
body position that was characteristic of patients suffering from kidney stones.
In the third case of the Polish patient with tuberculosis, the doctor was a native French speaker
and the patient and her companion were sufficiently able to speak French to support the rapport
building process and signal understanding. The distribution of the turns was more balanced in
this case study as compared to the others. When she ran into word-finding difficulties, her
companion tried to help her. Rather than speaking on behalf of the patient, in this case the
companion supported the patient in deciphering the doctor’s questions and formulating
answers. As the doctor started to ask more specific questions, their linguistic repertoire proved
to be too limited. Still, the patient actively signalled her (non-)understanding, reducing the risk
of false fluency by helping the doctor to detect miscommunication and encouraging him to
repair it.
The strong willingness to communicate of each of the three participants in the interaction was
reflected by their high level of engagement in the conversation which allowed them to
overcome the language barrier, albeit at the cost of protracted consultation time. In this
relatively favourable context, the multilingual communication tool UDR added value to the
communication process. The tool helped to calm the patient, as she was addressed in her mother
tongue. Member checking with the doctor revealed that the tool helped to cross-validate his
history taking.
In the first and the second case, the language barrier diverted the doctor’s attention away from
broader underlying problems hampering the information flow, notably a misalignment of goals
or frames. In the first case, the patient’s companion wanted the doctor to prescribe some
antibiotics and his impatience to achieve that goal partially reduced his engagement in the
Conclusion 161
consultation, and the quality of this responses to the doctor’s questions. This gave rise to
imbalances in turn-taking, and made the doctor nervous as he was constantly interrupted by the
companion. In the second case, the doctor did not realise that the patient just wanted some time
to call home, and instead continued to insist that the patient’s anxiety for the surgery he would
have to undergo was not necessary. In this case, their willingness to communicate and their
shared repertoire allowed them to align their roles after some discussion.
When during member checking sessions the protagonist doctors were shown the transcripts
and the translations of what had been said in Urdu, Punjabi, Moroccan Arabic, and Berber, they
expressed surprise over the amount of information loss that had passed by unnoticed. This was
less so in the case of the patient with tuberculosis, who showed a stronger degree of engagement
in the conversation, and always expressed clearly and proactively her understanding or non-
understanding of the doctor’s utterances. Accuracy checks and confirmation checks can as well
be useful tools to prevent the risk of false fluency.
The discussion above reveals that a language barrier is not an absolute concept. Rather, a
language barrier can exist in a continuum of degrees or intensities. The intensity of the language
barrier does not only depend on the language proficiency or skills of participants to the
interaction. It also depends on the communicative purpose (e.g. how specific/technical is the
information sought by the doctor?) and the availability of communicative resources or
communicative repertoire (e.g. the scope to use non-verbal communication, see Section 9.1.3).
There are also other elements which have a strong influence on the ultimate intensity of the
language barrier and its impact on the quality of the interaction. These elements comprise
contextual factors (such as stress, anxiety, and time pressure), para verbal (e.g. how words are
said) and non-verbal communicative resources (such as body posture, gestures, and the display
of relevant artefacts), which can mitigate miscommunication resulting from language barriers,
but also reinforce it.
Moreover, as the communicative purpose evolves over the course of the consultation, and the
availability of communicative resources varies, a language barrier is not a static, but rather a
dynamic concept which can vary in intensity over the course of an interaction. As the quality
of the exchange is continuously changing in response to the changing intensity of the language
barrier and the available communicative resources, a “communicative swing” can be observed,
as the exchange of information can be easy at one point and difficult at another point during
the same interaction.
Partial language barriers manifest themselves in word-finding difficulties, problems with

pronunciation and/or with the understanding of utterances. In some cases, mispronounced
words and ungrammatical sentences do not hamper understanding as full linguistic correctness
is to some extent redundant; especially when verbal resources are complemented by non-verbal
ones (e.g. as in the case of the patient with the kidney stone, who held his hand on his lower
back region). In other cases, ungrammatical and mispronounced utterances do hamper
understanding, especially if very specific or technical information is sought (e.g. as in the case
where the doctor tries to find out how old the baby is of the woman with mastitis). While
Conclusion 162
patients may have sufficient linguistic resources to engage in casual conversations, it is more
challenging to talk about more specific terms that are not used so often in common talk (e.g.
‘to spit’, in the case of the woman with tuberculosis).
One particular risk associated with a partial language barrier, is that of false fluency. While in
the case of a full-language barrier misunderstandings are easily detected as the communication
flow is overtly obstructed, in the case of a partial language barrier, misunderstandings are more
likely to remain unnoticed. For example, even if the interactants’ common repertoire is
sufficient to support rapport building, utterances comprising more specific terminology can be
misunderstood. In the case of mediation by a third party (patient companion), information may
get lost in translation. The risk of false fluency may be partially mitigated by strong engagement
of the interactants in the conversation.
When a language barrier is nearly complete, interactants have to rely on non-verbal

communication and the use of artefacts; and communication becomes very troublesome.
Relying on a patient companion for linguistic mediation and interpretation may mitigate this
problem to some extent, but it can as well generate new risks to the communication flow. There
may be a language barrier between the doctor and the companion, and sometimes there is even
an additional language barrier between the companion and the patient. Moreover, companions
may bring their individual agendas to the interaction, which can be an additional source of
miscommunication.
9.1.2 Lack of knowledge on the communicative context

In each of the three case studies, the patients came to see the doctor at the ED without their
medical records; and they were accompanied by a person unknown to the doctor. Important
background information was missing on the linguistic repertoire of the patient and the
companion; the identity of the companion and his/her reliability as a source of history.
From the three case studies and my other ethnographic observations in the ED, I learned that
in most of the interactions, there was no solid assessment of the communicative context at the
outset of the consultation. If such an assessment took place at all, it most often took place in
the course of the consultation. The same applied for the negotiation of the roles taken up by the
patient and the companion during the interaction. As a result, in most cases, the doctor, the
patient and the companion entered into the medical interaction without prior knowledge or
discussion on the “rules of communicative engagement”, hence without the doctor knowing in
what language he would be able to address the patient and/or the companion and to which
extent they would be able to understand him/her (or if they would be able to engage in mutual
verbal exchanges at all).
Doctors often started out the consultation by enumerating the languages they master themselves
(to a greater or lesser degree) to verify whether there was a match with the linguistic repertoire
of the patient or the companion. At other occasions, the companion was often observed to
proactively take the floor as the clinician entered, and to explain what the patient’s problem is,
Conclusion 163
either in the hospital’s language or in English as a lingua franca. In the latter case, clinicians
would start immediately with the consultation, skipping the assessment of the communicative
context. They would then learn “en route” what the patient’s mother tongue was and whether
(s)he understood French or possible other languages. When the companion takes the floor,
clinicians often assume that the companion is a close relative, which is however not always the
case. Recently arrived immigrants may be accompanied by friends or acquaintances they have
known for only a few days or weeks. These problems of identity may spill over into problems
in role dynamics and misaligned expectations.
The lack of investment in the assessment of the communicative context at the outset of the
consultation may point at the need to further highlight the importance of this aspect and the
potentially negative implications (time loss, misunderstandings, …) of its neglect in the
training of healthcare staff. It may as well be a consequence of the doctors’ lack of time and
sense of urgency; or of their feelings of lack of control over possible solutions to tackle
language barriers, besides relying on the patient’s companion, once they are identified.
Doctors rarely called in an on-site interpreter. In most cases, they either relied on the patient’s
companion or sought help from close colleagues for translation. In the hospital under study,
lists with language skills of close colleagues were available. One of the reasons why doctors
prefer to rely on close colleagues may be that calling in an on-site interpreter (who is typically
based at another department) seems more time consuming. However, there are several
disadvantages to this approach: first, close colleagues are typically not trained to intervene as
medical interpreters. Second, performing stints as interpreters implies that staff members are
spending time away from their “core assignments”, which does not always support efficient
human resource management. Finally, during night and weekend shifts, most staff on the list
and the onsite interpreters are not available.
9.1.3 Exploiting the communicative repertoire

With the “rules of communicative engagement” being a priori unknown, the interactants enter
the medical consultation each with their own “communicative repertoire”, which they will,
depending on the level of their engagement, source from to negotiate meaning with the other
interaction. The communicative repertoire is described by Rymes (2014, 290) as “the collection
of ways in which individuals use language and other means of communication (gestures, dress,
posture, accessories) to function effectively in the multiple communities in which they
participate.”
In this sense, the ED consultation can to some extent be considered as a piece of high-stake
improvisational theatre. While the setting of the interaction determines that the patient has a
medical complaint and that it is the doctor’s responsibility to identify the problem and propose
a solution, the way to accomplish this is unplanned and unscripted. Depending on the
Conclusion 164
communicative repertoire of the interactants, their creativity, inspiration and cognitive

efficacy, 30 different communication strategies will be called in.
Our analysis revealed a range of communication strategies that were pervasively used by the
participants in the consultations under study to get across to each other and prevent or repair
miscommunication, such as specific ways to formulate questions, multiple confirmation, and
accuracy checks, self- and other repetitions and rephrasing, message reduction and so on. These
strategies were most useful in a context where language barriers were not too strong. Other
useful strategies were linguistic accommodation, meta-communication, and, very pertinently,
non-verbal communication. Expressing (non-) understanding also proved to be a key strategy
in the negotiation of meaning and the prevention of misunderstanding. The extent to which
participants used this gave some indication of their level of engagement in the conversation.
Depending on the nature of the language barrier and the communicative context, these
strategies varied in success. For example, in the tuberculosis case study, there was a willingness
to communicate in combination with a relatively solid joint repertoire that allowed for the
exchange of information. In this case, the use of multiple confirmation checks and repetitions
largely helped the interactants to detect and overcome miscommunication. In the mastitis case
study, however, with an absolute language barrier between the patient and the doctor and a
severe language barrier between the doctor and the companion, these strategies were not quite
successful. In general, in those cases where there was a considerable gap in shared linguistic
repertoire, these communication strategies did not suffice to overcome the language barrier.
A notable example of how a certain communicative strategy can work in one context, while
adding confusion in another context, concerns the formulation of questions. Question strategies
that are commonly recommended by medical training guidelines for language concordant
consultations may cause problems in language discordant consultations. While doctors
commonly use open questions to get a more comprehensive view on the patient’s problems,
open questions require a much broader linguistic repertoire and are therefore more difficult for
non-native speakers to respond. While closed questions are easier to respond, they considerably
increase the risk of false fluency, in other words, that respondents echo the response or reply
affirmatively without really having understood the question. In the case of multiple choice or
multiple-part questions, non-native speakers might have difficulties distinguishing different
questions from each other. In practice, several question strategies may need to be tried out in
order to be able to cross-check the validity of answers against each other.
Doctors were also observed to use linguistic accommodation as a strategy to strengthen rapport
and reassure their patients. While the official languages of the hospital where the consultation
took place are French and Dutch, the case study consultations took place in Spanish, English
and French respectively. None of the patients (nor their companions) were native speakers of
the languages in which their consultation took place. Only in the tuberculosis case was the
30
This mostly means that if interactants are tired and/or nervous, their cognitive efficacy may be weaker than if
they are well-rested and relaxed.
Conclusion 165
doctor a native speaker of the language (French) in which the interaction took place. This
illustrates the basic willingness of the participants to the concerned interactions to find a shared
language, a “lingua franca” that allowed them to communicate and get across to each other;
one can consider this as a first layer of basic linguistic accommodation.
In several cases, doctors were observed to go beyond this first layer. For example, in the
mastitis case, when the doctor tried to identify the exact location of the patient’s pain, he
addressed her in Moroccan-Arabic, which he believed at that moment to be her native language.
In the tuberculosis case, the doctor used the UDR application to address the patient in her native
language. The UDR application allowed the doctor to engage in “language switching” by
offering him a “pop-up” shared linguistic repertoire. As the UDR application was used, the
patient’s voice changed from sounding anxious to sounding more relaxed. Reducing patient
anxiety can have a positive impact on communication, as stress reduces cognitive efficacy and
may prevent the patient from fully exploiting his/her linguistic repertoire.31
Finally, the analysis of the three case studies revealed that non-verbal communication was a
key tool to overcome difficulties at the verbal level, or to complement oral communication. In
many cases, the use of non-verbal resources proved to be non-redundant as they provided the
first indications for a candidate diagnosis, and as such a valid starting point for the consultation.
In the mastitis case study, despite her low level of engagement in the interaction, the patient
still pointed at her painful breast, giving the doctor an indication of the location of the pain. In
the kidney stone case study, non-verbal communication was used at various instances in time
to support the verbal communication process. The patient’s inclined position with his hand in
his lower back region provided an important and non-redundant cue on where his pain was
located. Non-verbal cues were also used when the companion was asking the patient whether
his condition was worse when remaining silent or rather when moving around; when the doctor
snapped his fingers to illustrate the simplicity and instantaneity of the surgery; and when the
patient pointed at one of his sides to ask which side would be undergoing surgery. In the
tuberculosis case, the doctor used non-verbal communication to explain what the verb ‘to spit,‘
meant, and the patient used an artefact (a napkin with sputum) to provide him with additional
information that was difficult for her to convey verbally.
9.1.4 Redundancy
In addition to the different strategies used intentionally by interactants to negotiate meaning,
our analysis has also provided several illustrations of communication’s inherent mechanisms
to prevent or accommodate misunderstandings, for example because the meaning of a message
may remain unchanged even if specific parts of it are removed. This characteristic of
communication was earlier referred to as the “redundancy of language” (see Section 4.3). In
31
As was mentioned before, there were also some disadvantages associated with the use of UDR: firstly, the UDR
was introduced as an additional “actor” in the interaction, and may occasionally produce undesirable interventions.
Secondly, at several points in the interaction, the doctor was distracted by the UDR application and failed to pay
proper attention to what the patient said. Still, it is conceivable that additional training on working with the
application can reduce the incidence of such undesirable side effects to a minimum.
Conclusion 166
each of the three cases, examples can be found that show how correct language can be
redundant, as listeners often fetch the correct implicature of the speaker’s utterance despite
vocabulary or grammar errors. In the tuberculosis case the doctor is even observed to make
explicit use of this redundancy feature of language as he tries to reduce his messages to only
those key concepts that are non-redundant (for example: “your mother TB?”) to facilitate
understanding. At other instances of the consultation, for example when more specific or
detailed information was sought from the patient, a higher level of precision and accuracy in
language was needed to support the process of differential diagnosis. For instance, in the
mastitis case, the companion’s focus on the key concept “one year” was not sufficient to make
the doctor understand the time frame of the patient’s disease: it was unclear whether his
implicature meant “for one year” or rather “one year ago”.
While for some purposes, full linguistic correctness is redundant; and information can be
successfully exchanged in spite of errors against vocabulary and/or grammar; for other
purposes syntax errors can generate confusion and hamper the correct transmission of
information.
9.1.5 Broader causes of miscommunication

When miscommunication was detected or experienced by the doctors, they would generally
ascribe it to the language gap between them and the other interactants. Because of this
misconception, they would focus their communication on making themselves understood by,
for instance, repeating or reformulating their messages. Our analysis revealed however that
beyond the language barrier in strictu sensu there are other, generally broader drivers of
miscommunication, that go beyond specific turns of an interaction, such as role conflicts, or
misalignments in frames and/or goals between different participants in an interaction.
These broader sources of miscommunication at the same time reinforce and are exacerbated by
the presence of a language barrier, as there are fewer opportunities to negotiate meaning across
a language barrier. Moreover, the absence of a common linguistic repertoire may prevent the
detection of these other sources of miscommunication. At times, the doctor generated a next
level of miscommunication by trying to repair an instance of miscommunication by
erroneously focusing on the language discordance dimension. As such, these broader drivers
may lead to overt confusion in that the interaction gets dominated by fuzzy communication and
discussion or to covert confusion in that the differences in frames or goals do not surface
directly but nevertheless hamper the smoothness and the validity of the information exchange.
Several instances of miscommunication were observed that stemmed from a misalignment in

frames between the doctor and the companion. One instance that caused considerable confusion
was observed in the mastitis case when the doctor tried to get a view on the patient’s pain (did
she suffer from similar pains one year ago, or had she been suffering for the past year from this
same pain?) and on her children (how many children does she have, how old is her latest
baby?). Another instance that generated significant confusion was observed in the kidney stone
case when the doctor mistakenly thought the patient was unwilling to undergo surgery, while
Conclusion 167
the patient just wanted some extra time to call his parents prior to confirming. In the
tuberculosis case, the participants seemed to have much fewer difficulties in achieving frame
alignment, possibly because they could draw on a more extensive common linguistic repertoire;
helped by the patient’s particularly strong engagement in the interaction and her proactive
stance in preventing miscommunication. The case study analysis also suggests that
misalignment of frames is often intertwined with a misalignment of goals or expectations
between different participants to an interaction. The pursuit of a specific agenda could reduce
an interactant’s engagement as an attentive listener, which makes him/her less receptive for the
implicatures of the other’s utterances.
It is not uncommon to observe a misalignment in expectations, goals or agendas between the

doctor and the patient in a medical consultation. As discussed before, the existing literature
describes frequently observed discrepancies between the patient’s “lifeworld” and the doctor’s
“medical agenda”. Among the reviewed case studies, the most salient example of such a
misalignment was observed in the mastitis case, where the companion seemed to have brought
the patient to the doctor just to obtain some specific medication to relieve her pain. As he
already had strong ideas about what he wanted to get out of the consultation, he was sometimes
less engaged in the consultation, and at times seemed to pay little attention to the doctor’s
questions. At the same time, the doctor wanted to make an accurate diagnosis and hence follow
his medical agenda rather than abide directly by the companion’s request.
The instances in the kidney stone case where the doctor tried to convince the companion to act
as an interpreter; or where the doctor tried to convince the patient to stay in the hospital for
surgery the next day can also be considered as instances where a misalignment in goals was
hampering the communication process. However, in both instances these problems were
eventually resolved.
A misalignment of goals (and/or its symptoms) can generate a noticeable burden on the quality
of the interaction; even when the misalignment (the cause of problems) remains unnoticed. For
instance, it can trigger imbalances in turn-taking, as was observed in the mastitis case study.
While the doctor wanted to follow his agenda, the patient’s companion seemed to want to get
as soon as possible at a prescription for medication. As a result, they were interrupting each
other to steer the conversation into their preferred direction. The turn-taking became more
balanced after they succeeded to align their respective agendas. A similar observation can be
made based on the kidney stone case study, where the misalignment in goals (firstly, when the
doctor wanted the companion to act as an interpreter but the companion resisted; secondly,
when the doctor wanted to convince the patient to agree to the surgery, while the patient wanted
to call his parents prior to doing so) disrupted the balance of turn-taking. It is conceivable that
a misalignment of goals triggers frustration and impatience among interactants, increasing the
likelihood that they will speak before their turn.
To reduce the negative impact of such misalignments in goals, doctors could ask patients
(and/or their companions) at the outset of the consultation what their expectations are. If
Conclusion 168
potential differences in individual agendas can be resolved upfront, their influence on the
communicative flow of the consultation is expected to be less disruptive.
9.1.6 Patient companions as ad hoc interpreters

Patient companions took up a variety of roles over the course of the consultation. Contrary to
what might be assumed in institutional settings such as the ED, they seldom spontaneously
took up the role of ad hoc interpreters. This led to a misalignment of the expectations the doctor,
the patient and the companion had over their respective roles in the consultation. The doctor
would generally expect the patient to take up an active role in the interaction, and, where
necessary, to address the doctor through the companion, who would then ideally act as an
interpreter. However, in none of the cases studied, the doctor explicitly communicated this to
the patient and the companion at the outset of the interaction.
As a result, in two of the case studies companions tended to speak on behalf of the patient,
without consulting the latter. In the mastitis case study, the companion took up the role of
interpreter only twice: once spontaneously at the time of the physical examination; once upon
explicit instigation by the doctor at the time of the treatment negotiation. In the kidney stone
case study, the companion did not take up the role of interpreter until the doctor had asked him
several times explicitly to do so, and explained to him why this was important. The patient
companion resisted mostly because he felt himself that he had sufficient information, as he
lived with the patient.
For the doctor, this was problematic, as he needed a first-hand account of the patient’s pain and
symptoms to make an accurate differential diagnosis. It was not clear how accurately the
companion’s words reflected the patient’s ideas. Moreover, it was not clear how reliable the
companion’s knowledge on the patient’s history was. Finally, there was a risk that the
companion had his own agenda.
This encouraged the doctors to engage in explicit role negotiation. In the mastitis case, the
doctor first accepted in a relatively passive way the role taken up spontaneously by the
companion (“Are you going to explain it to me?”). Only in the final stage of the consultation,
he explicitly instructed the companion to consult the patient (“Does she have questions? Can
you ask?”). In the kidney stone case this happened at a much earlier stage, during the
examination sequence (“Ask him”). Still, only after repeated efforts, negotiations, and the use
of meta-communication (which all took considerable time) the doctor managed to convince the
companion to translate his utterances to the patient. Interestingly, the respective patients
seemed to bother less about being excluded, as they did not undertake any major effort to
renegotiate the role distribution by themselves. This rather passive behaviour or lack of
engagement could reflect low self-display and/or be influenced by the specific power dynamics
in the relationship between the patients and their companions.
At the few instances where companions took up the role of language broker or interpreter, at
times errors in interpreting occurred. Companions would editorialize or change the contents of
Conclusion 169
what was said or substitute messages by their own content. There were also instances where
the companion expanded (by adding information) or, conversely, reduced (by omitting part of
the information) the doctor’s utterances. Both in the mastitis and the kidney stone case
instances were observed where the companion’s translation to the patient deviated considerably
from the original instance by the doctor. 32
It was not clear from the conversation whether these editorializations were “strategic” in the
sense that the companion had his own agenda and intended to steer the patient’s behaviour by
deviating from the doctor’s utterances; or whether they resulted from the companion’s own
limited textual comprehension (Li 2013) of what the doctor had really asked. As Gumperz and
Tannen (1979) have argued, these misunderstandings can also be caused by the companion’s
lack of background knowledge, which is required for a correct interpretation of the speaker’s
intention, and hence reflect pragmatic failure. As mentioned before, even though they had
known there was a high risk of miscommunication, the doctors involved in the consultations
were surprised over the amount of information loss that occurred through the translation by the
patient’s companions but remained unnoticed.
In the third case study, the patient companion acted mostly as a provider of comfort and
support, including linguistic support, for instance when the patient experienced word-finding
difficulties. She did not explicitly act as an ad hoc interpreter, as the doctor mostly interacted
directly with the patient. In this last case, the power balance between the patient and the
companion was also very different. This could be related to the nature of the personal
relationship between the patient and the companion (which seemed less hierarchical in this case
study, in contrast with the other two), but it could have been as well be reinforced by the fact
that the patient could and did directly reach out to the doctor.
The case analyses illustrate that it is not straightforward to expect patient companions to take
up the role of ad hoc interpreters, even if they have a better linguistic repertoire than the patient.
Doctors may need to engage explicitly in negotiations from the outset of the consultation to
establish adequate roles. As will be discussed below, even in those cases where companions
take up the role of ad hoc interpreters, there is a high incidence of interpreting errors, which
cannot always be detected by the doctor during the consultation. Still, the role played by patient
companions reaches beyond serving as ad hoc interpreters: they can have beneficial effects by
providing support of different forms, comfort, and can help reduce patient anxiety and stress
by contributing to a more familiar environment.
The contributions by the patient’s companion to the interactional dynamics in the consultation
is to some extent similar to Goffman’s (1967) metaphor of the “Telephone Booth Bias”, which
referred to the fact that what patients said or did not say during psychiatric consultations was
influenced by others (see also Section 2.2). In a similar way, next to the positive contributions
32
In the former case, the companion told the patient that the doctor had asked her whether she wanted him to give
her some pills; while the doctor had actually asked whether the patient had any questions remaining; in the latter
case, the companion told the patient that if he left the hospital, he would not be allowed to return; while the doctor
had actually said that returning home or going to another hospital would only make him lose time.
Conclusion 170
they can make, patient companions may as well “add noise” to the interactions between doctors
and patients.
9.1.7 Interactional conflicts of interest – opening Pandora’s box?

Communication problems may loom large in an interaction, but they may remain unnoticed. In
many cases, it is only when interactants make active efforts to check understanding, e.g.
through confirmation or accuracy checks, that miscommunication is revealed. Without such
efforts, communication may seem fluid even when it is problematic. When understanding is
actively checked by the interactants, and miscommunication is revealed, it becomes more
difficult to ignore it, and interactants are called upon to find solutions to overcome it. This
brings additional responsibilities and challenges.
For instance, a doctor may suspect that a patient (companion) is not answering his question
correctly, without knowing the underlying reason. Actively entering into a process of tackling
the misunderstanding may require substantial efforts and involve considerable time. In some
cases, the shared linguistic repertoire may suffice to communicate there is a misunderstanding,
but not necessarily to meta-communicate about underlying goals and expectations. In such
cases, interactants face a difficult choice between continuing to negotiate for meaning, or
instead make peace with the situation and try to achieve “good enough” communication.
While the process of actively checking understanding, and displaying understanding and non-
understanding may help to surface and mediate miscommunication at different levels, it is also
likely to strengthen the perception of diagnostic insecurity and makes the consultation more
cumbersome, long-winded and tiring. This generates a conflict of interest between the heavy
time constraints that are typical of the ED and the decision to engage in time-consuming
interactional strategies, especially when the likelihood of success and the ensuing gains in
terms of quality of care are not certain a priori.
9.1.8 Actual understanding and perceived understanding

As a result of communicative difficulties, the communicative purpose of an interaction may or
may not be achieved. Even if it is achieved, the process of exchanging information is typically
imperfect and fragile. Even if crucial information is transmitted, there can still remain a high
degree of uncertainty surrounding its validity.
Conversation analysts often apply the next-turn-proof-procedure to reveal misunderstandings.

This principle considers whether speakers display in the “next turn” an understanding of what
was said in the previous term (Jenks 2014, 279; Hutchby and Wooffitt 2008, 14). In the cases
under study, it was sometimes observed that while the application of this principle would imply
that the interactants indeed displayed understanding, they themselves expressed a lack of
understanding in follow-up interviews. This points at the existence of gaps between actual and
perceived understanding, possibly as a result of interactional uncertainties and difficult
contextual settings that add to the doctor’s insecurity. This discrepancy between actual and
Conclusion 171
perceived understanding can have relevant implications for care provision as low perceived
understanding can cause diagnostic insecurity on behalf of the doctor.
For example, in the case of the patient with mastitis, when the doctor asked about the onset of
the patient’s pain, the companion replied and the doctor displayed his understanding of the
companion’s subsequent answer. The doctor however did not get an answer to his question
regarding the timing of the pain and he explained to me in the after-action interview that he felt
frustration and doubt with regard to the quality of the information exchange. The doctor
perceived both his understanding of what the companion said and the companion’s
understanding of what he had said very low; even if the transcript of their conversation
indicated there was a relevant process of meaning making going on, as attested by the next-
turn proof procedure. Moreover, an external doctor that was consulted on the analysis of the
transcripts considered the interaction as “good enough”. She recognised that the
communication was far from perfect but indicated that the doctor still managed to get the
necessary information on the patient’s condition.
The opposite can also happen, notably in the case of false fluency (see Section 9.1.1). For
instance, in the case of the patient with the kidney stone, there seemed to be actual
understanding between the interactants. But since there was a severe language barrier between
the doctor and the patient, neither of the two realized that what was translated by the companion
was not always in line with the initial utterance of the doctor or the patient respectively.
In the third case of the patient with tuberculosis, there seemed to be a good correlation between
actual understanding and the perceived understanding. The next turn proof procedure was
clearly visible in that all parties actively displayed understanding and non-understanding. In
this case, the interactants’ willingness to communicate and their available joint repertoire in
French allowed them to negotiate meaning. But here also, although the doctor was generally
satisfied with the quality of the communication, he was not always sure whether the patient
and her companion would understand them.
In sum, in many forms of human communication people make meaning and the next turn proof
procedure is tangible in the interaction. But contingencies such as different language barriers,
hesitations and comprehension problems sometimes go beyond what can be captured by the
next turn proof procedure. The use of contextual data from ethnographic fieldnotes and after-
action interviews can reveal some of these features.
9.2 Implications for research

Our research has brought together insights from different strands of the literature. Coupland et
al. (1991)’s framework for analysing the different levels of miscommunication provided a
relevant basis for my research, as it helped to unravel and depict the complexity of
miscommunication of language discordant multiparty ED consultations. However, as their
framework was a “preliminary model” (N. Coupland, Giles, and Wiemann 1991, 16), I had to
Conclusion 172
adapt it to the specific context of my research, by enriching it with insights from interpreting
studies, interactional sociolinguistics, second language learning, and medical communication,
through an iterative and heuristic process. This is similar to the approach taken by Stubbe
(2010) in her study on miscommunication and problematic talk in the workplace, who
considers that it is impossible for Coupland et al. (1991)’s framework to cover all aspects of
miscommunication, but that it nevertheless does so in the best possible way.
In the area of language discordant medical interactions, most existing research has looked at
the context of primary care. Relevant insights have been developed in this area in particular by
Roberts and her co-authors (e.g. Roberts, Sarangi, and Moss 2004; Roberts et al. 2005; Roberts
2010; Roberts and Sarangi 2005) who analysed linguistically and culturally diverse dyadic
consultations in primary care. Several of my findings align with their conclusions. My analysis
also found that miscommunication manifested itself through lexical problems, weaknesses in
syntactic glue, word stress and pronunciation and differences in the way patients presented
themselves; and that non-verbal resources were used to mitigate miscommunication to some
extent. It added however another layer of complexity by considering medical consultations in
the ED, which is characterized by time pressure, simultaneity, unpredictability, and lack of
mutual background knowledge. Moreover, it looked at multiparty interactions, which included
interventions by companions who took up various roles and frequently faced significant
language barriers in their exchanges with doctors as well.
Research on language barriers in the ED is relatively scarce. Most of the existing research
pertains to the medical literature, which focuses on the impact of language barriers on relevant
outcomes such as patient satisfaction, diagnostic insecurity, length of stay, test utilisation, and
adherence to follow up instructions and does not look so much at the communicative processes
that lead to these outcomes. Only Flores et al. have considered in more detail the
communicative process, including by comparing the incidence of interpreting errors made by
professional medical interpreters with those made by non-professional ad hoc interpreters
(Flores et al. 2012). My analysis also finds that ad hoc interpreters lead to a high incidence of
significant interpreting errors. However, it takes a broader perspective and shows that relying
on patient companions for language intermediation poses problems beyond the incidence of
interpreting errors; as they may not even take up the role of ad hoc interpreter, and speak on
the patient’s behalf instead of consulting the latter, while it is difficult for the doctor to assess
the reliability of their statements due to the lack of background knowledge that is characteristic
of the ED.
Both the conceptual framework of this study (with a relatively open methodological stance and
an interdisciplinary approach), and the “superdiverse” context in which it has been applied,
present a novelty in the existing literature. This means that it should be considered as a first
exploration of what is possible and relevant in this domain; and that there remains wide scope
for further refinements of the conceptual framework. There is also plenty of scope to explore
the usefulness of this framework for application in other contexts where language discordance
weighs on the communicative process. Examples include medical contexts outside of the ED,
but also other contexts where public services are confronted with language-discordance such
Conclusion 173
as court cases involving migrants, military interventions abroad, or the management of public
services for refugees and asylum seekers.
To further expand my conceptual framework, it would be interesting to explore the subjective

experiences of patients and companions in the consultations under study. This could contribute
to further insights as to how communication in the presence of a language barrier contributes
to anxiety; and conversely, how anxiety compounds communication difficulties ensuing from
a language barrier. Additional relevant venues for future research can be the development of a
tool for quantitative analysis drawing on the framework of this study, to get a more
representative overview of which sources of miscommunication matter most, and which
strategies to prevent it are used most often. Finally, applied research that evaluates the impact
of targeted interventions such as clinical skills training on providing healthcare services in a
language discordant context can help improve practice.
9.3 Implications for practice

The study presented in this PhD mostly concerned applied research, and had a strong
underlying motivation of contributing relevant insights for the improvement of practice, more
specifically, communication in language-discordant medical interactions. As such, it is
important to highlight the main lessons one can draw for practice based on my analysis of the
three case studies, but also on wider observations made during the process of data collection
and on the insights gained from interacting with practitioners (doctors, as well as interpreters)
at various stages of the research process.
9.3.1 Addressing language barriers in the ED

Our research has shown that language barriers constitute significant hurdles for the provision
of quality healthcare. As is widely agreed in the literature, language barriers in medical practice
are best addressed by relying on professional medical interpreters. My analysis illustrates how
relying on companions to act as ad hoc interpreters is often problematic. One issue that
continues to be debated in the literature is the role(s) professional medical interpreters should
take up during a consultation according to their code of conduct (see e.g. Cox 2015).
The analysis in this study shows that language-related problems may go together with several
other, broader sources of miscommunication, that stretch far beyond the purely linguistic
aspect. It illustrates the impact of the lack of mutual pre-textuality or foreknowledge on the
patient’s and the companion’s background, on institutional and medical processes, on
expectations and emotions, on agendas, goals, and frames. The more recently immigrants have
arrived, the more significant the knowledge gap regarding institutional and medical processes
can be assumed to be, and the weaker their resources (e.g. a social network) may be to address
this knowledge gap. While this lack of mutual pre-textuality is common in language concordant
medical consultations as well, it is more easily addressed if the different participants to an
interaction have a sufficiently strong linguistic shared repertoire. In the absence of a shared
language, misunderstandings tend to persist for a longer time, and often remain undetected.
Conclusion 174
Conflicts are more difficult to overcome as the process of negotiation is complicated by the
complexity and indirectness of communication.
Consequently, to redress miscommunication in a language-discordant context, a mere focus on

the language barrier may not be sufficient. This means that professional medical interpreters
can have an important added value beyond machine translation tools. It may also mean that
professional medical interpreters may need a wider degree of discretion in their daily practice
than some codes of conduct currently allow for, to allow them to go (when needed) beyond
acting as a “machine translator”, to point out broader sources of miscommunication (when they
detect these, and it becomes clear that the other interactants do not) and assist interactants in
addressing these. This may need to be specifically addressed and further developed in the
training curriculum of both medical care providers and medical interpreters (see below).
9.3.2 Working without professional interpreters

Patient companions can contribute positively to medical consultations by providing comfort
and different types of support to patients. If they are closely related to the patient, they may as
well act as a reliable source of information on the patient’s medical history. Relying fully on
patient companions to replace professional interpreters is, however, problematic. My analysis
shows that, first, patient companions do not necessarily want to act as an interpreter. They may
have entered the consultation in their role as a husband, neighbour, passer-by. Additional
consultation time may be needed to invest in role negotiation.
Second, even if they agree to act as an interpreter, they may make a high number of interpreting
errors. These errors may stem both from the companion’s own lack of linguistic resources,
which causes errors in the interpretation of the doctor’s utterances, as well as in the
transmission of messages from the patient to the doctor. In some cases, there can be relevant
language barriers between the patient and the companion as well, further exasperating
difficulties in the exchange of information. Errors may also result from the companion’s own
agenda, which is not necessarily aligned with the doctor’s agenda. This adds further “noise” to
the information flow. Many of these misunderstandings remain undetected (for instance in the
case of “false fluency”). Textual comprehension becomes particularly important at stages of
the consultation where answers are sought to very specific questions. At such instances, correct
language use is much less likely to be redundant; and the risk of interpreting errors is
particularly high.
The member checking sessions with doctors involved in the consultations under study, revealed
that they were surprised by the amount of information loss that resulted from relying solely on
patient companions to mitigate the language barrier. This highlights the importance of using
professional interpreters. It also underscores the importance of training healthcare providers to
recognise and assess risks of miscommunication; and to raise awareness among them on the
tools and services at their disposal to address language barriers, so they do not need to rely on
patient companions.
Conclusion 175
9.3.3 Implications for healthcare staff and interpreter training

Both healthcare providers and professional medical interpreters should be trained on how to
assess the linguistic conditions of the encounter and on how to prevent or repair instances of
miscommunication.
Healthcare staff often do not perceive the need to call in an interpreter or are not always aware
of communicative mishaps (see e.g. Section 9.1.3). To raise awareness of the extent of
communication problems in the absence of professional interpreters, training curricula for ED
doctors should be enriched with training in language discordant communication, under the
specific conditions of the ED, and with particular attention for typical patient profiles. Such
training should be based on real-life examples from the medical field, like the cases presented
in this PhD and could be offered via online learning modules that can be accessed by physicians
in quieter moments in the ED, allowing them to learn while remaining at the disposal of
potential patients.
In the context of the research for this PhD, a first pilot version of such an e-module was
developed jointly with the ED doctors and the hospital training department, drawing on real-
life data and sticking closely to the typical structure of clinical reasoning. This module takes
the course participant into the interactional backstage and frontstage of the clinical encounters.
Based on real-life anonymised examples, the course participant is presented with a set of
dilemmas at the level of real communication dynamics, false fluency, and communication
difficulties. These interactional topics are embedded into the clinical skills logic as to make it
more recognisable for practitioners and to foster transfer of insights from the course to practice.
The structure of the course follows a similar logic as medical papers, discussing first the
medical motivation, then the rationale, a set of examples and a conclusion.
To generate joint ownership of the module with the medical staff, the course starts with a video
clip in which the Head of the ED unit explains why communication and hence this course is
important. In another clip, a doctor reviews briefly the medical literature on the negative
impacts of a language barrier. The course talks the course participant through a consultation
from opening to closing. Via multiple choice questions that relate to selected excerpts of
interactions and through voice off commentaries, course participants are familiarized with the
interactional characteristics and the associated clinical consequences of language discordant
medical consultations in the ED. The e-module was developed in such a way that clinical staff
can pause and restart the course at any time. This is crucial as medical staff often do not have
a lot of time in a row, as their main task is to take care of patients.
By undertaking the course, medical staff become increasingly aware and able to detect possible
communication mishaps. They may as well become increasingly aware of the benefit of
professional interpreters, and more willing to call them in. Furthermore, real-life examples can
be used to diffuse “best practices” on communication strategies and the prevention of
misunderstandings, and improve communication processes even in the absence of professional
Conclusion 176
medical interpreters (Diamond and Jacobs 2010; Leanza et al. 2014). For example, if, owing
to circumstances, doctors need to rely on companions to available to mediate the interaction,
they can be trained to ensure that, prior to the consultation, the relationship between the patient
and the companion is identified so that an adequate role distribution can be established ex-ante,
and that the companion understands the most basic rules of interpreting (Leanza et al. 2014).
Medical interpreters, on the other hand, need to become more familiar with the culture of
emergency departments, learn “how the system works and why clinicians and staff operate the
way they do” (Scheeres et al. 2008, 19). They should also become familiar with the often-
chaotic structure and the typical questions and narratives of ED encounters. In addition to basic
training in interpreting skills, the specificities of the ED context could be acquired via an on-
the-job “communicative immersion” in the ED. This would imply that interpreters shadow
doctors as they see patients and perform their typical tasks. Observation of language-
concordant medical interaction would be a crucial element of the interpreters’ learning process
as well, as to become familiar with the terminology. As healthcare staff and interpreters will
get to know each other and each other’s working culture, mutual trust can be built, which may
encourage doctors to call in an interpreter more readily when seeing foreign language speaking
patients. This “communicative immersion” will be enriching for healthcare staff as well,
because working with interpreters or cultural brokers can help healthcare staff to improve their
intercultural competence skills (Flores et al. 2002). Another advantage is that a better mutual
understanding is likely to contribute to smoothing communication in the stressful and chaotic
context of the ED.
9.3.4 Implications for organisational ED management

Despite the fact that the ED is typically characterised by intense time constraints for staff, there
is a lot of waiting involved for the patient, between intake, treatment and discharge (Slade et
al. 2008). To gain time, the “linguistic assessment” of patients can potentially already take
place prior to the consultation, such as at the time of the intake or triage. If a patient has a lot
of difficulty in expressing him- or herself, this can be mentioned in the patient’s record. Initial
steps can be undertaken to identify the patient’s native language, and if needed, an interpreter
can already be called in.
In those hospitals where on-site interpreters are present, the effective management of available
resources is key. For example, to make sure that phone numbers of on-site interpreters are
readily available, relevant phone numbers could be printed at the top of each patient file. This
would allow for a direct line to the interpreter, even when different doctors are treating the
same patient. When an interpreter is called in (s)he could be given the patient’s file number
and the internal number of the doctor who is following the patient. Such exchange of numbers
is crucial to find another back-up in the complexity of the ED (Engel et al. 2010). If the doctor
and the interpreter can easily locate each other, time is saved, and subsequent gatherings can
be organised in a flexible way, useless waiting time can be avoided, and the interpreter can
assist to other medical interactions during his/her idle time.
Conclusion 177
9.4 Limitations of the research

Clearly, the analysis in this PhD presents some limitations. First, it incorporates insights from
many different fields. Without doubt, there is still plenty of scope to further finetune its
methods for future research on similar topics. As it sought a holistic approach and therefore
tried to incorporate as much as possible relevant aspects of miscommunication, there was less
opportunity to enter into details on specific aspects.
As my analysis was based on three case studies, which were unique in their contents and their
constellation, the generalisability of the findings may be challenged. I did, however,
contextualise the data as to allow for a comparison with other contexts.
The medical interactions studied in my analysis were all of a triadic nature, involving a doctor,
a patient and the patient’s companion. The analysis was enriched with data from immediate
feedback interviews and member checking sessions with the involved doctors. For practical
and privacy reasons, however, it was not possible to perform member checking with the patient
and the patient’s companion. Furthermore, because a researcher was also present at the
interaction, there was some risk of observer’s paradox, notably that the observed interaction is
to some extent influenced by the presence of the researcher (Labov 1973, 209).
Finally, the absence of comparable counterfactuals implies that I cannot infer how the
interactional dynamics would have been different if there were no language barrier, or if a
professional medical interpreter would have been invoked. As a matter of fact, in the ED in
question, professional interpreters were rarely called in.
References 178
10 References
Agar, M. 1996. The Professional Stranger. London: Academic Press.

Ainsworth-Vaughn, N. 2001. ‘The Discourse of Medical Encounters’. In The Handbook of
Discourse Analysis, edited by D. Schiffrin, D. Tannen, and H.E. Hamilton, Blackwell
Publishers Ltd. Oxford: Blackwell Publishers Ltd.
Allison, G. 1971. Essence of Decision: Explaining the Cuban Missile Crisis. 1sted. Little
Brown.
Angelelli, C.V. 2000. ‘Interpretation as a Communicative Event: A Look through Hymes’
Lenses’. Meta: Journal des traducteurs 45 (4): 580.
———. 2003. ‘The Visible Co-Participant: The Interpreter’s Role in Doctor-Patient
Encounters’. In From Topic Boundaries to Omission: New Research on Interpretation,
edited by M. Metzger, S. Collins, V. Dively, and R. Shaw, 1:3–26. Studies in
Interpretation. Washington, D.C.: Gallaudet University Press.
———. 2004. Medical Interpreting and Cross-Cultural Communication. Cambridge
University Press.
Aranguri, C., B. Davidson, and R. Ramirez. 2006. ‘Patterns of Communication through
Interpreters: A Detailed Sociolinguistic Analysis’. Journal of General Internal
Medicine 21 (6): 623–29.
Atkins, S., C. Roberts, K. Hawthorne, and T. Greenhalgh. 2016. ‘Simulated Consultations: A
Sociolinguistic Perspective’. BMC Medical Education 16: 16. doi:10.1186/s12909-
016-0535-2.
Atkinson, P. 1995. Medical Talk and Medical Work. SAGE.
Auer, P., and C. Roberts. 2011. ‘Introduction — John Gumperz and the Indexicality of
Language’. Text & Talk 31 (4). doi:10.1515/text.2011.018.
Austin, J. L. 1962. How to Do Things with Words (William James Lectures). Oxford University
Press.
Avery, M.P.B. 2001. ‘The Role of the Health Care Interpreter: An Evolving Dialogue’. The
National Council on Interpreting in Health Care Working Papers Series. Boston:
NCIHC.
Back, A., and W. Baile. 2003. ‘Communication Skills: Myths, Realities and New
Developments’. J Support Oncol 1 (3): 169–171.
Ballinger, C. 2008. ‘Accountability’. In The SAGE Encyclopedia of Qualitative Research
Methods, by L.M. Given, 3–4. Los Angeles, California: SAGE Publications.
Bard, M. 1970. ‘The Price of Survival for Cancer Victims’. In Where Medicine Fails, edited
by A.L. Strauss, 99–110. Chicago: Aldine.
Barone, A.D., W.C. Yoels, and J.M. Clair. 1999. ‘How Physicians View Caregivers: Simmel
in the Examination Room’. Sociological Perspectives 42 (4): 673–90.
doi:10.2307/1389579.
Bavelas, J.B. 1983. ‘Situations That Lead to Disqualification’. Human Communication
Research 9: 130–45.
Bavelas, J.B., L. Coates, and T. Johnson. 2000. ‘Listeners as Co-Narrators.’ Journal of
Personality and Social Psychology 79 (6): 941–52. doi:10.1037//0022-3514.79.6.941.
Beach, W.A. 1995. ‘Preserving and Constraining Options: Okays and Official Priorities in
Medical Interviews’. In Talk of the Clinic, edited by G. Morris and R. Cheneil. Erlbaum.
———. 2006. ‘Conversation Analysis’. In The Encyclopedia of Language & Linguistics,
edited by E. K. Brown and A. Anderson, 2nd ed, 989–94. Amsterdam/Boston: Elsevier.
Becker, H.S., B. Geer, E.C. Hughes, and A.L. Strauss. 1961. Boys in White: Student Culture in
Medical School. Chicago: The University of Chicago Press.
References 179
Beckman, H.B., S. Kaplan, and R.M. Frankel. 1989. ‘Outcome-Based Research on Doctor-
Patient Communication: A Review’. In Communicating with Medical Patients, 223–
27. Newbury Park, CA: Sage Publications.
Berg, J., and P. Appelbaum. 2001. Informed Consent Legal Theory and Clinical Practice. New
York: Oxford University Press.
Bezemer, J. 2012. ‘What Is Multimodality?’ MODE: Multimodal Methodologies.
http://mode.ioe.ac.uk/2012/02/16/what-is-multimodality/.
Bhutta, Z. 2004. ‘Beyond Informed Consent’. Bulletin of the World Health Organization 82:
771–777.
Bialystok, E. 1983. ‘Some Factors in the Selection and Implementation of Communication
Strategies’. In Strategies in Interlanguage Communication, edited by C. Faerch and G.
Kasper, 100–118. Harlow, UK: Longman.
———. 1990. Communication Strategies. Oxford: Blackwell.
Bickley, L.S. 2013. Bates’ Guide to Physical Examination and History Taking. Philadelphia:
Wolters Kluwer/Lippincott Williams & Wilkins.
Biletzki, A., and A. Matar. 2014. ‘Ludwig Wittgenstein’. In The Stanford Encyclopedia of
Philosophy, edited by E.N. Zalta, Spring 2014.
BISA. 2016. ‘Bevolking: Nationaliteiten’. Statistische gegevens van het Brussels
Hoofdstedelijk Gewest. Brussel: Brussels Instituut voor Statistiek en Analyse.
Blommaert, J. 1998. ‘Different Approaches to Intercultural Communication’. Plenary lecture,
Lernen und Arbeiten in einer international vernetzten und multikulturellen
Gesellschaft, Expertentagung Universität Bremen, Institut für Projektmanagement und
Witschaftsinformatik (IPMI),. http://www.cie.ugent.be/CIE/blommaert1.htm.
———. 2007. ‘On Scope and Depth in Linguistic Ethnography’. Journal of Sociolinguistics
11 (5): 682–88. doi:10.1111/j.1467-9841.2007.00346.x.
Blommaert, J., and J. Dong. 2010. Ethnographic Fieldwork: A Beginner’s Guide. Channel
View Books.
Blommaert, J., and B. Rampton. 2012. Language and Superdiversity. MPIMMG.
www.mmg.mpg.de/fileadmin/user_upload/Subsites/Diversities/Journals_2011/2011_1
3-02_art1.pdf.
Blumer, H. 1969. Symbolic Interactionism: Perspective and Method. Berkeley: University of
California Press.
Bochner, S. 1983. ‘Doctors, Patients and Their Cultures’. In Doctor-Patient Communication,
edited by D. Pendleton and Hasler. London: Academic Press.
Boivin, I., Y. Leanza, and E. Rosenberg. 2013. ‘Conceptions des rôles des interprètes
professionnels et familiaux dans les entretiens médicaux et postures par rapport au
monde vécu’. Alterstice - Revue Internationale de la Recherche Interculturelle 2 (2):
35–46.
Bonacruz Kazzi, G., and C. Cooper. 2003. ‘Barriers to the Use of Interpreters in Emergency
Room Paediatric Consultations’. Journal of Paediatrics and Child Health 39 (4): 259–
263.
Bowen, G.A. 2006. ‘Grounded Theory and Sensitizing Concepts’. International Journal of
Qualitative Methods 5 (3): 12–23.
Bowen, S. 2001. ‘Language Barriers in Access to Health Care’. Health Canada 120.
Boyd, E., and J. Heritage. 2006. ‘Taking the History: Questioning during Comprehensive
History-Taking’. In Communication in Medical Care: Interaction between Primary
Care Physicians and Patients, edited by J. Heritage and D.W. Maynard. Studies in
Interactional Sociolinguistics 20. Cambridge University Press.
References 180
Boylorn, R.M. 2008. ‘Participants as Co-Researchers’. In The SAGE Encyclopedia of

Qualitative Research Methods. Volumes 1 & 2., edited by L.M. Given, 599–600. Los
Angeles, CA: Sage Publications.
Brindley, P.G., K.E. Smith, P. Cardinal, and F. LeBlanc. 2014. ‘Improving Medical
Communication: Skills for a Complex (and Multilingual) Clinical World’. Canadian
Respiratory Journal: Journal of the Canadian Thoracic Society 21 (2): 89.
Brisset, C., Y. Leanza, and K. Laforest. 2013. ‘Working with Interpreters in Health Care: A
Systematic Review and Meta-Ethnography of Qualitative Studies’. Patient Education
and Counseling 91 (2): 131–40.
Britten, N., and A. Shaw. 1994. ‘Patients’ Experiences of Emergency Admission: How
Relevant Is the British Government’s Patients Charter?’ Journal of Advanced Nursing
19 (6): 1212–20. doi:10.1111/j.1365-2648.1994.tb01206.x.
Britten, N., and O. Ukoumunne. 1997. ‘The Influence of Patients’ Hopes of Receiving a
Prescription on Doctors’ Perceptions and the Decision to Prescribe: A Questionnaire
Survey’. Bmj 315 (7121): 1506–1510.
Brown, G. 1995. Speakers, Listeners, and Communication: Explorations in Discourse
Analysis&DPEULGJHௗ1HZ<RUN1<86$&DPEULGJH8QLYHUVLW\3UHVV
Brown, P., and S.C. Levinson. 1987. Politeness: Some Universals in Language Usage.
Cambridge University Press.
Bührig, K., and B. Meyer. 2004. ‘Ad-Hoc Interpreting and the Achievement of Communicative
Purposes in Doctor-Patient Communication’. In Multilingual Communication, edited
by J. House and J. Rehbein, 43–62. Amsterdam; Philadelphia: J. Benjamins.
Burley, D. 2011. ‘Better Communication in the Emergency Department’. Emergency Nurse 19
(2): 32–36.
Burns, R.B. 2000. Introduction to Research Methods. 4th edition. London: Sage.
Byrne, P.S., and B.E.L. Long. 1976. Doctors Talking to Patients. A Study of the Verbal
Behaviour of General Practitioners Consulting in Their Surgeries. London, England,
Her Majesty’s Stationery Office.
Cambridge, J. 1999. ‘Information Loss in Bilingual Medical Interviews through an Untrained
Interpreter’. The Translator 5 (2): 201–19.
Candlin, C.N., and S. Candlin. 2003. ‘Health Care Communication: A Problematic Site for
Applied Linguistics Research’. Annual Review of Applied Linguistics 23: 134–154.
Caporael, L.R. 1981. ‘The Paralanguage of Caregiving: Baby Talk to the Institutionalized Aged
- the Field Study.’ Journal of Personality and Social Psychology 40: 876–84.
Carrasquillo, O, E J Orav, T A Brennan, and H R Burstin. 1999a. ‘Impact of Language Barriers
on Patient Satisfaction in an Emergency Department’. Journal of General Internal
Medicine 14 (2): 82–87.
Carrasquillo, O., E.J. Orav, T.A. Brennan, and H.R. Burstin. 1999b. ‘Impact of Language
Barriers on Patient Satisfaction in an Emergency Department’. Journal of General
Internal Medicine 14 (2): 82–87.
Cartwright, A. 1964. Human Relations and Hospital Care. Routledge & K. Paul.
Cerf, D. 2012. ‘Le Bon Usage Du Service Des Urgences’. presented at the Séminaires des
Services d’Urgences IRIS, IRIS Public Hospital Network, Brussels.
Chisholm, C.D., A.M. Dornfeld, D.R. Nelson, and W.H. Cordell. 2001. ‘Work Interrupted: A
Comparison of Workplace Interruptions in Emergency Departments and Primary Care
Offices’. Annals of Emergency Medicine 38 (2): 146–51.
doi:10.1067/mem.2001.115440.
Cicourel, A.V. 1987. ‘The Interpenetration of Communicative Contexts: Examples from
Medical Encounters’. Social Psychology Quarterly 50 (2): 217–26.
References 181
Clark, M. M. 1983. ‘Cultural Context of Medical Practice.’ Western Journal of Medicine 139
(6): 806.
Clayman, M.L., and M.A. Morris. 2013. ‘Patients in Context: Recognizing the Companion as
Part of a Patient-Centered Team’. Patient Education and Counseling 91 (1): 1–2.
doi:10.1016/j.pec.2013.02.004.
Cole, S.A., and J. Bird. 2000. 7KH0HGLFDO,QWHUYLHZࣟ7KH7KUHH-Function Approach. Elsevier
Health Sciences.
Collins, J., and S. Slembrouck. 2006. ‘“You Don’t Know What They Translate”: Language
Contact, Institutional Procedure, and Literacy Practice in Neighborhood Health Clinics
in Urban Flanders’. Journal of Linguistic Anthropology 16 (2): 249–268.
Coope, J., and D. H. H. Metcalfe. 1979. ‘How Much Do Patients Know? A Multiple Choice
Question Paper for Patients in the Waiting Room’. Journal of the Royal College of
General Practitioners 29 (205): 482–87.
Cossey K., and Jeanmonod R. 2012. ‘Impact of Language Discordance on Door-to-Room Time
and Patient Satisfaction in Triage’. Annals of Emergency Medicine 60 (4): S113–14.
Coupland, J., J.F. Nussbaum, and N. Coupland. 1991. ‘The Reproduction of Ageing and Agism
in Intergenerational Talk’. In ‘Miscommunication’ and Problematic Talk, edited by N.
Coupland, H. Giles, and J.M. Wiemann, 85–102. Newbury Park, California: Sage
Publications.
Coupland, N., H. Giles, and J.M. Wiemann, eds. 1991. ‘Miscommunication’ and Problematic
Talk. Newbury Park: Sage Publications.
Coupland, N., J.M. Wiemann, and H. Giles. 1991. ‘Talk as “Problem” and Communication as
“Miscommunication”: An Integrative Analysis’. In ‘Miscommunication’ and
Problematic Talk, edited by N. Coupland, H. Giles, and J.M. Wiemann, 1–17. Newbury
Park, California: Sage Publications.
Cox, A. 2015. ‘Do You Get the Message? Defining the Interpreter’s Role in Medical
Interpreting’. MonTI, Mongraphs in Translation and Interpreting Special Issue-
Interpreting: 161–84.
Cox, A., K. Kerremans, and R. Temmerman. 2013. ‘Niche Sourcing and Transexplanations for
the Enhancement of Doctor-Patient Comprehension in Multilingual Hospital Settings’.
In Proceedings of the 10th International Conference on Terminology and Artificial
Intelligence TIA 2013, 33–36. Paris.
Cox, A., and R. Lázaro Gutiérrez. 2016. ‘Interpreting in the Emergency Department: How
Context Matters for Practice’. In Mediating Emergencies and Conflicts, edited by
Federico M. Federici, 33–58. London: Palgrave Macmillan UK.
http://link.springer.com/10.1057/978-1-137-55351-5_2.
Crane, J.A. 1997. ‘Patient Comprehension of Doctor-Patient Communication on Discharge
from the Emergency Department’. The Journal of Emergency Medicine 15 (1): 1–7.
Creese, A. 2008. ‘Linguistic Ethnography’. In Encyclopedia of Language and Education,
3424–3436. Springer. http://link.springer.com/10.1007/978-0-387-30424-3_257.
Creese, A., and F. Copland. 2015. ‘Linguistic Ethnography’. In Research Methods in Language
and Education, edited by K. King, Y. Lai, and S. May, 1–13. Cham: Springer
International Publishing. http://link.springer.com/10.1007/978-3-319-02329-8_25-1.
Creswell, J.W. 2014. Research Design: Qualitative, Quantitative, and Mixed Methods
Approaches. 4thed. Thousand Oaks, CA: Sage.
David, M., I. Schwartau, H. Anand Pant, and T. Borde. 2006. ‘Emergency Outpatient Services
in the City of Berlin: Factors for Appropriate Use and Predictors for Hospital
Admission’. European Journal of Emergency Medicine 13 (6): 352.
References 182
Davidson, B. 2000. ‘The Interpreter as Institutional Gatekeeper: The Social-Linguistic Role of

Interpreters in Spanish-English Medical Discourse.’ Journal of Sociolinguistics 4 (3):
379–405.
Davis, F. 1960. ‘Uncertainty in Medical Prognosis Clinical and Functional’. American Journal
of Sociology 66 (1): 41–47.
———. 1963. Passage through Crisis: Polio Victims and Their Families. Bobbs-Merrill.
Debosscher, T. 2012. ‘Infogids Voor Hulpverleners En Begeleiders van Mensen Zonder Wettig
Verblijf’. Kruispunt Migratie-Integratie vzw.
Dewey, J. 1939. Logic: The Theory Of Inquiry. New York: Henry Holt And Company Inc.
Diamond, L.C., and E.A. Jacobs. 2010. ‘Let’s Not Contribute to Disparities: The Best Methods
for Teaching Clinicians How to Overcome Language Barriers to Health Care’. Journal
of General Internal Medicine 25 (S2): 189–93. doi:10.1007/s11606-009-1201-8.
Divi, C., R.G. Koss, S.P. Schmaltz, and J.M. Loeb. 2007. ‘Language Proficiency and Adverse
Events in US Hospitals: A Pilot Study’. International Journal for Quality in Health
Care 19 (2): 60–67.
Dörnyei, Z., and M.L. Scott. 1995a. ‘Communication Strategies: An Empirical Analysis with
Retrospection’. In Selected Papers from the Proceedings of the 21t Annual Symposium
of the Deseret Language and Linguistics Society, edited by J.S. Turley and K. Lusby,
155–68. Provo, UT: Brigham Young University.
———. 1995b. ‘Communication Strategies: What Are They and What Are They Not?’ Paper
presented at the Annual Conference of the American Association for Applied
Linguistics, Long Beach, California.
———. 1997. ‘Communication Strategies in a Second Language: Definitions and
Taxonomies’. Language Learning 47 (1): 173–210.
Dryden, C., and H. Giles. 1987. ‘Language, Social Identity, and Health’. In Psychology Survey
6, 115–39. Levister: British Psychological Society.
Duneier, M. 1992. Slim’s Table: Race, Respectability, and Masculinity. Chicago: University
of Chicago Press.
Duranti, A. 2005. ‘Ethnography of Speaking: Toward a Linguistics of the Praxis’. In
Intercultural Discourse and Communication: The Essential Readings, by S.F. Kiesling
and C.B. Paulston, 17–32. Malden, MA: Blackwell Pub.
Eisenberg, E.M., A.G. Murphy, K. Sutcliffe, R. Wears, S. Schenkel, S. Perry, and M.
Vanderhoef. 2005. ‘Communication in Emergency Medicine: Implications for Patient
Safety’. Communication Monographs 72 (4): 390–413.
Elderkin-Thompson, Virginia, Roxane Cohen Silver, and Howard Waitzkin. 2001. ‘When
Nurses Double as Interpreters: A Study of Spanish-Speaking Patients in a US Primary
Care Setting’. Social Science & Medicine 52 (9): 1343–1358.
Ellingson, L.L. 2002. ‘The Roles of Companions in Geriatric Patient–interdisciplinary
Oncology Team Interactions’. Journal of Aging Studies 16 (4): 361–82.
doi:10.1016/S0890-4065(02)00071-3.
Engel, K.G., B.A. Buckley, D.M. McCarthy, V.E. Forth, and J.G. Adams. 2010.
‘Communication Amidst Chaos: Challenges to Patient Communication in the
Emergency’. JCOM 17 (10). http://www.abim.org.www.turner-
white.com/pdf/jcom_oct10_emergency.pdf.
Faerch, C., and G. Kasper. 1983. ‘Plans and Strategies in Foreign Language Communication.’
In Strategies in Interlanguage Communication, edited by C. Faerch and G. Kasper.
Harlow, UK: Longman.
Fagan, M.J., J.A. Diaz, S.E. Reinert, C.N. Sciamanna, and D.M. Fagan. 2003. ‘Impact of
Interpretation Method on Clinic Visit Length’. Journal of General Internal Medicine
18 (8): 634–638.
References 183
Fedasil. 2011. ‘Fedasil: Centrum van Brussel (Klein Kasteeltje) Beschrijving.’

http://fedasil.be/nl/center/749/description.
Fetterman, D.M. 2008. ‘Ethnography’. In The SAGE Encyclopedia of Qualitative Research
Methods Volume 1 & 2, edited by L.M. Given, 288–92. Los Angeles, CA: Sage
Publications.
Fielding, N. 2001. ‘Ethnography’. In Researching Social Life, edited by Nigel Gilbert. SAGE.
Flores, G., M. Abreu, C.P. Barone, R. Bachur, and H. Lin. 2012. ‘Errors of Medical
Interpretation and Their Potential Clinical Consequences: A Comparison of
Professional Versus Ad Hoc Versus No Interpreters’. Annals of Emergency Medicine
60 (5): 545–53.
Flores, G., M. B. Laws, S. J. Mayo, B. Zuckerman, M. Abreu, L. Medina, and E. J. Hardt. 2003.
‘Errors in Medical Interpretation and Their Potential Clinical Consequences in Pediatric
Encounters’. Pediatrics 111 (1): 6–14. doi:10.1542/peds.111.1.6.
Flores, G., J. Rabke-Verani, W. Pine, and A. Sabharwal. 2002. ‘The Importance of Cultural
and Linguistic Issues in the Emergency Care of Children’. Pediatric Emergency Care
18 (4): 271–84.
Fosnocht, D.E., M.B. Hollifield, and E.R. Swanson. 2004. ‘Patient Preference for Route of
Pain Medication Delivery’. The Journal of Emergency Medicine 26 (1): 7–11.
doi:10.1016/j.jemermed.2003.03.002.
Fosnocht, D.E., E.R. Swanson, and E.D. Barton. 2005. ‘Changing Attitudes About Pain and
Pain Control in Emergency Medicine’. Emergency Medicine Clinics of North America
23 (2): 297–306. doi:10.1016/j.emc.2004.12.003.
Fosnocht, D.E., E.R. Swanson, and P. Bossart. 2001. ‘Patient Expectations for Pain Medication
Delivery’. The American Journal of Emergency Medicine 19 (5): 399–402.
doi:10.1053/ajem.2001.24462.
Gany, F., J. Leng, E. Shapiro, D. Abramson, I. Motola, D.C. Shield, and J. Changrani. 2007.
‘Patient Satisfaction with Different Interpreting Methods: A Randomized Controlled
Trial’. Journal of General Internal Medicine 22 (S2): 312–18. doi:10.1007/s11606-
007-0360-8.
Gao, G., N. Burke, C. P. Somkin, and R. Pasick. 2009. ‘Considering Culture in Physician--
Patient Communication During Colorectal Cancer Screening’. Qualitative Health
Research 19 (6): 778–89.
Garfinkel, H. 1994. Studies in Ethnomethodology. Wiley-Blackwell.
Garra, G., H. Albino, H. Chapman, A.J. Singer, and H.C. Thode. 2010. ‘The Impact of
Communication Barriers on Diagnostic Confidence and Ancillary Testing in the
Emergency Department’. The Journal of Emergency Medicine 38 (5): 681–85.
doi:10.1016/j.jemermed.2009.01.004.
Gass, S., and E. M. Varonis. 1991. ‘Miscommunication in Nonnative Speaker Discourse’. In
‘Miscommunication’ and Problematic Talk, edited by N. Coupland, H. Giles, and J.M.
Wiemann, 121–45. Newbury Park: Sage Publications.
Geertz, C. 1973. The Interpretation of Cultures: Selected Essays. Basic Books.
Gerwing, J., and M. Allison. 2009. ‘The Relationship between Verbal and Gestural
Contributions in Conversation: A Comparison of Three Methods’. Gesture 9 (3): 312–
36.
Gerwing, J., and A.M. Landmark Dalby. 2014. ‘Gestures Convey Content: An Exploration of
the Semantic Functions of Physicians’ Gestures’. Patient Education and Counseling 96
(3): 308–14.
Giles, H., J. Coupland, and N. Coupland, eds. 1991a. Contexts of Accommodation:
Developments in Applied Sociolinguistics. Studies in Emotion and Social Interaction.
References 184
New York: Cambridge University Press; Paris: Editions de la Maison des science de
l’homme.
Giles, H., N. Coupland, and J. Coupland. 1991b. ‘Accommodation Theory: Communication,
Context, and Consequence’. In Contexts of Accommodation: Developments in Applied
Sociolinguistics, edited by H. Giles, J. Coupland, and N. Coupland, 1–68. Studies in
(PRWLRQDQG6RFLDO,QWHUDFWLRQ&DPEULGJH>(QJODQG@ௗ1HZ<RUNௗ Paris: Cambridge
8QLYHUVLW\3UHVVௗ(GLWLRQVGHOD0DLVRQGHVVFLHQFHGHO¶KRPPH
Giles, H., A. Mulac, J.J. Bradac, and P. Johnson. 1987. ‘Speech Accommodation Theory: The
next Decade and Beyond.’ In Communication Yearbook 10, edited by M. McLaughlin,
13–48. Newbury Park, California: Sage Publications.
Glaser, B.G., and A.L. Strauss. 1965. Awareness of Dying. Chicago: Aldine Publishing
Company.
Gluckman, M. 1961. ‘Ethnographic Data in British Social Anthropology*’. The Sociological
Review 9 (1): 5–17. doi:10.1111/j.1467-954X.1961.tb01082.x.
Goffman, E. 1959. The Presentation of Self in Everyday Life. Penguin.
———. 1967. Interaction Ritual: Essays on Face-to-Face Behavior. Oxford, UK: Aldine.
———. 1974. Frame Analysis: An Essay on the Organization of Experience. Harper & Row.
———. 1981a. ‘Footing’. In Forms of Talk, 124–59. University of Pennsylvania Press.
———. 1981b. Forms of Talk. University of Pennsylvania Press.
———. 1983. ‘The Interaction Order: American Sociological Association, 1982 Presidential
Address’. American Sociological Review 48 (1): 1–17.
———. 1986. ‘Ordinary Troubles’. In Frame Analysis: An Essay on the Organization of
Experience, Northeastern University Press ed, 301–44. Boston: Northeastern
University Press.
Goldman, R.D., P. Amin, and A. Macpherson. 2006. ‘Language and Length of Stay in the
Pediatric Emergency Department’. Pediatric Emergency Care 22 (9): 640–43.
doi:10.1097/01.pec.0000227865.38815.ec.
Goodwin, C. 1981. Conversational Organization: Interaction between Speakers and Hearers.
Academic Press.
———. 2000. ‘Action and Embodiment within Situated Human Interaction’. Journal of
Pragmatics 32: 1489–1522.
———. 2007. ‘Interactive Footing’. Studies in Interactional Sociolinguistics 24: 16.
Goodwin, C., and A. Duranti. 1992. ‘Rethinking Context: An Introduction’. In Rethinking
Context: Language as an Interactive Phenomenon, edited by A. Duranti and C.
Goodwin, 1–42. Cambridge University Press.
Gordon, C. 2011. ‘Conversation and Interaction’. In The Cambridge Handbook of
Sociolinguistics, edited by R. Mesthrie. New York: Cambridge University Press.
Gray, B., J. Hilder, and H. Donaldson. 2011. ‘Why Do We Not Use Trained Interpreters for
All Patients with Limited English Proficiency? Is There a Place for Using Family
Members?’ Australian Journal of Primary Health 17 (3): 240. doi:10.1071/PY10075.
Greenfield, S., S. Kaplan, and J. Ware. 1985. ‘Expanding Patient Involvement in Care: Effect
on Patient Outcomes.’ Annals of Internal Medicine 102: 520–28.
Greenhalgh, T., N. Robb, and G. Scambler. 2006. ‘Communicative and Strategic Action in
Interpreted Consultations in Primary Health Care: A Habermasian Perspective’. Social
Science & Medicine 63 (5): 1170–87.
Grice, H. P. 1991. Studies in the Way of Words. Harvard University Press.
Grice, H.P. 1975. ‘Logic and Conversation’. In Speech Acts, edited by P. Cole and J. L.
Morgan, 5. ed, 41–58. Syntax and Semantics 3. New York u.a: Academic Press.
Guba, E.G. 1990. ‘The Alternative Paradigm Dialog’. In The Paradigm Dialog, 17–30.
Newbury Park, CA: Sage.
References 185
Gumperz, J.J. 1982a. Language and Social Identity. New York: Cambridge University Press.
———. 1982b. Discourse Strategies. Cambridge University Press.
———. 2001. ‘Interactional Sociolinguistics: A Personal Perspective’. In The Handbook of
Discourse Analysis, 215–88. Malden, Mass.: Blackwell Publishers.
Gumperz, J.J., and N. Berenz. 1993. ‘Transcribing Conversational Exchanges’. In Talking
Data: Transcription and Coding in Discourse Research, edited by Jane Anne Edwards
and Martin D. Lampert, 91–121. Hillsdale, N.J: Lawrence Erlbaum Associates.
Gumperz, J.J., and J. Cook-Gumperz. 1982. ‘Introduction: Language and the Communication
of Social Identity’. In Language and Social Identity, edited by J.J. Gumperz, 1–21. New
York: Cambridge University Press.
Gumperz, J.J., and D. Tannen. 1979. ‘Individual and Social Differences in Language Use’. In
Individual Differences in Language Ability and Language Behavior, edited by C.J.
Fillmore, D. Kempler, and W.S.Y. Wang, 305–25. Perspectives in Neurolinguistics and
Psycholinguistics. New York: Academic Press.
Günthner, S. 2008. ‘Interactional Sociolinguistics’. In Handbook of Interpersonal
Communication, edited by G. Antos and E. Ventola, Volume 2:53–76. Berlin/New
York: De Gruyter Mouton.
Ha, J.F., and N. Longnecker. 2010. ‘Doctor-Patient Communication: A Review’. The Ochsner
Journal 10 (1): 38–43.
Haidet, P. 2007. ‘Jazz and the “Art” of Medicine: Improvisation in the Medical Encounter’.
The Annals of Family Medicine 5 (2): 164–69. doi:10.1370/afm.624.
Hale, S., and J. Napier. 2014. Research Methods in Interpreting: A Practical Resource.
Bloomsbury Academic.
Hammersley, M., and P. Atkinson. 2010. Ethnography: Principles in Practice. London; New
York: Routledge.
Hampers, Louis C., Susie Cha, David J. Gutglass, Helen J. Binns, and Steven E. Krug. 1999.
‘Language Barriers and Resource Utilization in a Pediatric Emergency Department’.
Pediatrics 103 (6): 1253–1256.
Hansell, M., and C. Seabrook Ajirobutu. 1982. ‘Negotiating Interpretations in Interethnic
Settings’. In Language and Social Identity, edited by J.J. Gumperz, 85–94. New York:
Heinrichsmeier, R. 2016. ‘The Interactional Construction of Ageing Identities: A Linguistic
Ethnography of Older Women’s Narratives, Talk and Other Practices in a Hair-Salon’.
King’s College London.
Henley, N.M., and C. Kramarae. 1991. ‘Gender, Power, and Miscommunication’. In
‘Miscommunication’ and Problematic Talk, edited by N. Coupland, H. Giles, and J.M.
Wiemann, 18–43. Newbury Park, California: SAGE Publications.
Heritage, J., and D.W. Maynard. 2006. ‘Introduction: Analyzing Interaction between Doctors
and Patients in Primary Care Encounters’. In Communication in Medical Care
Interaction between Primary Care Physicians and Patients, 1–21. Studies in
Interactional Sociolinguistics. Cambridge University Press.
Heritage, J., and J.D. Robinson. 2006. ‘Accounting for the Visit: Giving Reasons for Seeking
Medical Care’. In Communication in Medical Care Interaction between Primary Care
Physicians and Patients, edited by J. Heritage and D.W. Maynard, 48–85. Cambridge,
UK; New York: Cambridge University Press.
Ho, A. 2008. ‘Using Family Members as Interpreters in the Clinical Setting’. J Clin Ethics 19
(3): 223–233.
Holtgraves, T.M., J. Bavelas, J. Gerwing, and S. Healing. 2014. ‘Hand and Facial Gestures in
Conversational Interaction’. In The Oxford Handbook of Language and Social
Psychology, edited by T.M. Holtgraves. Oxford University Press.
References 186
Hubbard, G., N. Illingworth, N. Rowa-Dewar, L. Forbat, and N. Kearney. 2010. ‘Treatment

Decision-Making in Cancer Care: The Role of the Carer’. Journal of Clinical Nursing
19 (13–14): 2023–31. doi:10.1111/j.1365-2702.2009.03062.x.
Hugh-Jones, P., A.F. Tanser, and C. Whitby. 1964. ‘Patient’s Views of Admission to a London
Teaching Hospital’. British Medical Journal 2: 660–64.
Hull, M. 2010. Medical English Clear & Simple: A Practice-Based Approach to English for
ESL Healthcare Professionals. F.A. Davis.
Hutchby, I., and R. Wooffitt. 2008. Conversation Analysis. Polity.
Hymes, D.H., ed. 1972. ‘Models of the Interaction of Language and Social Life’. In Directions
in Sociolinguistics: The Ethnography of Communication, by J.J. Gumperz and D.H.
Hymes. New York: Holt, Rinehart, Winston.
———. 1974. Foundations in Sociolinguistics; an Ethnographic Approach. Philadelphia:
University of Pennsylvania Press.
———. 1980. Language in Education: Ethnolinguistic Essays. Language and Ethnography
Series 1. Washington, D.C: Center for Applied Linguistics.
———. 2004. Ethnography, Linguistics, Narrative Inequality toward an Understanding of
Voice. London; Bristol, PA: Taylor & Francis.
Innes, J. M. 1977. ‘Does the Professional Know What the Client Wants?’ Social Science &
Medicine (1967) 11 (11): 635–38.
Jenks, C. 2014. ‘CONVERSATION ANALYSIS’. In The Routledge Handbook of English
Studies, edited by Constant Leung and Brian V. Street, 274–/RQGRQௗ1HZ<RUN
Routledge, Taylor & Francis Group.
Jones, R.H. 2013. ‘Analysis of Discourse and Interaction: Overview’. In The Encyclopedia of
Applied Linguistics, edited by C.A. Chapelle, 70–73. John Wiley & Sons.
Kalkan, A., O. Kose, M. Tas, and G. Meric. 2013. ‘Review of Techniques for the Removal of
Trapped Rings on Fingers with a Proposed New Algorithm’. The American Journal of
Emergency Medicine 31 (11): 1605–11.
Karliner, L. S., A. Auerbach, A. Nápoles, D. Schillinger, D. Nickleach, and E.J. Pérez-Stable.
2012. ‘Language Barriers and Understanding of Hospital Discharge Instructions’.
Medical Care 50 (4): 283–89.
Karliner, L.S., E.A. Jacobs, A.H. Chen, and S. Mutha. 2007. ‘Do Professional Interpreters
Improve Clinical Care for Patients with Limited English Proficiency? A Systematic
Review of the Literature’. Health Services Research 42 (2): 727–54.
Karliner, L.S., E.J. Pérez-Stable, and G. Gildengorin. 2004. ‘The Language Divide’. Journal
of General Internal Medicine 19 (2): 175–183.
Katz, A.M., L. Conant, T.S. Inui, D. Baron, and D. Bor. 2000. ‘A Council of Elders: Creating
a Multi-Voiced Dialogue in a Community of Care’. Social Science & Medicine 50 (6):
851–860.
Kaufert, J.M., and W.W. Koolage. 1984. ‘Role Conflict among “culture Brokers”: The
Experience of Native Canadian Medical Interpreters’. Social Science & Medicine 18
(3): 283–286.
Kecskés, I. 2004. ‘Editorial: Lexical Merging, Conceptual Blending, and Cultural Crossing’.
Intercultural Pragmatics 1 (1): 1–26.
———. 2010. ‘The Paradox of Communication: Socio-Cognitive Approach to Pragmatics’.
Pragmatics and Society 1 (1): 50–73. doi:10.1075/ps.1.1.04kec.
———. 2014. Intercultural Pragmatics. Oxford: Oxford University Press.
Knopp, R., S. Rosenzweig, E. Bernstein, and V. Totten. 1996. ‘Physician-Patient
Communication in the Emergency Department, Part 1’. Academic Emergency Medicine
3 (11): 1065–1076.
References 187
Korsch, B. M., and V. F. Negrete. 1972. ‘Doctor-Patient Communication’. Scientific American

227 (2): 66–74.
Korsch, B.M., E. Gozzi, and V. Francis. 1968. ‘Gaps in Doctor-Patient Communication I:
Doctor-Patient Interaction and Patient Satisfaction’. Pediatrics 42: 855–71.
Krystallidou, D. 2012. ‘On Mediating Agents’ Moves and How They Might Affect Patient-
Centredness in Mediated Medical Consultations’. Linguistica Antverpiensia, New
Series–Themes in Translation Studies 11: 75–93.
———. 2013. ‘The Interpreter’s Role in Medical Consultations as Perceived and as
Interactionally Negotiated: A Study of a Flemish Hospital Setting, Using Interview
Data and Video Recorded Interactions’. Dissertation, Ghent University.
———. 2014. ‘Gaze and Body Orientation as an Apparatus for Patient Inclusion Into/exclusion
from a Patient-Centred Framework of Communication’. The Interpreter and Translator
Trainer 8 (3): 399–417.
Kuhn, G.J. 2002. ‘Diagnostic Errors’. Academic Emergency Medicine 9 (7): 740–750.
Kurtz, S.M., J. Silverman, J. Benson, and J. Draper. 2003. ‘Marrying Content and Process in
Clinical Method Teaching: Enhancing the Calgary-Cambridge Guides’. Academic
Medicine 78 (8): 802–9.
Labov, W. 1973. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press.
Lachance, P.A. 2016. ‘La Communication À L’urgence Ou L’urgence de La Communication’.
In La Communication Professionnelle En Santé, edited by M.T. Lussier and C. Richard,
617–42. 2e Édition 2016. Montréal: Pearson.
Lacourt, I. 2007. ‘“What Are Your Plans?” The Socio-Occupational Integration of Brussels’
Welfare Offices’ Clients’. Brussels Studies 5.
http://www.briobrussel.be/assets/andere%20publicaties/en_36_bs5en.pdf.
Laidsaar-Powell, R.C., P.N. Butow, S. Bu, C. Charles, A. Gafni, W.W.T. Lam, J. Jansen, et al.
2013. ‘Physician–patient–companion Communication and Decision-Making: A
Systematic Review of Triadic Medical Consultations’. Patient Education and
Counseling 91 (1): 3–13. doi:10.1016/j.pec.2012.11.007.
Laxmisan, A., F. Hakimzada, O.R. Sayan, R.A. Green, J. Zhang, and V.L. Patel. 2007. ‘The
Multitasking Clinician: Decision-Making and Cognitive Demand during and after
Team Handoffs in Emergency Care’. International Journal of Medical Informatics 76
(11–12): 801–11.
Le, Heiskell, and Pasnau Ro. 1991. ‘Psychological Reaction to Hospitalization and Illness in
the Emergency Department.’ Emergency Medicine Clinics of North America 9 (1):
207–18.
Leanza, Y. 2005. ‘Roles of Community Interpreters in Pediatrics as Seen by Interpreters,
Physicians and Researchers’. Interpreting 7 (2): 167–192.
———. 2007. ‘Roles of Community Interpreters in Pediatrics as Seen by Interpreters,
Physicians and Researchers’. In Healthcare Interpreting: Discourse and Interaction,
edited by F. Pöchhacker and M. Shlesinger, 11–34. John Benjamins Publishing.
Leanza, Y., I. Boivin, and E. Rosenberg. 2010. ‘Interruptions and Resistance: A Comparison
of Medical Consultations with Family and Trained Interpreters’. Social Science &
Medicine 70 (12): 1888–95. doi:10.1016/j.socscimed.2010.02.036.
Leanza, Y., A. Miklavcic, I. Boivin, and E. Rosenberg. 2014. ‘Working with Interpreters’. In
Cultural Consultation, edited by L.J. Kirmayer, J. Guzder, and C. Rousseau, 89–114.
International and Cultural Psychology. New York, NY: Springer New York.
http://link.springer.com/10.1007/978-1-4614-7615-3.
Léon, J. 2006. ‘Goffman, Erving (1922-1982)’. In The Encyclopedia of Language &
Linguistics, edited by E. K. Brown and AnnH$QGHUVRQQGHG$PVWHUGDPௗ%RVWRQ
Elsevier.
References 188
Levinson, S.C. 1980. ‘Speech Act Theory: The State of the Art’. Language Teaching 13 (1–2):
5.
———. 1992. ‘Activity Types and Language’. In Talk at Work: Interaction in Institutional
Settings, edited by P. Drew and J. Heritage, 66–100. Cambridge [England]; New York:
Levitt Center. 2016. ‘What Is Social Emergency Medicine?’ Andrew Levitt Center for Social
Emergency Medicine. http://www.levittcenter.org/what-is-social-emergency-
medicine/.
Ley, P., and M.S. Spelman. 1967. Communicating with the Patient. London: Staples Press.
Li, S. 2011. Understanding Interactions in Interpreted Triadic Medical Consultations in
Primary Care. University of Leeds. http://etheses.whiterose.ac.uk/2109/.
———. 2013. ‘Co-Construction of Interpreted Conversation in Medical Consultations’.
Applied Linguistics Review 4 (1): 127–149.
———. 2015. ‘Nine Types of Turn-Taking in Interpreter-Mediated GP Consultations’.
Applied Linguistics Review 6 (1): 73–96. doi:10.1515/applirev-2015-0004.
Li, S., J. Gerwing, D. Krystallidou, A. Rowlands, A. Cox, and P. Pype. 2017. ‘Interaction—A
Missing Piece of the Jigsaw in Interpreter-Mediated Medical Consultation Models’.
Patient Education and Counseling (In press). doi:10.1016/j.pec.2017.04.021.
Li, S., D. Pearson, and S. Escott. 2010. ‘Language Barriers within Primary Care Consultations:
An Increasing Challenge Needing New Solutions’. Education for Primary Care 21 (6):
385–391.
Lin, C.-S., M.-Y. F. Hsu, and C.-F. Chong. 2008. ‘Differences between Emergency Patients
and Their Doctors in the Perception of Physician Empathy: Implications for Medical
Education’. Education for Health 21 (2): 144.
Lincoln, Y.S., S.A. Lynham, and E.G. Guba. 2011. ‘Paradigmatic Controversies,
Contradictions, and Emerging Confluences Revisited.’ In The SAGE Handbook of
Qualitative Research, edited by N.K. Denzin and Y.S. Lincoln, 4thed., 97–128.
Thousand Oaks, CA: Sage.
Lindblom, K. 2006. ‘Cooperation Principle’. In The Encyclopedia of Language & Linguistics,
HGLWHGE\(.%URZQDQG$$QGHUVRQQGHG$PVWHUGDPௗ%RVWRQ(OVHYLHU
Lipkin, M. 2008. ‘The Medical Interview’. In Behavioral Medicine a Guide for Clinical
Practice, edited by Mitchell D Feldman and John F Christensen. New York: McGraw-
Hill Medical. http://accessmedicine.mhmedical.com/book.aspx?bookId=373.
Long, M.H. 1983. ‘Linguistic and Conversational Adjustments to Non-Native Speakers’.
Studies in Second Language Acquisition 5: 177–93.
———. 1996. ‘The Role of the Linguistic Environment in Second Language Acquisition.’ In
Handbook of Research on Second Language Acquisition, edited by W. Ritchie and T.
Bhatia, 413–69. New York: Academic Press.
Lussier, M.T., and C. Richard. 2008. ‘Because One Shoe Doesn’t Fit All A Repertoire of
Doctor-Patient Relationships’. Canadian Family Physician 54 (8): 1089–1092.
Mack, N., C. Woodsong, K. M. Macqueen, G. Guest, and E. Namey. 2005. Qualitative
Research Methods: A Data Collector’s Field Guide. Family Health International &
USAID.
Mahmoud I., Hou X.-Y., Chu K., and Clark M. 2013. ‘Language Affects Length of Stay in
Emergency Departments in Queensland Public Hospitals’. World Journal of
Emergency Medicine 4 (1): 5–9.
Marshall, P. 2007. Ethical Challenges in Study Design and Informed Consent for Health
Research in Resource-Poor Settings. Geneva, Switzerland: World Health Organization.
References 189
Martin, G.S. 2015. ‘“Sorry Can You Speak It in English with Me?” Managing Routines in
Lingua Franca Doctor–Patient Consultations in a Diabetes Clinic’. Multilingua 34 (1):
1–32.
Maryns, K. 2004. ‘The Asylum Speaker: An Ethnography of Language and Communication in
the Belgian Asylum Procedure’. Universiteit Gent.
Maryns, K., and J. Blommaert. 2002. ‘Pretextuality and Pretextual Gaps: On Re/defining
Linguistic Inequality’. Pragmatics 12 (1): 11–30.
Maslow, A.H. 1966. Psychology of Science. First Edition edition. Place of publication not
identified: Joanna Cotler Books.
Mayo Clinic. 2015. ‘Mastitis’. Rochester, Minnesota, USA. www.mayoclinic.org/diseases-
conditions/mastitis/basics/definition/con-20026633.
———. 2016a. ‘Diseases and Conditions: Kidney Stones Symptoms’. Diseases and
Conditions. www.mayoclinic.org/diseases-conditions/kidney-
stones/basics/symptoms/con-20024829.
———. 2016b. ‘Symptoms and Causes - Tuberculosis’. www.mayoclinic.org/diseases-
conditions/tuberculosis/symptoms-causes/dxc-20188557.
Mazzi, M.A., M. Rimondini, M. Deveugele, C. Zimmermann, F. Moretti, L. van Vliet, G.
Deledda, I. Fletcher, and J. Bensing. 2015. ‘What Do People Appreciate in Physicians’
Communication? An International Study with Focus Groups Using Videotaped
Medical Consultations’. Health Expectations 18 (5): 1215–26. doi:10.1111/hex.12097.
McCaslin, M.L. 2008. ‘Pragmatism’. In The SAGE Encyclopedia of Qualitative Research
Methods, edited by L.M. Given, 671–75. Los Angeles, Calif: SAGE Publications.
McIntosh, J. 1974. ‘Processes of Communication, Information Seeking and Control Associated
with Cancer: A Selective Review of the Literature’. Social Science & Medicine (1967)
8 (4): 167–87. doi:10.1016/0037-7856(74)90107-3.
Meeuwesen, L. 2003. ‘Sequential Analysis of the Phasing of the Medical Interview’.
Epidemiologia E Psichiatria Sociale 12 (2): 124–129.
Meeuwesen, L., and E. Ani. 2011. ‘If You Don’t Understand What I Mean...’ interpreting in
Health and Social Care. Utrecht: Centre for Social Policy and Intervention Studies.
Meuter, R.F.I., C. Gallois, N.S. Segalowitz, A.G. Ryder, and J. Hocking. 2015. ‘Overcoming
Language Barriers in Healthcare: A Protocol for Investigating Safe and Effective
Communication When Patients or Clinicians Use a Second Language’. BMC Health
Services Research 15 (September). doi:10.1186/s12913-015-1024-8.
Meyer, B., B. Pawlack, and K. Ortun. 2010. ‘Family Interpreters in Hospitals: Good Reasons
for Bad Practice?’ mediAzioni, no. 10: 297–324.
Milroy, L. 1984. ‘Comprehension and Context: Successful Communication and
Communicative Breakdown’. In Applied Sociolinguistic, edited by P. Trudgill, 7–32.
Acad. Press.
Mishler, E.G. 1984. The Discourse of Medicine: Dialectics of Medical Interviews. Greenwood
Publishing Group.
Mitchell, C.J. 1984. ‘Producing Data: Case Studies’. In Ethnographic Research: A Guide to
General Conduct., edited by R.F. Ellen, 237–41. New York: Academic Press.
Mondada, L. 2013. ‘Conversation Analysis and Institutional Interaction’. In The Encyclopedia
of Applied Linguistics, edited by C.A. Chapelle, 1005–11. Oxford: Wiley-Blackwell.
NHS. 2014. ‘Mastitis’. Website of the National Health Service, UK. NHS Choices. October 23.
http://www.nhs.uk/conditions/Mastitis/Pages/Introduction.aspx.
———. 2015. ‘Coughing up Blood’. NHS Choices. http://www.nhs.uk/conditions/coughing-
up-blood/Pages/Introduction.aspx.
O’Leary, S.C.B., S. Federico, and L.C. Hampers. 2003. ‘The Truth about Language Barriers:
One Residency Program’s Experience’. Pediatrics 111 (5): e569–e573.
References 190
Padilla Cruz, M. 2013. ‘Understanding and Overcoming Pragmatic Failure in Intercultural

Communication: From Focus on Speakers to Focus on Hearers’. International Review
of Applied Linguistics in Language Teaching 51 (1): 23–54.
Paltved, C., and P. Musaeus. 2012. ‘Qualitative Research on Emergency Medicine Physicians:
A Literature Review’. International Journal of Clinical Medicine 3 (7): 772–89.
doi:10.4236/ijcm.2012.37A136.
Paribakht, T. 1985. ‘Strategic Competence and Language Proficiency.’ Applied Linguistics 6:
132–46.
Patil, S., and P. Davies. 2014. ‘Use of Google Translate in Medical Communication: Evaluation
of Accuracy’. BMJ 349 (dec15 2): g7392–g7392. doi:10.1136/bmj.g7392.
Patton, M.Q. 1990. Qualitative Evaluation and Research Methods. 2nded. Newbury Park, CA:
Sage.
Peräkylä, A., and S. Vehvilfinen. 2003. ‘Conversation Analysis and the Professional Stocks of
Interactional Knowledge’. Discourse & Society 14 (6): 727–750.
Person, J., L.A. Spiva, and P. Hart. 2012. ‘The Culture of an Emergency Department: An
Ethnographic Study’. International Emergency Nursing, December.
doi:10.1016/j.ienj.2012.10.001.
Petrovic, M. 2012. ‘Belgium: A Country of Permanent Immigration’. Migration Information
Source: The Online Journal of the Migration Policy Institute.
Phelan, M. 2001. The Interpreter’s Resource7RSLFVLQ7UDQVODWLRQ&OHYHGRQௗ%XIIDOR
Multilingual Matters.
Phillips, E., and D.S. Pugh. 2010. +RZ7R*HW$3KGࣟ$+DQGERRNIRU6WXGHQWVDQG7KHLU
Supervisors. 5thed. Berkshire: McGraw-Hill Education.
Pilnick, A., and R. Dingwall. 2011. ‘On the Remarkable Persistence of Asymmetry in
Doctor/patient Interaction: A Critical Review’. Social Science & Medicine 72 (8):
1374–82. doi:10.1016/j.socscimed.2011.02.033.
Pinker, S. 1995. The Language Instinct: The New Science of Language and Mind. Vol. 7529.
Penguin UK.
———. 2007. The Language Instinct How the Mind Creates Language. New York:
HarperPerennial ModernClassics.
Plaja, A.O., L.M. Cohen, and J. Samora. 1968. ‘Communication between Physicians and
Patients in Outpatient Clinics: Social and Cultural Factors’. The Milbank Memorial
Fund Quarterly 46 (2): 161.
Pöchhacker, F., and M. Kadric. 1999. ‘The Hospital Cleaner as Healthcare Interpreter: A Case
Study’. The Translator 5 (2): 161–78. doi:10.1080/13556509.1999.10799039.
Pope, C., and N. Mays. 2006. Qualitative Research in Health Care. 3rded. Oxford: Blackwell
Publishing.
Poulisse, N. 1993. ‘A Theoretical Account of Lexical Communication Strategies’. In The
Bilingual Lexicon, edited by R. Schreuder and B. Weltens, 157–89. Amsterdam: John
Benjamins.
Poulisse, N., T. Bongaerts, and E. Kellerman. 1987. ‘The Use of Retrospective Verbal Reports
in the Analysis of Compensatory Strategies’. In Introspection in Second Language
Research, edited by C. Faerch and G. Kasper, 213–29. Clevedon, UK: Multilingual
Matters.
Priebe, S., S. Sandhu, S. Dias, A. Gaddini, T. Greacen, E. Ioannidis, U. Kluge, A. Krasnik, M.
Lamkaddem, and Vincent Lorant. 2011. ‘Good Practice in Health Care for Migrants:
Views and Experiences of Care Professionals in 16 European Countries’. BMC Public
Health 11 (1): 187.
Ragan, S.L., and R. Hopper. 1984. ‘Ways to Leave Your Lover: A Conversation Analysis of
Literature’. Communication Quarterly 32: 310–17.
References 191
Ramirez, D., K. G. Engel, and T. S. Tang. 2008. ‘Language Interpreter Utilization in the
Emergency Department Setting: A Clinical Review’. Journal of Health Care for the
Poor and Underserved 19 (2): 352–62.
Rampton, B. 2005. Crossing: Language & Ethnicity among Adolescents. 2nd ed. Encounters.
0DQFKHVWHU8.1RUWKDPSWRQ0$ௗ6W-HURPH3XE
———. 2006. Language in Late Modernity Interaction in an Urban School. Cambridge; New
York: Cambridge University.
———. 2007. ‘Neo-Hymesian Linguistic Ethnography in the United Kingdom’. Journal of
Sociolinguistics 11 (5): 584–607. doi:10.1111/j.1467-9841.2007.00341.x.
———. 2017. ‘Interactional Sociolinguistics’. Working Papers in Urban Language &
Literacies, no. 205.
Rampton, B., J. Maybin, and C. Roberts. 2014. ‘Methodological Foundations in Linguistic
Ethnography’. Working Papers in Urban Language & Literacies.
Rampton, B., J Maybin, and C. Roberts. 2015. ‘Theory and Method in Linguistic
Ethnographhy’. In Linguistic Ethnography: Interdisciplinary Explorations, edited by J.
Snell, S. Shaw, and F. Copland, 1st ed. 2015 edition, 14–50. Houndmills, Basingstoke,
+DPSVKLUHௗ1HZ<RUN1<3DOJUDYH0DFPLOODQ
Raymond, G. 2008. ‘Ethnomethodology’. In The International Encyclopedia of
Communication, edited by Wolfgang Donsbach, 1600–1604. Malden, MA: Blackwell
Pub.
Redding, W.C. 1972. Communication within the Organization. New York: Industrial
Communication Council.
Reddy, M. 1979. ‘“The Conduit Metaphor: A Case of Frame Conflict in Our Language about
Language”.’ In Metaphor and Thought, 284–324. Cambridge University Press.
Rhodes, K.V., T. Vieth, T. He, A. Miller, D.S. Howes, O. Bailey, J. Walter, R. Frankel, and
W. Levinson. 2004. ‘Resuscitating the Physician-Patient Relationship: Emergency
Department Communication in an Academic Medical Center’. Annals of Emergency
Medicine 44 (3): 262–67. doi:10.1016/j.annemergmed.2004.02.035.
Richard, C., and M.T. Lussier. 2014. ‘The Art of Medical Information Exchange’. In The
Oxford Handbook of Health Communication, Behavior Change, and Treatment
Adherence, edited by L.R. Martin and M.R. DiMatteo. Oxford Library of Psychology.
2[IRUGௗ1HZ<RUN2[IRUG8QLYHUVLW\3UHVV
———. 2016a. ‘23. La Littératie En Santé’. In La Communication Professionnelle En Santé,
edited by M.T. Lussier and C. Richard. 2e Édition 2016. Montréal: Pearson.
———. 2016b. ‘Une Approche Dialogique de La Consultation’. In La Communication
Professionnelle En Santé, edited by M.T. Lussier and C. Richard. 2e Édition 2016.
Montréal: Pearson.
Richards, V. 1978. &DQFHU7KH:D\ZDUG&HOOࣟ,WV2ULJLQV1DWXUHDQG7UHDWPHQW. University
of California Press.
Rimondini, M., M.A. Mazzi, M. Deveugele, and J.M. Bensing. 2015. ‘How Do National
Cultures Influence Lay People’s Preferences toward Doctors’ Style of Communication?
A Comparison of 35 Focus Groups from an European Cross National Research’. BMC
Public Health 15 (1). doi:10.1186/s12889-015-2559-7.
Rivadeneyra, R., V. Elderkin-Thompson, R.C. Silver, and H. Waitzkin. 2000. ‘Patient
Centeredness in Medical Encounters Requiring an Interpreter’. The American Journal
of Medicine 108 (6): 470–474.
Roberts, C. 1996. ‘A Social Perspective on Understanding: Some Issues of Theory and
Method’. In Achieving Understanding: Discourse in Intercultural Encounters, edited
by K. Bremer, C. Roberts, M. Vasseur, M. Simonot, and P. Broeder, 9–36. London and
New York: Longman.
References 192
———. 2006. ‘Doctors and Patients in Multilingual Settings’. In The Encyclopedia of

Language & Linguistics, edited by E. K. Brown aQG$$QGHUVRQQGHG$PVWHUGDPௗ
Boston: Elsevier.
———. 2010. ‘Communicating Decisions with a Multilingual Patient Population’. Language
Barriers in Clinical Settings Barrières Linguistiques En Contexte Médical, Les cahiers
de l’ILSL, , no. 28: 9–25.
———. 2015. ‘Part Three - “Steps in Doing a Transcription”’. King’s College - Qualitative
Research Methods.
www.kcl.ac.uk/sspp/departments/education/research/ldc/knowledge-
transfer/DATA/qualitative.aspx.
Roberts, C., B. Moss, V. Wass, S. Sarangi, and R. Jones. 2005. ‘Misunderstandings: A
Qualitative Study of Primary Care Consultations in Multilingual Settings, and
Educational Implications’. Medical Education 39 (5): 465–475.
Roberts, C., and S. Sarangi. 2005. ‘Theme-Oriented Discourse Analysis of Medical
Encounters’. Medical Education 39 (6): 632–640.
Roberts, C., S. Sarangi, and B. Moss. 2004. ‘Presentation of Self and Symptoms in Primary
Care Consultations Involving Patients from Non-English Speaking Backgrounds’.
Communication & Medicine 1 (2): 159–169.
Robinson, J.D. 2003. ‘An Interactional Structure of Medical Activities during Acute Visits and
Its Implications for Patients’ Participation’. Health Communication 15 (1): 27–59.
———. 2008. ‘Doctor-Patient Talk’. In The International Encyclopedia of Communication,
edited by W. Donsbach, 1396–1400. Malden, MA: Blackwell Pub.
Robinson, J.D., and T. Stivers. 2001. ‘Achieving Activity Transitions in Physician-Patient
Encounters.’ Human Communication Research 27 (2): 253–98.
Rosenberg, E., Y. Leanza, and R. Seller. 2007. ‘Doctor–patient Communication in Primary
Care with an Interpreter: Physician Perceptions of Professional and Family
Interpreters’. Patient Education and Counseling 67 (3): 286–92.
Rosenberg, E., R. Seller, and Y. Leanza. 2008. ‘Through Interpreters’ Eyes: Comparing Roles
of Professional and Family Interpreters’. Patient Education and Counseling 70 (1): 87–
93. doi:10.1016/j.pec.2007.09.015.
Rosenzweig, S. 1993. ‘Emergency Rapport’. Journal of Emergency Medicine 11 (6): 775–778.
doi:10.1016/0736-4679(93)90651-M.
Rosland, A.M., J.D. Piette, H. Choi, and M. Heisler. 2011. ‘Family and Friend Participation in
Primary Care Visits of Patients with Diabetes or Heart Failure: Patient and Physician
Determinants and Experiences’. Medical Care 49 (1): 37–45.
doi:10.1097/MLR.0b013e3181f37d28.
Roter, D.L. 1977. ‘Patient Participation in the Patient-Provider Interaction: The Effects of
Patient Question Asking on the Quality of Interaction, Satisfaction and Compliance’.
Health Education & Behavior 5 (4): 281–315.
Roth, J. A. 1963. ‘Information and the Control of Treatment in Tuberculosis Hospitals’. In The
Hospital in Modern Society, by Eliot Freidson, 293–317. Free Press of Glencoe.
Roy, C. 1999. Interpreting As a Discourse Process. Oxford University Press.
Rymes, B. 2014. ‘Communicative Repertoire’. In The Routledge Handbook of English Studies,
edited by C. Leung and B.V. Street, 281–/RQGRQௗ1HZ<RUN5RXWOHGJH7D\ORU
& Francis Group.
Salmon, P., and B. Young. 2011. ‘Creativity in Clinical Communication: From
Communication Skills to Skilled Communication: Communication Skills and Skilled
Communication’. Medical Education 45 (3): 217–26. doi:10.1111/j.1365-
2923.2010.03801.x.
References 193
Samora, J., L. Saunders, and R. Larson. 1961. ‘Medical Vocabulary Knowledge among
Hospital Patients’. Journal of Health and Human Behavior 2: 83–89.
Sarangi, S. 2010. ‘Healthcare Interaction as an Expert Communicative System: An Activity
Analysis Perspective’. In New Adventures in Language and Interaction, edited by J.
Streeck, 167–97. Pragmatics & beyond New Series, v. 196. Amsterdam/Philadelphia:
John Benjamins Pub. Co.
———. 2015. ‘The Communicative and Ethical Ecology of Interpreter-Mediated Healthcare
Encounters’. In . Ghent.
Sarver, J., and D.W. Baker. 2000. ‘Effect of Language Barriers on Follow-up Appointments
after an Emergency Department Visit’. Journal of General Internal Medicine 15 (4):
256–264.
Saville-Troike, M. 2003. The Ethnography of Communication: An Introduction. 3rd ed.
Language in Society 3. Malden, Mass: Blackwell Pub.
Scheeres, H., D. Slade, M. Manidis, J. McGregor, and C. Matthiessen. 2008. ‘Communicating
in Hospital Emergency Departments’. Prospect Journal, AMEP Research Centre,
Macquarie University, Sydney, Australia, 23 (2): 13–22.
Schegloff, E.A. 1977. ‘The Preference for Self-Correction in the Organization of Repair in
Conversation.’ Language 53 (2): 361–82.
———. 1987. ‘Some Sources of Misunderstanding in Talk-in-Interaction’. Linguistics 25 (1).
———. 1992. ‘Repair after next Turn: The Last Structurally Provided Place for the Defence
of Intersubjectivity in Conversation.’ American Journal of Sociology 95 (5): 1295–
1345.
———. 2000. ‘When “Others” Initiate Repair’. Applied Linguistics 21 (2): 205–43.
Schenker, Y., A. Fernandez, R. Sudore, and D. Schillinger. 2010. ‘Interventions to Improve
Patient Comprehension in Informed Consent for Medical and Surgical Procedures: A
Systematic Review’. Medical Decision Making 31 (1): 151–73.
Schenker, Y., B. Lo, K.M. Ettinger, and A. Fernandez. 2008. ‘Navigating Language Barriers
under Difficult Circumstances’. Annals of Internal Medicine 149 (4): 264–269.
Schenker, Y., F. Wang, S.J. Selig, R. Ng, and A. Fernandez. 2007. ‘The Impact of Language
Barriers on Documentation of Informed Consent at a Hospital with On-Site Interpreter
Services’. Journal of General Internal Medicine 22 (S2): 294–99.
Schillinger, D., and A.H. Chen. 2004. ‘Literacy and Language Disentangling Measures of
Access, Utilization, and Quality’. Journal of General Internal Medicine 19 (3): 288–
290.
Schwei, R.J., S. Del Pozo, N. Agger-Gupta, W. Alvarado-Little, A. Bagchi, A.H. Chen, L.
Diamond, F. Gany, D. Wong, and E.A. Jacobs. 2015. ‘Changes in Research on
Language Barriers in Health Care Since 2003: A Cross-Sectional Review Study’.
International Journal of Nursing Studies, March. doi:10.1016/j.ijnurstu.2015.03.001.
Schwimmer, B. 1996. ‘Objectivity in Ethnography’. University of Manitoba.
www.umanitoba.ca/faculties/arts/anthropology/courses/122/module1/objectivity.html.
Searle, J.R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge University
Press.
Seife, C. 2007. Decoding the Universe: How the New Science of Information Is Explaining
Everything in the Cosmos, from Our Brains to Black Holes. Penguin Books.
Shands, H.C., J.E. Finesinger, S. Cobb, and R.D. Abrams. 1951. ‘Psychological Mechanisms
in Patients with Cancer.’ Cancer 4: 1159–70.
Shaw, S., F. Copland, and J. Snell. 2015. ‘An Introduction to Linguistic Ethnography:
Interdisciplinary Exploration’. In Linguistic Ethnography: Interdisciplinary
Explorations, edited by J. Snell, S. Shaw, and F. Copland, 1sted. Houndmills,
Basingstoke, Hampshire; New York, NY: Palgrave Macmillan.
References 194
Shepherd, H.L., M.H.N. Tattersall, and P.N. Butow. 2008. ‘Physician-Identified Factors
Affecting Patient Participation in Reaching Treatment Decisions’. Journal of Clinical
Oncology: Official Journal of the American Society of Clinical Oncology 26 (10):
1724–31. doi:10.1200/JCO.2007.13.5566.
Shotter, J. 2016. ‘Wittgenstein and Language Games in Qualitative Research’. Edited by C.A.
Chapelle. The Encyclopedia of Applied Linguistics. Accessed March 16.
Silverman, D., and B. Torode. 1980. The Material Word: Some Theories of Language and Its
Limitations. London: Routledge & Kegan Paul.
Slade, D., E. Chandler, J. Pun, M. Lam, C. Matthiessen, G. Williams, E Espindola, et al. 2015.
‘Effective Healthcare Worker-Patient Communication in Hong Kong Accident and
Emergency Departments’. Hong Kong Journal of Emergency Medicine, no. 22(2): 69–
83.
Slade, D., M. Manidis, J. McGregor, H. Scheeres, E. Chandler, J. Stein-Parbury, R. Dunston,
and M. Herke. 2015. Communicating in Hospital Emergency Departments. Berlin,
Heidelberg: Springer Berlin Heidelberg. http://link.springer.com/10.1007/978-3-662-
46021-4_2.
Slade, D., and S. McCarthy. 2011. ‘Doctor-Patient Communication in Emergency
Departments: Contextual Complexity and Competing Discourses’. EMA - Emergency
Medicine Australasia 23: 3–4.
Slade, D., H. Scheeres, M. Manidis, R. Iedema, R. Dunston, J. Stein-Parbury, C. Matthiessen,
M. Herke, and J. McGregor. 2008. ‘Emergency Communication: The Discursive
Challenges Facing Emergency Clinicians and Patients in Hospital Emergency
Departments’. Discourse & Communication 2 (3): 271–98.
Slembrouck, S. 2017. ‘What Is Meant by Discourse Analysis? Online Course Material.’ UGent.
http://www.english.ugent.be/da.
Sontag, S. 1979. Illness as Metaphor. Vintage Books.
Stenton, C. 2008. ‘The MRC Breathlessness Scale’. Occupational Medicine 58 (3): 226–27.
Stevenson, F.A., C.A. Barry, N. Britten, N. Barber, and C.P. Bradley. 2000. ‘Doctor–patient
Communication about Drugs: The Evidence for Shared Decision Making’. Social
Science & Medicine 50 (6): 829–40. doi:10.1016/S0277-9536(99)00376-7.
Stivers, T. 2002. ‘Presenting the Problem in Pediatric Encounters: “Symptoms Only” Versus
“Candidate Diagnosis” Presentations’. Health Communication 14 (3): 299–338.
doi:10.1207/S15327027HC1403_2.
———. 2006a. ‘The Interactional Process of Reaching a Treatment Decision in Acute Medical
Encounters.’ In Communication in Medical Care: Interactions between Primary
Care Physicians and Patients, by J. Heritage and D.W. Maynard, 279–312. Cambridge:
———. 2006b. ‘Treatment Decisions: Negotiations between Doctors and Patients in Acute
Care Encounters’. In Communication in Medical Care, 279–312. Studies in
Interactional Sociolinguistics. Cambridge University Press.
Stivers, T., N. J. Enfield, P. Brown, C. Englert, M. Hayashi, T. Heinemann, G. Hoymann, et
al. 2009. ‘Universals and Cultural Variation in Turn-Taking in Conversation’.
Proceedings of the National Academy of Sciences 106 (26): 10587–92.
doi:10.1073/pnas.0903616106.
Street, R.L. 1991. ‘Accommodation in Medical Consultations’. In Contexts of Accommodation:
Developments in Applied Sociolinguistics, edited by H. Giles, J. Coupland, and N.
Coupland, 131–56. Studies in Emotion and SoFLDO,QWHUDFWLRQ&DPEULGJH>(QJODQG@ௗ
1HZ<RUNௗ3DULV&DPEULGJH8QLYHUVLW\3UHVVௗ(GLWLRQVGHOD0DLVRQGHVVFLHQFHGH
l’homme.
References 195
Stubbe, M. 2010. ‘“Was That My Misunderstanding?”: Managing Miscommunication and

Problematic Talk at Work’. Victoria University of Wellington.
http://researcharchive.vuw.ac.nz/handle/10063/1462.
Swinglehurst, D. 2014. ‘Displays of Authority in the Clinical Consultation: A Linguistic
Ethnographic Study of the Electronic Patient Record’. Social Science & Medicine 118
(October): 17–26.
Swinglehurst, D., C. Roberts, and T. Greenhalgh. 2011. ‘Opening up the “Black Box” of the
Electronic Patient Record: A Linguistic Ethnographic Study in General Practice’.
Communication & Medicine 8 (1): 3–15.
Tajfel, H., and J.C. Turner. 1979. ‘An Integrative Theory of Intergroup Conflict’. In The Social
Psychology of Intergroup Relations, edited by W.G. Austin and S. Worchel,
Brooks/Cole, 33–47. Monterey, CA.
Tannen, D. 2009. ‘Framing and Face: The Relevance of The Presentation of Self to Linguistic
Discourse Analysis’. Social Psychology Quarterly 72 (4): 300–305.
———. 2011. That’s Not What I Meant!: How Conversational Style Makes or Breaks
Relationships. HarperCollins.
Tannen, D., and C. Wallat. 1987. ‘Interactive Frames and Knowledge Schemas in Interaction:
Examples from a Medical Examination/Interview’. Social Psychology Quarterly 50 (2):
205.
Tarone, E. 1977. ‘Conscious Communication Strategies in Interlanguage: A Progress Report’.
In On TE-SOL ’77, edited by H.D. Brown, C.A. Yorio, and R.C. Crymes, 194–203.
Washington: TESOL.
Thomas, J. 1983. ‘Cross-Cultural Pragmatic Failure’. Applied Linguistics 4: 91–109.
Tracy, K. 2015. ‘Editor’s Introduction’. In The International Encyclopedia of Language and
Social Interaction, edited by K. Tracy, C. Ilie, and T. Sandel, Volume One A-D:xxvii–
xlviii. Wiley Blackwell.
Turabián, J. L., and B. Pérez Franco. 2015. ‘El Acompañante En La Entrevista Clínica de
Atención Primaria’. SEMERGEN - Medicina de Familia 41 (4): 206–13.
doi:10.1016/j.semerg.2014.05.012.
Tzanne, A. 2000. Talking at Cross-Purposes: The Dynamics of Miscommunication.
Amsterdam: John Benjamins.
Valero-Garcés, C. 2002. ‘Interaction and Conversational Constrictions in the Relationships
between Suppliers of Services and Immigrant Users’. Pragmatics 12 (4): 469–495.
———. 2006. Formas de Mediación Intercultural E Interpretación En Los Servicios Públicos.
Conceptos, Datos, Situaciones Y Práctica. Granada: Comares.
Van de Poel, K., E. Vanagt, U. Schrimpf, and J. Gasiorek. 2013. Communication Skills for
Foreign and Mobile Medical Professionals. Berlin Heidelberg: Springer.
Van Spall, H.G.C., A. Toren, A. Kiss, and R.A. Fowler. 2007. ‘Eligibility Criteria of
Randomized Controlled Trials Published in High-Impact General Medical Journals’.
JAMA: The Journal of the American Medical Association 297 (11): 1233–1240.
Vandermotten, C., P. Marissal, G. Van Hamme, C. Kesteloot, K. Slegers, L. Vanden Broucke,
B. Ippersiel, S. de Bethune, and R. Naiken. 2011. ‘Analyse Dynamique Des Quartiers
En Difficulté Dans Les Régions Urbaines Belges’. PPS Social Integration, anti-Poverty
Policy, Social Economy and Federal Urban Policy. Brussels: Politique des Grandes
Villes, Intégration Sociale. http://mi-is.be/sites/default/files/doc/atlasfr_0.pdf.
Verrept, H. 2000. ‘Interculturele Bemiddeling in de Ziekenhuizen’. Eindrapport Brussel:
Ministerie van Sociale Zaken, Volksgezondheid En Leefmilieu (Coördinatiecel
Interculturele Bemiddeling) 182: 12.
References 196
VLC. 2017. ‘The Virtual Linguistics Campus - Linguistic Courses and Course Material -
Developed by the Linguistic Engineering Team at Marburg University.’
http://linguistics.online.uni-marburg.de/.
Wacquant, L. 2002. ‘Scrutinizing the Street: Poverty, Morality, and the Pitfalls of Urban
Ethnography 1’. American Journal of Sociology 107 (6): 1468–1532.
Wadensjö, C. 1992. ‘Interpreting as Interaction: On Dialogue-Interpreting in Immigration
Hearings and Medical Encounters’. Doctoral dissertation. Linköping University,
Department of Communications Studies, Faculty of Arts and Sciences.
———. 1998. Interpreting As Interaction. Oxon/New York: Routledge.
Wagley, L.K., and S.E. Newton. 2010. ‘Emergency Nurses’ Use of Psychosocial Nursing
Interventions for Management of ED Patient Fear and Anxiety’. Journal of Emergency
Nursing 36 (5): 415–19. doi:10.1016/j.jen.2009.07.022.
Wallbrecht, J., L. Hodes-Villamar, S.J. Weiss, and A.A. Ernst. 2014. ‘No Difference in
Emergency Department Length of Stay for Patients with Limited Proficiency in
English’. Southern Medical Journal 107 (1): 1–5.
doi:10.1097/SMJ.0000000000000037.
Watt, I.S. 2008. ‘Consultation, Medical History and Record Taking’. In Clinical Examination,
edited by O. Epstein, G. D. Perkin, J. Cookson, I.S. Watt, R. Rakhit, A. Robins, and
G.A.W. Hornet, Fourth Edition, 1–19. London: Mosby Elsevier.
Waxman, M.A., and M.A. Levitt. 2000. ‘Are Diagnostic Testing and Admission Rates Higher
in Non-English-Speaking versus English-Speaking Patients in the Emergency
Department?’ Annals of Emergency Medicine 36 (5): 456.
West, C., and R.M. Frankel. 1991. ‘Miscommunication in Medicine’. In ‘Miscommunication’
and Problematic Talk, edited by N. Coupland, H. Giles, and J.M. Wiemann, 166–164.
Newbury Park: Sage Publications.
Wierzbicka, A. 2006. ‘Intercultural Pragmatics’. In The Encyclopedia of Language &
Linguistics, edited by E. K. Brown and A. Anderson, 2nd ed, 735–$PVWHUGDPௗ
Boston: Elsevier.
Wikivisual. 2016. ‘How to Prevent Kidney Stones from Recurring’. wikiHow. Accessed
August 19. http://www.wikihow.com/Prevent-Kidney-Stones-from-Recurring.
Willems, G. 1987. ‘Communication Strategies and Their Significance in Foreign Language
Teaching.’ System 15: 351–64.
Wilson, W. J., and A. Chaddha. 2009. ‘The Role of Theory in Ethnographic Research’.
Ethnography 10 (4): 549–64. doi:10.1177/1466138109347009.
Wisten, A., and K. Zingmark. 2007. ‘Supportive Needs of Parents Confronted with Sudden
Cardiac death—A Qualitative Study’. Resuscitation 74 (1): 68–74.
Wittgenstein, L. 1953. Philosophical Investigations (Translated by Anscombe, G.E.M.).
Oxford: Basil Blackwell.
WMA. 2013. ‘World Medical Association Declaration of Helsinki Ethical Principles for
Medical Research Involving Human Subjects’. World Medical Association.
Wolff, J.L., and D.L. Roter. 2011. ‘Family Presence in Routine Medical Visits: A Meta-
Analytical Review’. Social Science & Medicine (1982) 72 (6): 823–31.
World Bank. 2016. ‘International Migrant Stock. Based on: United Nations Population
Division, Trends in Total Migrant Stock: 2012 Revision.’
http://data.worldbank.org/indicator/SM.POP.TOTL.
Yin, R.K. 1994. Case Study Research: Design and Methods. 2nded. Vol. 5. Applied Social
Research Method Series. Thousand Oaks, CA: Sage.
Zola, I.K. 1963. ‘Problems of Communication, Diagnosis and Patient Care: The Interplay of
Patient, Physician and Clinic Organization.’ Journal of Medical Education 38: 829–38.
References 197
Zoppi, K. A., and C.P. McKegney. 2002. ‘The Difficult Clinical Conversation’. In
Fundamentals of Clinical Practice, edited by Mark B Mengel, Warren Lee Holleman,
and Scott A Fields. New York, N.Y.: Kluwer Academic/Plenum Publishers.
Zuengler, J. 1991. ‘Accommodation in Native-Nonnative Interactions: Going beyond The
“what” to The “why” in Second-Language Research’. In Contexts of Accommodation,
edited by H. Giles, J. Coupland, and N. Coupland. Studies in Emotion and Social
Interaction. Cambridge University Press.
Zun, L.S., T. Sadoun, and L. Downey. 2006. ‘English-Language Competency of Self-Declared
English-Speaking Hispanic Patients Using Written Tests of Health Literacy.’ Journal
of the National Medical Association 98 (6): 912.
Annex 1 198
Annex 1. Theoretical foundations of Language and Social Interactions research
This Annex reviews, selectively, the main theoretical foundations of the Language and Social
Interactions literature, to which this PhD belongs. It reviews a set of seminal studies in different
disciplines which were brought together at some point in time under the umbrella of Language
and Social Interaction (LSI) research and complements Chapter 2 by providing additional
details on some of the subdisciplines in the domain of Language and Social Interaction (LSI)
research that have been influential for methodological approach developed in this PhD.
Language and Interaction (LSI) research sets out from the idea that language, rather than a
mere medium of communication, is an integral part of a social activity. In the late 1980s,
Sanders and Sigman, two language scholars, proposed the term “Language and Social
Interaction” to bring together different research disciplines which until then had been
developing relatively independently from each other, most importantly conversation analysis,
ethnography of communication, discourse analysis, interactional sociolinguistics, language and
social psychology, and pragmatics (Tracy 2015).
In what follows, the basic concepts and founding theories of these disciplines (notably
pragmatics, conversation analysis, ethnography of communication, interactional
sociolinguistics, and social psychology) will be discussed, focusing on those aspects that are
more relevant to the analysis envisaged in this dissertation.
Pragmatics
Philosopher Wittgenstein already argued in the 1950s that the meaning of words should not be
restricted to the objects they refer to or to the mental associations they can bring about. Instead,
he advocated the idea that the meaning of words can only be inferred from their context, more
specifically, from the activities during which they are produced (Wittgenstein 1953). To
Wittgenstein, language was more than just a “conduit” to convey a message from a sender to a
receiver (Reddy 1979); and he considered that narrow definitions of language obscured the
simplicity of it.
Wittgenstein introduced the concept of a “language game” to explain that every utterance is
part of a social activity, and to understand its meaning, it is more insightful to look at how a
particular word is used in a particular activity, than to use an explanatory definition or a
generalisation (Wittgenstein 1953; Biletzki and Matar 2014). Hence, Wittgenstein emphasised
that words are produced within a specific system of communication or action, in which their
meaning may be different from the usual meaning of the same words in a particular language
in the wider context. His use of the term “language game” stems from the observation that
events occurring within the context of a game usually have quite different meanings from
similar events occurring outside of that game (Shotter 2016). The same applies to language
used in different activities.
Annex 1 199
For example, if a gay and lesbian rights activist performs the utterance “I have a dream” at a
rally for equal rights for gays and lesbians, he very likely means that his dream is to end
discrimination against gays and lesbians. However, if an African American actor who plays
the role of Martin Luther King in a film uses those very same utterances, he is more likely to
refer to the idea of ending racial discrimination. If a court attendant says “All rise” as the judge
walks into the courtroom on a Monday morning, all people present will likely get up. However,
uttering the same words on a Saturday evening in his local bar after having had a few beers, is
unlikely to make many get up from their chairs. Finally, also the utterances “Show your tongue”
or “Lay down on your back” have a different meaning inside a doctor’s cabinet than outside of
it.
Wittgenstein did not provide an explicit definition of the concept “language game”. His main
objective was to promote a more fluid, diversified, and more activity-oriented perspective on
language (Biletzki and Matar 2014). His idea of connecting language to the social situation it
creates and which houses it was a major source of inspiration for later scholars such as Austin
(1962) and Searle (1969), who laid the foundations of the “speech act theory”. Their theory
develops the idea that speech acts are communicative acts which are produced by a speaker
with a certain intention, such as expressing a promise, a regret, or a command, asking a
question, or making a declaration. Hence, whereas earlier philosophers mainly considered
language as an instrument to describe the reality (either correctly or falsely), Austin and Searle
advocated the idea that words can be used for a variety of other purposes as well.
More specifically, they argued a speech act is made up of locutionary, illocutionary and
perlocutionary content. The locutionary act refers to the literal meaning of a speech act. The
illocutionary act refers to the pragmatic force of the utterance, in other words, what the speaker
intends by the speech act. For instance, how does he want the hearer to act in response to his
speech act? Finally, the speech act may as well have a perlocutionary effect, which refers to
the impact the speech act has on the hearer (such as scaring, inspiring, upsetting, …) whether
or not it is intended.
The speech act theory has brought new insights in how people communicate. However, it can
be difficult to apply this theory to the description of actual conversations, as it is not always
straightforward “to assign speech acts in a non-arbitrary way” (Levinson 1980, 20). Indeed,
sometimes the meaning of a whole set of sentences can be the same as the meaning of one
speech act. For instance, the speech act “I apologise for being late” has the same illocutionary
force as saying, “it was not my intention to be late, but I locked myself out of my house and
missed my train”. While the second utterance does not have the format proposed by the speech
act theory, it could still be considered as a speech act if the wider context of activity in which
communication takes place is considered, in a similar way as what had been proposed earlier
by Wittgenstein.
Another important theoretical contribution was made by Grice (1975, 1991), who describes
conversations as cooperative processes, in which participants cooperate to achieve mutual
Annex 1 200
conversational objectives and bring meaning across. 33 In his view, “normal” behaviour in a
conversation obeys the “Cooperative Principle”, which is formulated as follows: “Make your
contribution such as it is required, at the stage at which it occurs, by the accepted purpose or
direction of the talk exchange in which you are engaged” (Grice 1975, 46).
The implications of the Cooperative Principles are described in more detail in Grice’s four
“Maxims of Communication”, notably the Maxim of Quantity, the Maxim of Quality, the
Maxim of Relation, and the Maxim of Manner. The Maxim of Quantity refers to the quantity
of information provided and prescribes that a message should provide as much information as
required, but not more than that. The Maxim of Quality refers to the quality of the information
provided and recommends that a message should be truthful, implying that participants to a
conversation do not say things they do not know to be true, nor things they know or believe to
be false. The Maxim of Relation suggests to only provide information that is relevant. Finally,
the Maxim of Manner refers to the style of communication, notably that it be clear,
unambiguous, brief, and orderly.
However, these Maxims are broken very often. If a Maxim is broken covertly, so that other
participants are not aware, one speaks of a “violation” of that Maxim. If a Maxim is broken
openly, one speaks of a “flout” of that Maxim. For example, if a doctor asks a patient’s
companion to a medical visit: “Are you the patient’s husband?” obeying the Cooperative
Principle as well as the four maxims would imply that the companion answers “Yes”, if he is
indeed the patient’s husband. Answering “Yes” while he is actually not the patient’s husband
would imply a violation of the Maxim of Quality. If the companion is the patient’s husband,
but he replies “Yes, we got married 4 years ago, in Spain, and we had a great honeymoon,” or
if he replies “That is none of your business,” he flouts the Maxim of Quantity for giving too
much or too little information respectively. Replying by saying “I got married in Las Vegas”
would mean a flout of the Maxim of Relation, as this would not answer the doctor’s question.
Finally, replying “Well, it depends on what you consider husband. We have been living
together for a while but I would not consider her my wife, you know…” would imply flouting
the Maxim of Manner.
In Grice’s view, if a Maxim is openly opted out of or flouted by a speaker, this usually means
that the speaker intends to convey additional meaning over and above what is literally said. He
refers to such additional meaning as “an implicature”. For instance, in the last example, where
the companion says “I would not consider [the patient] my wife”, this could imply that the
companion wants to take distance from the patient, to express that his commitment and
assumed responsibility towards her are weaker than what would be the case if they were
married. Maxims can also be flouted by using understatements, overstatements, or metaphors.
Understatements can be used, for instance, to imply modesty, humour, or politeness.
33
Even if participants are not socially cooperative, for example if they are arguing, they can be conversationally
cooperative to bring about the verbal argument (Kecskés 2014, 27).
Annex 1 201
This insight inspired Brown and Levinson (1987) to develop their Politeness Theory, in which
they describe how Grice’s maxims sometimes need to be flouted in order to be polite, in other
words to save another participant’s face. 34 For instance, a doctor may say to a patient’s
companion “I am going to physically examine the patient”, implying that he would like the
companion to leave the consultation room. Following Grice’s maxims, he could communicate
more clearly by adding the question “Can you leave the room, please?”. However, by being
less unambiguous, the doctor leaves space to the companion to independently take the decision
to leave the room, which is less face-threatening than the direct question would have been.
Another possible example concerns the use of politeness expressions which may not be relevant
for the exchange of information but may serve to reconfirm the hierarchical relation between
participants to an interaction (e.g. if a doctor who is treating the Prime Minister, starts all his
utterances with the words “Your Excellency”.)
Together with Grice (1975)’s work on “implicature”, simply put, the implicitly suggested
meaning of an utterance, Austin (1962) and Searle (1969)’s work laid the foundations of the
discourse analytical field of pragmatics (Jones 2013). In the field of pragmatics, the meaning
of an utterance cannot be analysed without acknowledging the intention of the speaker who
produced it, and the effect it has on the receiver.
Levinson (1992) revisits Wittgenstein’s idea that in order to understand what utterances mean,
one has to know the type of activity in which these utterances play a role. While the majority
of linguists and philosophers did not agree with Wittgenstein’s rather radical idea that the
meaning of language was entirely linked to its usage; Levinson (1992) further refined
Wittgenstein’s ideas by introducing the notion of “activity types”. Levinson‘s activity types
are instances of communication that are goal-defined, involving people, with constraints
regarding their setting and their participants, as well as constraints on what can be done or said
within the context of that activity type (Levinson 1992, 67). An example of an activity type is
a consultation in the ED; another one is an informal conversation in a bar. As the participants
and the setting of both activities tend to be different, also the type of talk (e.g. the degree of
informality) will be different. Patients seeing a doctor may prefer him not to tell them the same
jokes as he would to his friends in a bar. Levinson further sees activity types as a structure
consisting of subparts or episodes. For example, if one considers as an activity type an ED
consultation, the different medical interview stages (such as opening, history taking, physical
examination, and diagnosis) can be considered as episodes. Alternatively, an activity can also
be defined by an overarching goal, such as for example finding out whether a patient is having
a heart attack or not.
Levinson (1992, 72) also considers that each activity type is associated with a set of inferential
schemas, which participants to an activity use to interpret what is said. These inferential
schemas are equivalent to a set of shared background knowledge on what the activity is about
and what would be a common way to respond to what is said or done in the context of that
34
The notion of “face” was initially coined by Goffman (1967), who described it as our “socially situated identity”
which is constantly being constructed, challenged, enhanced or negotiated through social interaction.
Annex 1 202
activity. However, when the activity type is not clear or if there is uncertainty surrounding the
relevant inferential schemas, communication can be challenged and prone to
miscommunication.
Levinson’s concept of inferential schematas is related to the ideas put forward by Goffman
(1974)’s framing theory, which argues that individuals respond to events by referring back to
one or more “primary frameworks” or schemas of interpretation. These frameworks are ways
in which individuals organize their experiences, and they allow individuals to locate, perceive,
identify and label concrete events experiences in their daily life, mostly unconsciously. They
define the principles that are followed when one decides how to behave and communicate in
particular contexts.
However, the complexities of communication that arise from illocutionary forces, implicature,
and inferential schemas, also imply that specific knowledge and effort is required from the
hearer to correctly interpret the speaker’s intention. When the hearer does not have the
necessary knowledge to understand what is being implied; or if the speaker does not clearly
formulate his utterance in a way that his implication can be understood, miscommunication can
arise. For instance, in a medical visit, if a doctor asks a patient at the end of the consultation:
“Any further questions?”, he may actually mean, “Are there any other concerns or things you
would like to talk about?”. If this is, however, not correctly understood by the patient, the latter
may answer “No”, while still having some unaddressed concerns. Such issues can be
particularly problematic in linguistically and culturally discordant communication.
Intercultural pragmatics
Although Grice's Cooperation Principle and his Maxims can be useful to analyse
communication in different contexts, they have not gone without criticism (Lindblom 2006).
A number of authors have questioned his theory, in particular, on the one hand, the idea that
people are always cooperative to communicate, and on the other hand, the universality of
speech styles (Tannen 2011; Wierzbicka 2006).
Theories by early pragmaticians, including Grice, were predominantly based on monolingual

Anglo-Saxon contexts. This means that their applicability could be problematic in language
discordant interactions where participants do not have sufficient skills in a shared language for
smooth mutual understanding. In such interactions, communication can be difficult even if
participants do all the best they can to cooperate and obey Grice’s maxims. On the other hand,
participants to a language discordant interaction may attribute miscommunication, triggered by
a violation or a flout of Grice’s maxims, wrongly to a language barrier, which then again can
give rise to miscommunication. As researchers started to study Grice’s conversational maxims
in linguistically more diverse settings in the 1990s, they laid the foundations for the new sub
discipline of intercultural pragmatics, which came into existence in the 2000s (Kecskés 2014).
Intercultural pragmatics studies the way language is used between human beings who have
different first languages, communicate in a common language, and usually represent different
Annex 1 203
cultures (Kecskés 2004, 2010). Its aim is not to describe communication by highlighting or
categorising cultural differences between interactants; it rather proposes to study in detail how
meaning is made in communicative events, through a sociocognitive approach (Blommaert
1998; Kecskés 2014). The language discordant situations under consideration can range from
situations in which one of the participants in the interaction is not speaking in his/her mother
tongue, to situations where none of the participants speaks in his/her mother tongue and
participants thus use a lingua franca, situations in which multiple languages are used, or
situations in which people are learning a language (Kecskés 2004).
Conversation analysis
Goffman was one of the pioneers of the research field of microsociology. He did not believe
that macro-level social factors such as religion, family, social position, or wealth were the
dominant determinants of individuals’ attitudes, beliefs, or values, said in a different way, of
their social reality. Instead, he promoted the idea that individuals build up their own social
reality through daily interaction with other individuals. In his essay “The Interaction Order”,
Goffman (1983) argues that social interaction is made up of individuals meeting each other
“face to face” (or over the phone, over mail), and that in these interactions, macro-level
sociological distinctions are not necessarily relevant: individuals are primarily meeting each
other.
In Goffman’s view, nearly everything we do or say, we do or say in the presence of others. In

other words, our doing and sayings are socially situated. The social situatedness of talk has an
impact on how speech is used in particular interactions, in other words, how we speak and what
we mean by what we say. To study social interactions in more detail, Goffman used concepts
and metaphors from theatre performances, such as framing, footing, and face, the front stage,
and the backstage. Goffman (1959) argued that in social interaction, individuals cannot
determine the role they take up in isolation from others. Instead, they are part of a team of
actors, by analogy with a cast of actors in a theatre setting. At the start of each interaction,
individuals negotiate their roles with other participants. This implies that an individual’s role
depends crucially on the context of the interaction as well as on the other participants to the
interaction.
Goffman (1959) also discovered that, like in theatre, individuals’ communicative behaviour
may change between the “front stage”, the public sphere, and the “backstage”, the private
sphere. For example, in the context of a medical ED, the “front stage” is made up of the
consultation room where doctors examine patients and address them in a formal way. The
backstage is the office where doctors fill out their files on the computer and meet other doctors
on an informal basis. Alternatively, in multilingual triadic ED encounters, the instances where
a patient or his companion address the doctor can be considered as “front stage”; while the
patient and his companion may have informal conversations in between, in a language the
doctor does not understand (“the backstage”).
Annex 1 204
Goffman (1981b) introduced the term “footing” to refer to our communicative behaviour and
the role we assume as a speaker. In his view, when we interact with someone, we will constantly
try to adapt our “footing” to keep the conversation going. A speaker can behave as an animator
(the “sounding box” through which utterances are produced), an author (the individual who
thought out and composed the utterances in the first place) or a principal (the individual or
party whose beliefs are represented by the words uttered, and who personally commits to the
promises or threats that are made). A single individual can assume each of these roles
simultaneously, but the different roles can as well be separated across different individuals or
entities (for example, a phone or television can act as an animator; a political party can act as
the principal).
A hearer can be ratified or unratified (Goffman 1981b). The ratified hearer is a hearer who is
fully entitled to listen to the speaker; contrary to the non-ratified hearer, who is an accidental
overhearer or bystander. Within a group of ratified hearers, a further distinction can be made
between a “primary addressee”, the person whom the speaker focuses his visual attention on
and to whom he expects to hand over the speaker role in the next turn, and the unaddressed
ratified hearers, who are entitled to listen, but are not directly addressed. The respective roles
taken up by the speaker and the hearer define the “participation framework” of the social
interaction, and participants rely on linguistics (e.g. code switching, pitch, volume stress and
tonal quality) as well as non-verbal behaviour (e.g. gestures, body orientation and touch) to try
to shift participation frameworks to a more preferred one (e.g. to re-attract attention from
potential hearers) (Goffman 1981b). In a medical interview, a doctor may for example shift
participation frameworks when he/she switches between talking to a foreign language speaking
patient and talking to his/her companion with whom the doctor shares a language.
Another famous scholar of social reality was sociologist and ethno-methodologist Garfinkel.
Like Goffman, he believed in bottom-up approaches to studying social reality, starting from
the idea that the society is built up from the micro-level of interaction between people
(Garfinkel 1994). More specifically, Garfinkel believed that social reality was made of a set of
“traffic” rules, of people who would or would not follow those rules, and of others’ reaction to
that behaviour. Garfinkel and his students were most famous for conducting so-called
“breaching” experiments (Raymond 2008). This implied that a researcher would breach
apparently common communicative patterns in a particular setting. For instance, in a medical
setting, a researcher could encourage a patient who enters the doctor’s office to ask the doctor,
“Hello, what can I help you with?”, rather than waiting for the doctor to initiate the
conversation; and then describe and discuss the doctor’s reaction.
Goffman’s and Garfinkel’s theories laid the foundations of the discipline of conversation
analysis. While Goffman’s focused on the interactional rituals as describe above, conversation
analysis promotes the idea that interaction should be studied at the pure interactional level via
transcriptions of audio recordings (Léon 2006). Conversation analysis sets out from the idea
that context is best understood at the level of the interaction. Hence, in contrast to Goffman’s
theories and those of interactional sociolinguistics or the ethnography of communication,
Annex 1 205
conversation analysis does not take into account the speaker’s mental states or background
knowledge if it cannot be traced back or locally inferred from the transcript itself (Beach 2006).
Conversation analysis discusses the turn-by-turn organisation of interactions.

Misunderstandings are also studied at the level of the turn taking: both the start and the repair
of misunderstandings are identified within the context of the interaction, through what
Gumperz calls “local inferencing”. Shegloff (1987, 203) identifies the turn which holds the
trigger for a potential misunderstanding, in other words the turn of which the contents was
apparently misunderstood by the hearer, as the “trouble source turn”. He finds that if this turn
is counted as the first turn, in many cases, repair comes about in the third (or the fourth) turn.
The reason conversation analysis’s primary locus of analysis is the immediate context of talk
is that it starts from the assumption that participants to a conversation most often refer to the
immediately preceding talk when constructing their own talk. Hence, conversation analysis
does not systematically attempt to understand communication speculating about people’s
mental states, or background knowledge in the interaction itself (Beach 2006), in contrast with
other methodological strands in the field of sociolinguistics and sociology.
Ethnography of communication
"It is a truism, but one frequently ignored in research, that how something
is said is part of what is said" (Hymes 1972, 59)
The ethnography of communication was founded by Hymes (1972). Its main asset lies in the
wide context scope it considers when studying processes of interpersonal communication
(Gordon 2011, 108). The main purpose of ethnography of communication is to unravel what a
speaker needs to know to communicate properly within a particular speech community
(Saville-Troike 2003, 2). The collection of ethnographic data occurs mostly through participant
observation (Saville-Troike 2003, 97).
According to ethnography of communication, rather than a “string of undifferentiated

discourse”, communication between individuals is structured into different communicative
events or sequences with clear boundaries (Saville-Troike 2003, 108). A good example of a
clearly delineated communicative event is a phone conversation, which starts by picking up the
phone and clearly ends when hanging up. In a similar way, a medical interview consists of
different sequences or communicative events (the opening, the problem presentation,...), each
with its specific purpose or function (Watt 2008; Bickley 2013). In the realm of interpreting
studies, analysts speak of “interpreted communicative events” (Angelelli 2000, 2004).
Hymes provides a taxonomy to contextualise spoken communication, which is well-known as

the “SPEAKING” framework (Hymes 1972, 65). Within this framework, the major contextual
components of communication are: the scene of interaction (S), the participants involved (P),
the ends (purposes or goals; E) of the interaction, the act sequences (A), the key (K) or tone,
the instrumentalities (or forms and channels of communication; I), the norms of interpretation
Annex 1 206
(N) and finally the genre (G) or the type of story which is told. A brief description of each of
these components is provided in Table A1.
Table A1: Review of the components of the SPEAKING framework
The scene or refers to the physical environment or extra-personal environment

setting of the (Saville-Troike 2003, 111) and timing of the event. The essential
interaction (S) descriptive questions to be addressed are: what kind of communicative
event is it? (e.g. hospitalisation negotiation); what is it about? (e.g. a
patient with severe heart complaints who will have to be hospitalised
for one night); why is it happening? (e.g. the negotiation is happening
because the doctor wants to keep the patient in observation for one night
as the cardiologist will only be able to see the patient at 8 am in the
morning, but the patient is afraid and hesitant to spend the night in the
hospital); where and when does it occur? (e.g. in the ED resuscitation
room at 1 am); what does the setting look like? (e.g. there are six beds
divided by green curtains.). Note that a scene can also change while
remaining in the same physical setting, for example if the position of
the interactants vis-à-vis each other changes.
The participants relate to who are the participants in the interaction and how they relate
of the EC (P) to each other (Saville-Troike 2003) (e.g. a female Dutch-speaking
doctor trainee, a male French-speaking consultant, a twenty-odd year
old Pakistani patient who appears to live on the street; an ad hoc
interpreter with unknown relationship to the patient). Personal elements
or autonomous factors (education, personal context, knowledge,
assumptions, experiences, ideas of the ED staff and patients) are also
considered (see e.g. Li 2011).
The ends or the looks at the purposes of the event and of the participants’ individual
purpose of the interactional actions (Saville-Troike 2003, 110) (e.g. the purpose of
interaction (E) history-taking maybe to identify the cause of a stomach ache; the
purpose of the patient’s groaning may be to express the amount of pain
(s)he has).
The act sequence reflects the ordering of communicative acts within an event. These
of the interaction communicative acts may be verbal, or non-verbal. According to Saville-
(A) Troike (2003), the act sequence may convey important information. A
possible act sequence for the opening sequence of the ED consultation
is as follows: the clinician introduces himself and shakes the patient’s
hand. Next, the clinician asks the patient what the problem is, the patient
answers, and the doctor may or may not ask a follow-up question.
The key of the relates to Hymes (1974, 57)’s notion of the “key” of an interaction
interaction (K) which refers to the “tone, manner, or spirit" of the speech act (e.g. is the
doctor talking in a sarcastic way; is the patient speaking in a hesitant
way?). According to Saville-Troike (2003, 113), the interpretation of
key can be culture-specific, which means it could be important to
consult people who belong to the same speech community as the
Annex 1 207
interlocutor(s) to understand what is meant by a particular key in a

particular context.
Instrumentalities refers to the choice of the channel/medium (e.g. written, oral in person,
or channels of the oral by phone) and the style which is used (e.g. formal/informal,
interaction (I) specialist/lay terminology)
The norms of refers to “prescriptive statements of behaviour, of how people ‘should’

interaction and act, which are tied to the shared values of the speech” (Saville-Troike
norms of 2003, 123). Norms of interpretation refer to the way in which specific
interpretation (N) behaviour should be interpreted. For example, while in some cultures,
looking a speech partner in the eyes is a token of respect; in other
cultures, it can mean the opposite. The same applies to a patient who
sits in silence for a while. In some countries, it is a sign of respect, in
others it means the opposite.
The genre of the describes the kind of event that is taking place. Examples of genre are a
interaction (G) joke, an anecdote, a dialogue, a monologue. In the context of an ED
consultation, the genre of the history-taking sequence tends to be a
question and answer session, the genre of the physical examination
tends to be imperative speech and the genre of diagnosis delivery tends
to be an argumentative monologue.
Source: Based on Hymes (1972)
Hymes’ framework has been used extensively by later scholars in the field of ethnography of
communication. Angelelli (2000) uses Hymes’ framework to compare the contextual
differences between conference interpreting and community interpreting, and recommends this
approach for context-specific analyses of interpreter-mediated encounters. Also Hale (2007,
228) argues that the ethnography of communication is a relevant methodology to study
language-discordant interactions, “where issues of bilingual competence, interactional skills
and cross-cultural knowledge are crucial.”
A1.1. Interactional sociolinguistics
“A main purpose of interactional sociolinguistics analysis is to show how

diversity affects interpretation.” (Gumperz 2001, 220)
While the ethnography of communication mainly draws on participant observation to analyse

in which context language is used and how different factors such as topic, genre, and
participants interact, interactional sociolinguistics goes a step further by analysing in addition
the turn-by-turn interpretive processes between the speakers of a communicative event
(Günthner 2008, 54). The main principle underlying interactional sociolinguistics is that “any
utterance can be understood in numerous ways, and that people make decisions about how to
interpret a given utterance based on their definition of what is happening at the time of the
interaction” (Gumperz 1982b, 130).
Annex 1 208
In order to describe and analyse misunderstandings, Gumperz (1982b) coined the concept of
“contextualisation cues”. This concept refers to any verbal and non-verbal sign that appears in
co-occurrence with lexical signs (words) and grammatical signs (sentences) and creates a
contextual ground, giving these words and sentences a context-specific meaning. Examples of
such verbal signs include code-switching (e.g. switching from dialect to standard language),
prosody (e.g. laying emphasis on a particular word or sentence), or the use of particular
language in a particular context. We learn these “contextualisation strategies” through direct
close contact with the family, a peer group, a community of practice, or close friends with
whom we share background knowledge (Gumperz 2001).
This shared background knowledge allows people to understand allusions and ways of
speaking. It allows us to interpret what is being said in the way that it has been intended. In
intercultural encounters interpretations of what is said are likely to be different for participants
according to their diverging cultural and linguistic backgrounds, and different conventions on
how to interpret what is said. 35 As contextualisation cues are not always apparent, and go
beyond what is literally being said, misunderstandings are often attributed to the speaker’s
negative attitude rather than to an asynchrony of shared linguistic and cultural background
knowledge (Gumperz 1982b). Tannen (2009, 302) sees contextualisation cues also as
“differences in conversational style”. In her view, conversational style is not something one
adds “like icing on a cake”. It rather is “the very stuff of which the linguistic cake is made.”
More generally, both the ethnography of communication and interactional sociolinguistics take
a broad perspective on the communication process. As in conversation analysis, the interaction
is the starting point for the ethnography of communication and interactional sociolinguistics as
well, but for the latter two disciplines, the broader socioeconomic and cultural background of
the speakers and the particular social occasion (e.g. an ED encounter) are considered crucial
elements of the context in addition (Duranti 2005, 26–27). In other words, to analyse the origin
of misunderstandings, the ethnography of communication and interactional sociolinguistics
look at elements at the level of the interaction (local inferencing), as well as at the level of the
broader context of the interaction and its participants (global inferencing).
Cicourel (1987) further elaborates on the impact of context, distinguishing the “narrow
context” and the “broad context” of an interaction. The “narrow context” of an interaction is
constituted of elements created by the discourse itself, such as the local organisation of turn-
taking, the internal syntactic structure, lexical choices, semantic networks, footing, pauses and
hesitations, and topic shifts. This is the scope of context usually considered by conversation
analysts who have a special interest in this context. However, Cicourel particularly highlights
the relevance of the “broad context” for discourse analysis, to account for the situatedness of
talk. The “broad context” refers to participant attributes and patterns of social, cultural, and
35
For example, in a hospital setting: if a doctor asks a patient “What brings you here?”, he wants to hear about
the patient’s complaints. A foreign patient may wrongly interpret this question as the doctor wanting to know how
the patient has arrived in the hospital. Also, if a doctor recommends a patient “You might want to rest for a few
days”, the patient may understand this merely as a potential development of his condition.
Annex 1 209
institutional organisation which are, in his view, indispensable for the understanding of
linguistic and non-linguistic aspects of communicative events. Therefore, he stresses the
importance of ethnographic fieldwork to complement transcript-based data.
A1.2. Social psychology

A major contribution in the field of social psychology is the communication accommodation
theory, as developed mostly by Giles and his co-authors (Giles, Coupland, and Coupland
1991a; Giles et al. 1987). This theory posits that when interacting, individuals may adapt
linguistic, prosodic, and non-verbal communicative behaviour to each other, to reduce
differences between themselves and others. This is referred to as convergence. According to
Street (1991), convergence is used by interactants in order to win approval, affiliate, establish
rapport, and/or communicate effectively. In a language-discordant doctor-patient consultation,
convergence may be evident in language-switching. A doctor may for instance switch form the
institutional language to the patient’s mother tong as to accommodate the later. Non-verbal
behaviour such as expressing involvement through gaze, gestures, body orientation and facial
expression can help to establish rapport if a language barrier complicates verbal
communication.
Conversely, divergence refers to the way in which such differences are accentuated by
speakers, and may particularly be observed in situations where speakers feel “threatened” in
their social identity (Giles, Coupland, and Coupland 1991a, 9). In doctor-patient interactions,
divergence may as well be evidenced by hierarchical behaviour by the doctor, if this runs
counter the patients’ wishes (Street 1991). Similarly, if doctors try to finish the consultation
quickly, exercise control by shifting topics or switching style abruptly and leave too little space
for patients to express concerns or discuss treatment alternatives, this can also be referred to as
‘divergence’ (Dryden and Giles 1987).
Using baby talk in conversations with young children may be appropriate and effective
instances of convergence. However, excessive convergence (over-accommodation) is often
perceived as downgrading, patronising, and hence irritating, for example when nurses use baby
talk to elderly, irrespective of their capabilities (e.g. Caporael 1981).
Annex 2 210
Annex 2. Application form for ethics committee

Annex 2 211
Annex 2 212
Annex 2 213
Annex 2 214
Annex 2 215
Annex 2 216
Annex 3 217
Annex 3. Initial informed consent form

Annex 3 218
Annex 3 219
Annex 3 220
Annex 4 221
Annex 4. Simplified informed consent form
Consentement éclairé
Projet BABELIRIS (2011-2015)
Etude sur la communication aux urgences de xxxxxxxxxx
Chercheur:
Antoon Cox
Centrum voor Vaktaal en Communicatie – Vrije Universiteit Brussel
Pleinlaan 5, 1050 Brussel
But de l’étude : analyse de la communication avec les patients.
Le chercheur m’a expliqué les objectifs de l’étude.
J’ai eu l’occasion de poser des questions concernant l’étude.
Je participe volontairement à cette étude.
Je peux à tout moment arrêter ma participation à cette étude.
Je donne mon accord pour l’enregistrement en format audio de la consultation.
Je sais que mes données personnelles resteront totalement confidentielles.
Nom du participant Date Signature
Nom du chercheur Date Signature

Annex 5 222
Annex 5. Transcription symbols used in transcript analysis
Symbol Explanation
// Final fall
/ Slight fall that indicates that more could be said
? Final rise
, Slight rise
- Truncation
.. Pause of less than .5 of second
... Pause of more than . 5 of a second (unless precisely timed)
<2> Precise units of time (number enclosed in brackets indicates the number
of seconds)
= Overlap
== Latching on to previous utterance
* Accent; normal prominence
** Extra prominence
() Unclear word
(did) Guess at unclear word
[clears throat] Non-lexical phenomena, vocal and non-vocal, which interrupts the
lexical stretch
(( )) Contextual information such as body movements, objects and the
language used
Source: based on Roberts (2015) and Gumperz and Berenz (1993)
Annex 6 223
Annex 6. Full transcript case study Midnight mastitis
Participants to the interaction

DOC: Doctor
PAT: Patient
COM: Patient companion
RES: Researcher (Antoon Cox)
GYN: Gynaecologist
UNK: Unknown person
UDR: Universal doctor (software)
Part A
1. 00:07.0 DOC ¿cuál es el problema? What is the problem?

…
¿usted me va a explicar? Are you going to explain it to me?
2. COM ((incomprehensible))
3. COM ..hmm?
4. 00:10.0 PAT = = ah oui, Oh yes
... aquí problema/ Problem here
((points at the right-hand breast of patient))
5. PAT = [groans]
6. 00:13.0 COM el...el pecho The... the breast
..
7. 00:16.0 DOC La... la..../ The… the…
como se llama, How does one call this...
= = el pecho? the breast?
8. 00:19.0 COM = pichu..pichu// Brist... brist....
= breast.. breast=
9. DOC [ hmhm ]
10. COM = = pichu problema/ Brist problem
= problem with the breast=
11. DOC y cuanto tiemp (tiempo)- And how long
= for how long? =
12. 00:27.0 COM = = cuando el dolor* When the pain
cuando el dolor el pichu*, When the pain, the breast...
el dolor y la cabiza::, la barriga::, la todos* The pain and head, the stomach, the
everything
= she has a headache, stomach ache, she
has pain all over=
<1>
¿desde cuando? ¿cuánto tiempo? Since when? how long?

.. ¿cuántos días*? How many days?
14. 0:35.0 COM = (pero si,):: But yes
casi casi un año // pero- Almost, almost one year, but
Annex 6 224
15. 0:35.5 DOC == ¿un * año? One year?

16. 0:38.0 COM = un año, oui One year, yes
17. 0:38.3 COM pero que me yo cuando la traelo aqui*// But when I bring her here
= but when I brought over here ==
18. 0:38.6 PAT ((incomprehensible))*
19. 0:43.0 COM y una pastilla* et c’est fini, c’est fini// and a tablet and it’s finished, it’s
finished
<1> = she took a tablet, and the pain was
gone =
20. 0:44.0 DOC a ver/ Let’s see
21. 0:54.0 COM = = un año cuando está, One year ago
..
le duele/ It hurts her
=?????=
22. DOC [ hmhm ]
23. 0:56.0 COM = = pero yo me la he traido aquí he, But I have brought her over here
..
he pasado un (TAC):: he pasado y (que I went through a (CAT), she drank a
bebe un) una pastilla y c’est fini// tablet and it’s finished
(1) = they did a CAT scan, gave her a tablet
and the pain was gone =
C’est fini// no hay le duele// It’s finished. She has no pain

= she did not have pain anymore=
24. 0:57.0 DOC = = oké Okay
25. 1:03.0 COM = = ahora un año que pasa otra vez/ Now, one year that it happens again.
y cuando el picho mm el bebé, And when the breast mm the baby

.. = she is in pain when she is
breastfeeding her baby=

26. 1:04.0 DOC ¿ella tiene bebé? Does she have a baby?
27. 1:04.3 COM = =ah, oui Oh yes
..
28. 1:04.6 DOC ¿le da leche*? Is she giving it the breast?
<1>
29. 1:08.0 COM lichi, cuando está lichi y el le duele// Mealk, when the mealk is there and it
<1> hurts her
= when she breastfeeds, it hurts=
30. 1:08.5 DOC Oké Okay
…
31. 1:13.0 COM me entiendes? Do you understand me?
32. 1:13.5 DOC …más o menos More or less
<1>
((speaks in Moroccan Arabic)) A little
[ laughs ]
Annex 6 225

..
entonces, So
la señora le duele el pecho/ The lady’s breast is hurting?
34. 1:16.5 COM == mmh
..
35. 1:19.0 DOC ¿desde ya hace un año? Already since one year?
...
o es lo mismo* que hace un año? Or is it the same as one year ago?
<1>
36. 1:22.0 COM Mis ..mo:: The same
…
mismo un año// The same one year
= the same as one year ago=
37. 1:26.0 DOC oké, ahora es nuevo*? Okay, now it is new?
.. = Now, she is in pain again? =
ahora nuevo? New now?
= Has the pain started over again?=
38. 1:26.5 COM ahora * está nuevo Now, it is new
39. 1:27.0 DOC = = oké,…¿de cuánto días, cuántos días? Okay, how many days, how many days?
= For how many days has she been in
pain? =
40. 1:29.0 COM = = sólo hoy Only today
41. 1:31.0 DOC = = hoy, hoy empezó?... comienza hoy? Today? Today it started? Does it start
today?
= did it start today?=
42. 1:32.0 COM = = claro après de midi/ Of course, afternoon
43. 1:34.0 DOC = = y ¿es el mismo* que hace un año? And is it the same as one year ago?
44. 1:35.0 PAT = = claro Of course
45. 1:36.0 DOC eso es? That’s it?
<1> = are you sure?=
46. 1:44.0 DOC o::ké// Okay
<1>
y está dando el pecho, and is she breastfeeding;
¿está dando la leche a su bebé desde hace Has she been giving milk to the baby for
un año? one year?
...
47. 1:51.0 COM no*, no*, no:: No no no
y ahora es (tres,). and now it is (three/a lot) =?=
y hace un año le duele.. and one year ago she was in pain
y ahora le bastante le duele.. and now she is having a lot of pain
48. 1:53.0 COM no poder comer, no poder leche a bebé no/ Not able to eat, not able to milk to baby
no
= she is unable to eat, unable to
breastfeed the baby=
49. 1:55.0 DOC = = le duele mucho? Is she having a lot of pain?
Annex 6 226
50. 1:56.0 COM = = demasiado* Too much

51. 1:59.0 DOC oké, ¿cuántos años tiene su bebé? Right, how old is the baby?
...
52. 2:02.0 COM euh:: un año mmh un año casi casi/… Euh one year mmm one year almost,
almost
53. 2:04.0 DOC = = oké *… y ¿tiene fiebre? Okay, does she have fever?
<1>
54. 2:06.0 COM ha fiebre* .. oui She has fever, yes
55. 2:09.0 DOC = = sí *, mucho calor? yes, very warm?
56. 2:10.0 COM mucho duele * y/ A lot of pain, and
57. 2:11.0 DOC = = ¿ha vomitado? Has she vomited?
58. 2:13.0 COM = =ha vomitado She has vomited
<2>
<2>
¿comprende español usted? ((He gazes at Do you understand Spanish?
the patient))
...
60. 2:19.0 COM = = el picho es muy raro, rojo me entiendes The brist is very weird, red, do you
... understand me?
= the breast looks very weird, reddish,
do you understand me? =
61. 2:22.0 DOC sí, sí, vamos a ver// Yes yes, we will see.
<1>
62. 2:23.0 COM tiene algo o? Does she have something?
.. = What is the problem?
63. 2:35.0 DOC = = se, señora, ma, madam,
..
¿me comprende? o no? do you understand me or not?
= nada. español? francés? Arabe? Nothing? Spanish? French? Arabic?
= [PAT groans]
((he addresses the companion:))
¿qué habla? árabe. Solo* árabe? What does she speak? Arabic? Only
Arabic?
64. 2:36.0 COM = = Árabe Arabic
65. 2:36.5 DOC ¿de donde es? de marruecos? Where is she from? From Morocco?
66. 2:37.0 COM = =Marruecos, Marruecos// Morocco, Morocco
<3>
67. 2:44.0 DOC y cuando- And when
..
en Bélgica, ¿hace mucho tiempo ha estado? in Belgium? Since a lot of time has she
been?
= Since when has she been living here?
68. 2:45.0 COM = = 2005 ((in French)) 2005
Annex 6 227
69. 2:49.0 DOC 2005 ((in Spanish)) ella, ah, oké oké 2005, she, ah, okay, okay
<1> = okay, so she has been here since
2005=
70. 2:50.0 DOC y usted vive aquí o de And you live here since?
= and what about you, since when have
you been living here?
71. 2:54.0 COM = = no* vive aquí, en 2005, que vendrá de No live here, in 2005, she will come
España hasta aquí from Spain to here
(( I think he is talking about the patient)
=?=
72. 2:57.0 DOC = = ah, oké, oké Ah, okay, okay
<1>
73. 2:58.0 COM se (casi) en nuevos, diez años She (almost/got married) in 9, 10 years
<9>
74. 3:04.0 DOC ((feeling PAT’s pulse)) Does she have a baby?
¿tiene un* bebé?
75. 3:08.0 COM …no tengo quatre ((French)), cuatro* No I have four, four
<2> =???=
76. 3:08.5 DOC …cuatro? Four?
<2>
77. 3:09.0 COM ((COM gazes at PAT))
78. 3:12.0 DOC oké, oké, on va voir ça// Okay, okay, let’s see
<2>
80. 3:16.5 DOC ¿ha vomitado? Has she been vomiting?
= yes, she has=
82. 3:19.0 DOC ..¿sí? Yes?
84. 3:20.0 DOC ¿mucho? A lot.
<1>
85. 3:21.0 COM mucho A lot
<1>
86. 3:21.5 DOC ¿no está embarazada? She is not pregnant?
87. 3:22.0 COM = = no*, no está embarazada No she is not.
88. 3:24.0 DOC = = ¿seguro? Are you sure?
89. 3:25.0 COM = = solo* tengo bebé de un año I only have a one year old baby
==my baby is only one year old==
90. 3:28.0 DOC …sí pero todo es posible// Yes, but anything is possible
91. 3:29.0 COM No* No
<2>
92. 3:31.0 DOC ça c’est bon ((he looks at the blood That’s good
pressure)) [clears his throat]
Annex 6 228
((physician puts on his rubber gloves)

(clears his throat))
93. 3:41.0 DOC oké, on va regarder, hein Okay, let’s have a look, (see, see)
((Moroccan Arabic:)) tchouf?... Tchouf? (see, see)
((DOC performs auscultation of the
breasts))
94. 3:43.0 PAT = =[groans]
95. 3:44.0 DOC = =oké? Okay?
<1> = may I ? =
96. 3:44.5 DOC ¿dónde // le duele? Where does it hurt ?
97. 3:47.0 COM = = (ça va?) * ((addressing the researcher)) Everything okay?
<12>
98. 3:59.0 COM ((DOC is feeling patient’s breasts))
99. 4:00.0 DOC ((Moroccan Arabic))… Pain?
101. 4:03.0 PAT = [groans][breathes out]
102. 4:05.0 DOC c'= = est ça? ((DOC is palpating)) This is it?
103. 4:11.0 DOC <2,5>{incomp}
105. 4:14.0 PAT = [groans]
106. 4:20.0 PAT =[groans]
107. 4:21.0 DOC …et ici non?...((pointing at the other And here not?
breast))
108. 4:23.0 PAT …((Berber:)) {incomp}
109. 4:25.0 COM …Tambien le duele This is also hurting her =This one also
hurts =
110. 4:28.0 PAT = = ((Berber:)) {incomp}
111. 4:28.3 COM ici (a veces) ou c’est ceci? Here (sometimes) or it is here
((pointing at the other breast))
112. 4:28.6 DOC …Como? What do you mean?
113. 4:30.0 COM = = Et- And…
114. 4:31.0 DOC = là c'est normal? ((talking about breasts Is this one normal?
while palpating))
115. 4:40.0 COM = =No*.. No, how many … it is okay that she
cuanto ((incomprehensible)) está bien que changes it to the other one, now the
la cambia para otro/.. ahora la otra está other one is fine.
bien// = it is okay if she uses the other breast,
<2,5> which is fine now =
116. 4:45.0 COM no hay problema pero ici, pase There is no problem, but this one
<3> = with that breast there is no problem,
but with this one there is =
117. 4:46.0 DOC solo esto*? Only this one?
Annex 6 229
..
Aquí* le duele ((pointing at breast))? Is she having pain here?
118. 4:48.0 COM = = no, solo esto No, only this one
119. 4:51.0 DOC = =oké Okay
120. 4:58.0 COM esto, qué problema para fiebre, para This one, that it is a problem for fever,
to
=?=
121. 4:58.5 PAT ((patient moves as doc is palpating))
122. 5:05.0 DOC xxxxxxxxxxxxxx Moroccan Arabic
xxxxxxxxxxxxxxxx
tr: may I have a look?
escuchar, voy a escuchar hein, oké?
123. 5:07.0 DOC ¿se puede sentar? (( doc is touching pat’s Can you sit?
shoulder)) = would you like to sit up straight?
124. 5:07.5 COM ah oui Oh yes
125. 5:12.0 DOC sentar, sentar. Sit, sit, ok, take a deep breath
oké,
((uses stethoscope on the back))
..
respirar bien Take a deep breath.
((physician is ausculating patient’s back))
((physicien demonstrates the breathing))
(7)
126. 5:12.5 COM
127. 5:17.0 PAT ((breaths in and out))
128. 5:17.5 PAT ((silence))
129. 5:27.0 DOC xxxxxxxxxxxxxx Moroccan Arabic OK, pain?

xxxxxxxxxxxxxxxx
tr: oké, {enough} dolor?
((palpating patient’s back))
xxxxxxxxxxxxxxxx
{dolor?}¿duele?
131. 5:30.0 DOC non? No?
132. 5:32.0 DOC non oké, ça va No ok?

All right then
xxxxxxxxxxxxxxxx
tr: enough ennough
134. 5:36.0 DOC y aquí, ¿le duele? Is she having pain here?
135. 5:36.5 PAT {incomp}
136. 5:39.0 DOC ¿le duele*? Is this hurting here?
Annex 6 230
137. 5:40.0 PAT {incomp}

138. 5:44.0 DOC xxxxxxxxxxxxxx Moroccan Arabic A little?
xxxxxxxxxxxxxxxx
(1)
139. 5:46.0 COM le duele también a la barriga por aqui el // She has also stomach ache and over here
picho the breast
140. 5:47.0 DOC ha comido * Has she eaten?
¿ha comido hoy? ((doc gestures with his Has she eaten today?
hand and his mouth the action of eating))
141. 5:49.0 COM oh, está comiendo Oh, she is eating
142. 5:50.0 DOC ¿sí? ¿está comiendo? Yes?
Is she eating?
143. 5:54.0 COM muy bien Very well
144. 5:54.5 COM no, está comiendo No, she is eating
145. 5:56.0 DOC huh? huh?
146. 5:56.5 COM está comiendo en mediodía She is eating at lunch
= she ate at lunch time=
147. 5:57.0 DOC Bien? good
148. 5:58.0 COM Mediodía Lunch time
149. 5:59.0 DOC Ah ah
150. 6:00.0 COM sólo mediodía Only lunch time
151. 6:04.0 COM pero está comiendo (aquí) But she is eating(here-
(1)
152. 6:04.3 DOC (vomitado) Vomited
153. 6:04.6 COM Claro sure
154. 6:05.0 DOC y ahora, ¿quiere vomitar? And right now, does she want to vomit?
155. 6:05.5 COM Claro sure
156. 6:08.0 DOC ¿ahora*? Now?
157. 6:08.5 COM ahora (misma fuera?) Right now
=????=
158. 6:16.0 DOC (6)
159. 6:18.0 DOC l’anamnèse n’est pas terrible I am not too happy with the history
mais euh bon ((physician adresses the taking.
researcher.))
(3)
¿le puede- Can you
Le puede preguntar si ahora quiere Can you ask her if she wants to vomit?
vomitar?
161. 6:30.0 COM xxxxxxxxxxxxxx Berber
xxxxxxxxxxxxxxxx
TR: He is asking whether want to vomit at
this moment
Annex 6 231
162. 6:30.5 COM xxxxxxxxxxxxxx Moroccan Arabic

xxxxxxxxxxxxxxxx
TR: Do you want to vomit?
163. 6:32.0 COM xxxxxxxxxxxxxx Moroccan Arabic
xxxxxxxxxxxxxxxx
TR: no
164. 6:32.5 COM No
165. 6:34.0 DOC ¿no?
166. 6:34.5 COM No
167. 6:40.0 DOC muy bien Very good
(2)
y, ¿puede mostrar donde le duele más? Can you/she indicate where it hurts
most?
168. 6:40.5 COM …((Berber)) He is telling you: where do you have
(feel) most pain.
169. 6:42.0 PAT = = ((Berber)) Here

((she touches her right-hand breast))
170. 6:46.0 DOC …mostrar, la, Show, there
oké,.. c'est tout// Okay, that’s all
171. 6:46.5 COM = = solo el pecho Only the breast
172. 6:50.0 DOC ha tenido, frío, ca - calor frío, calor She has been feeling cold, hot, cold, hot
173. 6:52.0 COM ((Moroccan Arabic)) He wants to know whether you feel cold
or warm?
174. 6:52.5 PAT ((Moroccan Arabic)) I am feeling cold and I am shivering.
175. 6:55.0 COM frío // y calor// Cold and hot
176. 7:00.0 DOC = = oké *
(1)
Ok
(2)
177. 7:00.5 COM frío que - Cold that-
178. 7:02.0 DOC = voy a hablar con mi colega/ I am going to talk to my colleague
179. 7:02.5 COM = = huh?
180. 7:04.0 DOC Ginecólogo* …entiendes? Gynaecologist, do you understand?
181. 7:04.3 COM = = mmh?
182. 7:04.6 DOC = = Hm*?
183. 7:07.0 COM …¿cómo? What?
184. 7:08.0 DOC mi colega/ My colleague
185. 7:08.5 COM = = Mmh
186. 7:10.0 DOC = =mi compañero, *ginecólogo// My companion, gynaecologist
187. 7:10.5 COM = = Mmh
188. 7:11.0 DOC ¿comprende*? Do you understand?
189. 7:12.0 COM = = mmh, claro Sure
Annex 6 232
190. 7:15.0 DOC = = mi compañero, que se ocupa de las My companion, who takes care of
mujeres,.. especialidad// women
Specialty
191. 7:18.0 COM = = Mmh
192. 7:18.3 DOC = = ¿ya*?... y… probablemente, es una Ok? It is probably an infection here
infección aqui
193. 7:18.6 COM = = Mmh
194. 7:26.0 DOC Ya* eum. < 1,5>
probablemente, hay que darle She is probably to be given antibiotics
antibióticos/… ¿comprendes?
195. 7:27.0 COM = =((incomprehensible))
196. 7:27.5 DOC = = huh?
197. 7:28.0 COM = = ((incomprehensible))(muito sabes?) A lot you know =?=
198. 7:29.0 DOC = = huh?
199. 7:31.0 COM = = sabes como va las cosas pero yo no You know how things go but I know
sabes nada nothing
200. 7:33.0 DOC = = sí*.., estoy explicando* Yes, I am explaining it
201. 7:34.0 COM = = Ay* claro oké Ah sure okay
202. 7:36.0 DOC …oké? Okay?
203. 7:36.5 COM …Claro Sure
204. 7:40.0 DOC …y vamos saber si hacemos euhm un And we will know whether we will do a
laboratorio// lab
==we will find out whether further
medical checks will be needed==
205. 7:41.0 COM …Mmm Mmm
206. 7:45.0 DOC …comprende? Do you understand?
<3>
207. 7:45.5 COM Merci Thank you
<1>
208. 7:49.0 DOC ¿ella tiene preguntas?... Does she have questions?
puedes preguntar// Can you ask?
209. 7:49.3 COM = = ((talks in Berber)) He is asking you, whether you want him
to give you one so you can take it.
210. 7:49.6 PAT …((talks in Berber)) why?
211. 7:51.0 COM ..((talks in Berber)) It has to do with the pain you are having.
212. 8:01.0 PAT …((talks in Berber)) Yes
213. 8:01.3 COM …{incomp} huah?
214. 8:01.6 DOC = =huah?
215. 8:02.0 COM {incomp}
216. 8:02.5 DOC = = ((talks in Moroccan Arabic)) Yes
217. 8:03.0 COM = = sí,
Annex 6 233
218. 8:04.0 DOC = =oké, ça va.
Part B
219. 00:17.0 DOC tiene que quedar en el hospital She will have to stay in the hospital
220. 00:17.5 COM ¿cómo? How?
221. 00:18.0 DOC hospital Hospital
222. 00:32.0 COM oui ((sound of a door that opens)) Yes
223. 00:30.0 UDR ((in Moroccan Arabic)) You need to be admitted to hospital
224. 00:33.0 DOC t'as compris? Have you understood me?
225. 00:34.0 COM ah oui Ah yes
226. 00:34.5 DOC huh? Huh?
227. 00:42.0 (break)
228. 00:44.0 UDR ((in Moroccan Arabic)) Do you understand the question?
229. 00:47.0 DOC d'accord Okay
230. 00:47.5 COM Mmh Mmh
231. 00:52.0 DOC tiene que recibir euh medicamentos por la She has to receive the medication
vía oral through the oral way
232. 00:53.0 COM por vía through the way
233. 00:57.0 DOC euh, ¿toma medicamentos ella? Eeuh, Is she taking medication?
234. 00:59.0 COM Non No
235. 00:59.5 DOC huh? Huh?
236. 01:00.0 COM Non No
237. 01:02.0 DOC je vais te montrer ce que c'est (heel stil) I’m going to show you what it is
238. 01:10.0 (break)
239. 01:14.0 PAT ((patient moves))
240. 01:13.0 RES faut aller au début You have to go back to the startpage
241. 01:15.0 DOC de l'antécédente Of the pathology
242. 01:15.5 RES Oui Yes
243. 01:21.0 ((sounds of hospital’s hall))
244. 01:27.0 (break)
245. 01:40.0 DOC ah, c'est ça. mais il n'a pas d'allergie? Ah, this is it. But doesn’t he have
((almost whispering)) allergies?
246. 01:52.0 RES euh, c'est une étape là-bas, mais je ne sais Euh… that is a step further, but I dont
pas si euh know whether euh
247. 01:59.0 DOC straight forward
248. 02:02.0 (break)
249. 02:05.0 DOC ah ouais. Ah yes
250. 02:11.0 UDR ((Moroccan Arabic)) Are you allergic to any medication?
Annex 6 234
251. 02:12.0 COM ((incomprehensible))

252. 02:12.5 PAT (incomprehensible))
253. 02:13.0 COM (incomprehensible))
254. 02:13.5 COM Oui Yes
255. 02:15.0 DOC oui quoi? Yes what?
257. 02:25.0 DOC c'est quel médicament? Which medication is it?
258. 02:31.0 COM que hay que le duele cuando hay una cosa There is what hurts her when things are
muy mal e bad
259. 02:31.5 DOC Mmh Mmh
260. 02:36.0 COM la douche la toilette, cuando va entrer The shower, the toilet, when she is going
dans la toilette, ayuda to enter the toilet, help
261. 02:37.0 DOC ah, oui, ce n'est pas vraiment un Ah, oui, it is not really medication then.
médicament, alors.
262. 02:40.0 DOC Mmh Mmh
263. 02:43.0 (break)
264. 02:53.0 UDR ((Moroccan Arabic)) Can you tell me the name of the
medicine or medicines?
265. 02:53.5 COM Non No
266. 02:54.0 DOC Non No
267. 02:54.5 UDR ((Moroccan Arabic)) Penicillin?
268. 02:55.0 DOC //non No
269. 02:55.5 COM non* No
270. 03:05.0 DOC ((clears his throat)) oké, ça va, oké, Eeuh, it is ok, good, we are going to
d'accord, on va donner les antibiotiques, give you antibiotics, ok?
euh, vamos a los antibióticos, ¿vale?
271. 03:05.5 COM oui Yes
272. 03:07.0 COM que duerma aquí non? She sleeps here no?
273. 03:08.0 DOC sí, en hospital Yes, in hospital
274. 03:08.5 COM à l’hospital In the hospital
275. 03:21.0 DOC ((clears throat)) oké?
276. 03:34.0 PAT ((incomprehensible))
277. 03:34.3 COM hum?
278. 03:34.6 PAT ((incomprehensible))
279. 03:46.0 COM pardon, après ((incomprehensible)) el Excuse me, after, the breast for the child,
picho para el niño, no hay ningún there is no problem?
problema?
280. 03:49.0 DOC no. pero el niño no puede quedarse aquí o No, per the child cannot stay her or
que no? what?
281. 03:53.0 COM no es es a la casa No, it is at home
Annex 6 235
282. 03:56.0 DOC tiene que sacar la leche, sacar, ¿entiendes? She has to take out (drain) milk, to take
out? Do you understand?
284. 03:59.0 DOC con una máquina ((imitates the sound of a With a machine
machine))
285. 03:59.3 COM ah oui
286. 03:59.6 DOC Umh
287. 04:09.0 DOC oké? euh, ¿tiene preguntas? Ok, euh, do you have any questions?
288. 04:12.0 DOC ¿preguntas? questions
289. 04:13.0 COM No
290. 04:16.0 DOC non, oké
291. 04:18.0 DOC ha comprendido // todo? Have you understood everything?
292. 04:20.0 COM ¿qué problema * tiene? What is here problem?
293. 04:24.0 DOC hein?
294. 04:25.0 COM problema y para el picho? Problem with the breast
295. 04:26.0 DOC bon, je vous explique // hein ((sighing)) Ok, I will explain it to you
296. 04:26.5 COM pardon hein *
297. 04:42.0 DOC on va voir si pas facile à expliquer
(mumbling)
298. 04:54.0 DOC ((il écrit quelque chose sur une feuille))
299. 04:55.0 RES mastite, het stond er niet in We are looking for the word mastitis in
UDR, I cannot find it
300. 04:57.0 DOC mmh, d'accord
301. 04:57.5 RES Oké
302. 04:58.0 DOC Euh
304. 05:03.5 PAT ((mumbles))
Part C
305. 00:03 UDR You have the following illness or

condition
306. 00:08 UDR An infection?
307. 00:13.3 DOC vous comprenez?
308. 00:13.6 COM Non
309. 00:17 COM He tells you that this disease is
contagious.
310. 00:17.5 DOC vous comprenez?
311. 00:18 COM He tells you that this disease is
contagious.
Annex 6 236
312. 00:22 DOC tu vois, ici, il n'y a pas

313. 00:24 COM {incomprehensible}
314. 00:27 DOC tout ce qui est gynéco
315. 00:27.5 PAT {incomprehensible}
317. 00:35 RES il y a universal woman
318. 00:36 DOC oui, tu ne l’as pas?
319. 00:40 COM // {incomprehensible}
321. 00:40.4 DOC peut-être * je ne sais pas //
323. 00:43 DOC peut-être ici non il n’ y a pas
325. 00:59 PAT {incomprehensible}
326. 01:02 (break)
327. 01:05 ((a door is opened))
328. 01:09 DOC [Ok, bon, on lui a expliqué // euh que
329. 01:18 GYN (to the doctor :) [donc, en fait, voilà *
c'est pour ça que je viens te voir, je viens
de regarder dans la bible, ils disent que
si l'état clinique de la patienteest
correcte, que je trouvais // le cas, hein?
330. 01:20 DOC ouais *
331. 01:26 GYN euh, il ne faut pas spécialement
hospitaliser mais pour peu qu'elle soit
compliante et qu’ elle appelle lundi à la
clinique du sein.
332. 01:27 DOC vas-y pour lui expliquer ça, quoi (heel
stil)
333. 01:28 GYN Ouais
334. 01:29 DOC qu'elle appelle lundi?
335. 01:29.5 GYN Ouais
336. 01:35 DOC mais comment vont-ils appeler, ils ne
vont jamais appeler, pas avec ce qu'ils ne
savent pas parler, ils vont appeler,
qu'est-ce qu'ils vont faire
337. 01:44 GYN prendre un rendez-vous pour qu'on voie
comment ça évolue
338. 01:52 GYN donc de toute façon, on va donner des
antiinflammatoires et des des, euh et des
antibiotiques (per os?) si tu penses que
ils peuvent revenir lundi, alors, ça sert à
rien de l'hospitaliser mais euh parce que
cliniquement elle est douloureuse
Annex 6 237
339. 01:51 DOC ouais

340. 01:54 GYN mais ça * va ce n'est pas euh c'est pas,
c'est pas // non plus//
341. 01:55 DOC un antiseptique*
342. 02:10 GYN voilà, donc euh il faut vu qu'elle a fait
quand même 38,5 faut piquer pour faire
une hémato CRP hémoculture et euh, et
normalement, elle va a césar de paepe, à
clinque du sein, ça c'est ce qu'ils mettent
dans une note de leur bible pour les //
mastites
343. 02:12 DOC ouais * mais
344. 02:13 GYN ((lacht))
345. 02:15 DOC tu penses qu'ils seront compliants?
346. 02:22 GYN faut demander à monsieur, demande-lui
s’il veut bien revenir avec sa femme
lundi César De Paepe ben euh
347. 02:22.5 DOC Lundi?
348. 02:24 GYN ouais
349. 02:29 DOC otra opción, dice mi colega, ella es la Another option, my colleague says, she
especialista is a specialist
350. 02:29.5 COM Mmh
351. 02:33 DOC dice que usted puede irse a la casa She says, that you can go homme and
llevarla antibióticos take the antibiotics with you
352. 02:33.5 COM mmh
353. 02:35 DOC tiene que tomar pastillas She has to take tablets
354. 02:35.5 COM mmh
355. 02:38 DOC pero tiene que volver el Lunes But she has to come back on Monday
356. 02:39 COM lundi oui
357. 02:39.5 DOC lundi
358. 02:40 COM Mmh
359. 02:43 DOC à césar de paepe, connaissez césar de ((this is is another campus named “César
paepe? de Paepe »))
360. 02:43.3 COM ¿aquí? Here?

361. 02:43.6 DOC césar de paepe
362. 02:45 DOC ¿conoce? Do you know it?
363. 02:45.5 COM ah oui
364. 02:47 DOC oui?
365. 02:48 COM ¿aquí en [name hospital] ? Here in [name hospital] ?
366. 02:48.5 DOC Non
367. 02:49 COM mmh
Annex 6 238
368. 02:50 DOC César de Paepe, conaissez? De you know César de Paepe
369. 02:51 COM non (incomprehensible) adresse ningún No, address, no problem
problema
370. 02:52 DOC oké
371. 02:52.5 GYN oké
372. 02:55 DOC clinique du sein Breast clinic -
373. 02:57 GYN ouais, on va leur donner le numéro
374. 02:57.5 DOC tu vas leur donner
375. 02:59 COM (incomprehensible) adresse no hay address, no problem
ningún problema
376. 03:02 DOC tu lui donn' tous les anti parce que je ne
sais pas combien il a d' enfants à la
maison
377. 03:03 GYN ouais, quatre
378. 03:05 DOC ah
379. 03:07 GYN oké, pas d'allergies?
380. 03:08 DOC non, non
381. 03:09 GYN prend, elle prend quoi comme
médicament, tu sais?
382. 03:10 DOC elle prend rien
383. 03:13 GYN parce que si ça ne va pas l(es antastiqu')
tout à l'heure, ben euh alors, on se
comprend pas
384. 03:14 DOC je pense qu'ici on lui a donné un
dafalgan
385. 03:17 GYN ah, c’est ça, oké
386. 03:17.5 DOC Mh
387. 03:27 GYN [parfait oké, ben alors on va faire ça. elle
va appeler la clinique du sein, je vais lui
donner un papier pour savoir si c'est
quand même plus facile vu que
cliniquement ça a l'air d'aller ((ze haalt
adem))
388. 03:34 GYN euh ce qu'il faut absolument lui dire c'est
qu'elle doit continuer à donner le sein
des deux côtés et du côté qui où il y a la
mastite aussi
389. 03:35 (break)
390. 03:36 DOC ¿comprende?
391. 03:38 COM ah oui (incomprehensible)
392. 03:44 DOC non, non, tiene que tomarla, tiene que No no, she has to take it, she has to give
darla, leche, pecho it (milk) milk, breast. ((she has to take
the antibiotics and she has to give breast
with her two breasts)
Annex 6 239
393. 03:44.5 COM mmh

394. 03:45 DOC de dos lados The two sides
395. 03:48 COM de dos lados. pero con la infección, ¿no The two sides, but with the infection,
hay problema? there is no problem?
396. 03:50 DOC no hay problema los dos Non, there is no problem, the two
397. 03:52 GYN c'est très important de bien // libérer
398. 03:52.5 DOC portar *
399. 03:54 GYN le sein // qui
400. 03:56 DOC sino, va * a inchar más, va a infectar Otherwise, it will swell even more, and
más, comprendes? get more infected. Do you understand?
401. 04:00 COM He is telling you that you have to breast-
feed with the two sides so the milk will
not come again here and you get ill.
402. 04:06 DOC {twee} (in het Arabisch)

404. 04:13 COM No, it is on Monday, we are going to a
hospital that is far away. They are going
to give me the address.
405. 04:15 COM merci
406. 04:20 GYN [ça va? je vais mettr' tout ça à l'examen
407. 04:21 COM pardon
408. 04:24 COM la otro, la otro, hospital, ¿c’est aquí? The other, the other hospital, is it here?
409. 04:24.5 GYN on va donner une adresse ((iemand We are going to give you the address
hoest))
410. 04:25 COM ¿aquí en Brussel? Here in Brussels?
411. 04:25.5 GYN oui // oui oui
412. 04:27 DOC oui oui *
413. 04:28 GYN c'est tout près
414. 04:29 DOC c'est par là, c'est pas loin
415. 04:29.5 COM mmh
416. 04:31 DOC no es muy lejos, no es muy lejos. en el It is not very far, it is not very far. It is in
centro the centre
417. 04:35 COM centro Centre?
418. 04:39 COM [aquí, no ? Not here?
419. 04:39.5 DOC [eu::h
420. 04:40 COM flamand non? Flemish? ((he may be referring to the
VUB university hospital which is Dutch
speaking))
421. 04:42 DOC hein?
422. 04:43 COM flamand non? Not Flemish?
Annex 6 240
423. 04:44 GYN non, non

424. 04:45 DOC cerca de Anneessens, conoces Close to Place Anneessens, do you know
Anneessens? Aneessens?
425. 04:47 COM ah oui le tengo Oh yes, I have it
426. 04:47.5 DOC Lemonnier
427. 04:48 COM Lemonnier vive allí Lemonier lives there ((maybe he wants
to say, I live close to Lemonier))
428. 04:53 DOC voilà, ça va aller, alors ((lacht))
429. 04:54 GYN c'est plus gare centrale que lemmonier
hein.
430. 04:55 DOC (chupinal?)
431. 04:57 GYN Oké
433. 05:00 COM merci monsieur.
434. 05:00.5 DOC oui
Part D
435. 00:02 GYN j'ai donné toutes les prescriptions // pour
ce
436. 00:04 DOC et il faut une formation * avec, c'est tout
437. 00:07 GYN ouais, c'est mais je pars pendant deux
semaines, mais je ne sais si le
pharmacien (onverstaanbaar)
438. 00:11 DOC ouais (onverstaanbaar)
439. 00:14 DOC tchouf (staticide?) see
440. 00:14 GYN je vais mettre l’écran
441. 00:15 DOC antibiótico antibiotics
442. 00:15 COM mmh
443. 00:21 DOC Cuatro por día Four a day
444. 00:21 COM cuatro e four
445. 00:22 DOC Cuatro four
446. 00:22 COM ah oui me entiendes Yes, do you understand me?
447. 00:28 COM cuatro pastillas para el día Four tablets per day
448. 00:30 DOC 3 (in het Arabisch)
449. 00:30 COM quatro
450. 00:30 DOC 2 (in het Arabisch)
451. 00:31 COM 4 (in het Arabisch)
452. 00:33 DOC ouais {3(in het Arabisch)
453. 00:37 PAT no, le entiende I do not understand you
Annex 6 241
454. 00:37 DOC cuatro por día, catorce días Four per day,14 days
455. 00:38 PAT catorce días 14 days
456. 00:38 DOC dos semanas 2 weeks
457. 00:41 PAT mmh
458. 00:42 DOC dafalgan¿conoces dafalgan? paracetamol Do you know dafalgan? paracetamol
459. 00:43 PAT ((groans))

460. 00:45 DOC es bueno para el dolor It is good for the pain
461. 00:45 PAT dolor
462. 00:49 DOC nurofen, antiinflamatorio Nurofen, anti-inflammatory
463. 00:49 PAT ((groans))

464. 00:56 DOC et le a recetar también contra los vómitos She (gynaecologist) will also prescribe
something against vomiting.
465. 00:56 PAT mmh
466. 00:59 DOC ¿me entiendes? tabletas, pastillas contra Do you understand me? Tablets,
los vómitos. against vomiting.
467. 01:04 COM mmh
468. 01:11
469. 01:12 RES je ferme la porte
470. 01:17 ((break))
471. 01:18 DOC ahorita va a volver She (gynaecologist) will back
immediately
472. 01:19 COM ah oui
473. 01:21 DOC ¿vale? Ok?
474. 01:25 ((geluiden uit de gang))
475. 01:26 GYN parfait
476. 01:28 DOC il ne faut pas trop Ubiprofem, doc elle il
vomit
477. 01:33 GYN Ben, ouais mais euh dans pas longtemps,
il faut absolument lui donner des
inflammatoires pour sa mastite
478. 01:35 DOC des zantac en plus peut-être
479. 01:35 GYN mmh
480. 01:36 DOC elle a eu du zantac
481. 01:37 GYN ((incomprehensible))((noise from the
corridor))
482. 01:40 DOC ((incomprehensible))((noise from the
corridor))
483. 01:40 GYN ouais
Annex 6 242

corridor))
485. 01:45 GYN mais si elle ne l'a pas, tu lui mets d'office
si jamais elle n'a pas de d'antécédents
((noise from the corridor))
486. 01:50 DOC Ouais
corridor))
488. 01:51 GYN ((incomprehensible))((noise from the
corridor))
489. 01:52 GYN tu rajoutes ça?
490. 01:58 DOC ouias, mets-en ranitidine 150 ((noise
from the corridor))
491. 01:59 DOC ((clears throat))
492. 01:59 GYN combien pour l'instant?
493. 02:01 DOC tu lui mets une fois par jour
494. 02:01 ((noise from the corridor))
495. 02:04 DOC là, tu lui donnes
497. 02:07 GYN ouais.
498. 02:11 DOC [lleva la tableta ahora para los vómitos She will bring tablets against vomiting.
500. 02:15 DOC otra tableta para el estómago And another one for the stomach.
502. 02:20 DOC para que no le duele, ¿vale? To relief her from the pain, ok?
503. 02:29 GYN donc, là, ça va? Du mouvement, pas de
médicament. ça pour l'étourdissement et
ça pour l'estomac ((sound of medicines
being passed around; sound of opening
door))
504. 02:30 UNK ((incomprehensible)) numéro 4
505. 02:30 GYN Ouais
506. 02:31 DOC Ouais
Annex 7 243
Annex 7. Full transcript case study The patient with a kidney stone
DOC: doctor
PAT: patient (woman with high voice)
COM: companion (woman with low voice)
NUR: nurse
Part 1
1. DOC What happened?

<3.5>
2. COM He has got a problem there* with the kidney** problem. ((he points at the
patient’s lower back))
<2>
3. COM He has pain* here.
<2>
4. DOC Since when?
<1>
5. COM Euh...
…
6. COM Last four days//.
<2>
7. 00:00:17.3 DOC Is it the right* side?
..
8. 00:00:18.7 COM Euh… "rait na lepht ?” ((mix Urdu-English: right or left?))
<2>
9. 00:00:21.8 PAT “lepht" ((mix Urdu-English: left))
10. 00:00:23.8 COM = ((incomprehensible)) ((Urdu)
11. 00:00:28.8 COM = Left side//
<3>
12. 00:00:25.7 DOC Since four days?
13. 00:00:27.1 COM = = Yes//

<3>
15. 00:00:36.9 COM No/
…
Continuous//
<1>
16. 00:00:38.6 DOC Ask him
…
17. 00:00:39.6 COM Yes ((incomprehensible))
18. 00:00:40.6 DOC = Is it continuously* .. also at night or is it ((snaps fingers)) sometimes heavy
pain and afterwards no pain.
<1.5>
19. 00:00:49.2 COM Is it continuous or intervals ((in Urdu))
Annex 7 244
…
20. 00:00:52.6 PAT "thora thora chakkar aya tha" ((little little pain to start with, then it intensifies, I
am seeing stars == I feel dizzy))
21. 00:00:55.2 COM = ahh
<1>
Euh now he has no pain/
<1>
After some time/
…
He has a big* pain//.
<1>
22. 00:01:01.6 DOC Yes
(1)
My question is, the past four* days/
23. 00:01:04.9 COM = Yes
24. 00:01:05.7 DOC = Was the pain continuously?
25. 00:01:07.6 COM = Continuously
26. DOC = Or was it sometimes big* pain after which no* pain?
27. 00:01:11.7 COM = Sometimes big* pain sometimes, sometimes small* pain.
(4)
28. 00:01:15.9 COM Sometimes big* pain, but sometimes big* pain.
(1.5)
29. 00:01:20.0 DOC hmhm
30. 00:01:20.4 COM = Ahhhh
(5)
Euhh
(.)
32. 00:01:28.1 DOC did he have fever?
(1)
33. 00:01:30.0 COM no fever
(2)
35. 00:01:33.5 DOC = Did it came ((snaps fingers)) suddenly?
(.)
Or did it came little by little.
(.)
36. 00:01:38.7 COM Little by little
(2)
37. 00:01:41.7 DOC Ask* him//
…
38. 00:01:41.7 COM Yes eeehhhhe, he live with me, I know// ..
39. 00:01:45.0 COM He live* with me*
40. 00:01:46.1 DOC = = Yes yes yes
…
Ok
Annex 7 245
…
But he can have other feelings than you think*.. So you must*
…
Translate//
41. 00:01:53.3 COM = =Hmhm//
42. 00:01:53.3 DOC = =Did the pain came ((snaps fingers)) suddenly or.. did it came
…
first a little pain and afterwards more:: and afterwards still* more::.
43. 00:02:02.0 COM = "kabhi kabhi hota hai ya ahista ahista chalta hai” ((is it now and then or does
it come slowly slowly?))
…
44. 00:02:05.1 PAT ahista ahista kr kay chalta hai … ((it comes slowly slowly))
45. 00:02:04.7 COM = aha
47. 00:02:10.5 COM = solosolosolo ((slowly slowly slowly)) and then fast* pain
48. 00:02:12.6 DOC = Huh?
49. 00:02:15.9 COM his pain // solosolosolo
50. 00:02:18.3 DOC = solosolosolo?
= What does it?
51. 00:02:20.0 COM = Little little little//
52. DOC = = OK
<4>
53. 00:02:26.1 COM ((incomprehensible))((in Urdu))
<1,5>
55. 00:02:30.5 COM Euh
56. 00:02:32.1 DOC If he is in pain*.
…
Is
…
Is it better if he stays calm* or is it better that he moves* all the time.
<1.5>
57. COM ((starts in Urdu))-
58. DOC = ask him
59. 00:02:41.5 COM = if you sit down, is it okay then? ((In Urdu)) …
60. 00:02:44.0 PAT While seating, the pain becomes less. ((In Urdu))
…
61. 00:02:45.3 COM If you do like this, do you feel less pain? ((movement))((In Urdu))
62. 00:02:46.5 PAT = Yes ((in Urdu))
63. 00:02:46.0 COM ((Urdu))
(.)
64. 00:02:48.9 PAT sometimes it stops and sometimes it starts. ((in Urdu))
65. 00:02:52.2 COM = OK
Annex 7 246
67. 00:02:55.2 COM = He says he sit down <1> no pain// … But he moving* he still have pain
<7>
68. 00:03:06.7 DOC Is… is he in pain when he urinates*?
…
When he makes pipi*.
69. 00:03:12.2 COM No pipi is ok.
70. 00:03:13.9 DOC = When he makes pipi, is it painful?
(.)
71. 00:03:17.3 COM Euh when you urinate, do you feel pain? ((Urdu))
72. 00:03:19.2 PAT = no ((Urdu))
73. 00:03:19.2 COM = no
(3)
74. 00:03:21.3 DOC Is the urine with blood?
(1)
75. 00:03:24.1 COM euh, does the urine have blood? ((Urdu))
(4)
Did you undergo a urine blood test? ((Urdu))
(1)
76. 00:03:33.0 DOC Yes, my question is did he saw blood in the pipi?
77. 00:03:36.7 COM = No
78. 00:03:37.4 PAT [no]((probably understands it))
79. 00:03:38.4 COM = No
(5)
80. 00:03:42.4 DOC Did he vomit?
(1.5)
81. 00:03:44.5 COM (unclear to whom he is speaking). ((incomprehensible))
82. 00:03:47.1 PAT ((Urdu))
83. COM = EUEH before one hour.
(.)
84. 00:03:50.6 COM before four days euh first time he vomited (not clear)
(1)
He vomited he.
85. 00:03:55.3 DOC = yes
(1)
since the first vomit four days ago
(.)
86. 00:03:59.4 COM Yes
87. 00:03:59.4 DOC = Did he vomit still or not
(.)
88. 00:04:02.2 COM Not
(.)
nothing
(4.5)
89. 00:04:06.9 DOC Does he have diarrhea?
(1)
kaka?
Annex 7 247
(.)
90. 00:04:10.5 COM Euh
(.)
91. 00:04:10.6 DOC Liquid
92. 00:04:12.1 COM = Euh
93. 00:04:12.3 DOC = Liquid
(.)
94. 00:04:13.2 COM Liquid heh? ((no proper Urdu expression, but understandable given context))
(.)
toilet?
95. 00:04:14.1 PAT = he
(.)
96. 00:04:14.2 COM Is toilet ok? ((Urdu)
(.)
97. 00:04:15.2 PAT ((Urdu, incomprehensible))
98. 00:04:16.3 COM = ok ((in Urdu))
99. 00:04:16.7 PAT =((Urdu, incomprehensible))
100. 00:04:21.0 COM speaks Urdu
101. 00:04:21.1 PAT speaks Urdu
102. 00:04:24.3 COM = sometimes better
(.)
toilet
(1)
103. 00:04:25.9 DOC No no.
(1)
The past days did he have liquid kaka?
(2)
104. 00:04:31.7 COM Was toilet ok during the last few days? ((Urdu))
(1.5)
105. 00:04:34.5 PAT sometimes it is ok, sometimes not ((Urdu))
106. 00:04:36.7 COM = sometimes ((incomprehensible)), sometimes liquid
(2)
107. 00:04:38.9 DOC the
(.)
the past days also?
108. 00:04:42.2 COM = the past days also
(7.5)
109. 00:04:48.3 DOC Can he still eat? The past days or not?
(.)
110. 00:04:52.7 COM ((incomprehensible)) ((in Urdu))
(1)
111. 00:04:55.2 DOC He is still eating?
(1)
112. 00:04:56.5 COM eeuh no he euhh
(.)
Annex 7 248
last time he eat

(1)
113. 00:05:00.9 DOC Yes
(.)
114. 00:05:02.4 COM EUH
115. 00:05:03.4 DOC = This morning and noon he eat? Or not
116. 00:05:03.6 COM [no]thing Nothing a little
(1.5)
no coffee no tea
(4)
117. 00:05:12.1 PAT coughs
118. 00:05:10.3 DOC Is he coughing more? ((makes the gesture of coughing))
(1)
119. 00:05:13.0 COM No
(4.5)
120. 00:05:17.6 DOC Is it only?
(.)
Show me exactly where it hurts.
(.)
121. 00:05:21.7 COM Tell correctly where you have pain.((Urdu))
122. 00:05:24.3 PAT = It hurts here. ((Urdu))
123. 00:05:28.4 COM = ((incomprehensible)) ((Urdu))
(.)
124. 00:05:28.4 PAT Speaks Urdu
125. 00:05:34.8 COM Speaks Urdu
126. 00:05:34.8 DOC Ok
(.)
Are the testicles also painful?
(2.5)
Painful? (.)Here? ((points))
(1)
127. 00:05:41.5 COM Do you have pain below.((Urdu))
128. 00:05:43.1 PAT = no ((Urdu))
(9)
130. 00:05:51.5 DOC Is he taking medicaments?
(.)
Pills?
(2)
131. 00:05:55.0 COM ((starts speaking))
132. 00:05:55.0 DOC [Is] he taking pills
(.)
133. 00:05:57.7 COM Pills?
(.)
What pills?
(3)
Annex 7 249
134. 00:06:00.6 DOC My question is

(.)
Does
(.)
Euh
= is he normally taking medicaments.
135. COM = ((incomprehensible))
136. 00:06:12.4 PAT = ((incomprehensible))
137. 00:06:16.0 COM = hmm
((medication is taken out of plastic bag))
138. 00:06:16.0 PAT speaks urdu
139. 00:06:17.8 DOC [he always takes mofikol?]
(.)
Every day?
(1)
140. 00:06:22.6 COM Euhh ((incomprehensible)) ((Urdu))
(.)
141. 00:06:23.1 PAT ((incomprehensible)) ((Urdu))
142. 00:06:24.6 DOC = Only now or every day?
(.)
143. 00:06:26.2 COM Euuuh how many days? (( Urdu))
145. 00:06:29.5 COM = Two days
(1)
only two days.
(2)
146. 00:06:32.5 PAT = This one I am taking every day. ((Urdu))
147. 00:06:34.1 DOC [Buscopan?]
148. 00:06:35.1 COM [since two days?]((Urdu))
[Since?]((Urdu))
149. 00:06:36.0 PAT = yes ((Urdu))
150. 00:06:36.0 COM = every day two tablets ((to check)) ((Urdu))
152. 00:06:40.5 COM = ((incomprehensible)) ((Urdu))
153. 00:07:40.3 ((physical examination starts, patient is lying on his back))
Pain? ((deep palpation))
DOC (3)
Pain? ((deep palpation))
(.)
155. 00:07:45.7 DOC = yes?
((palPATion continues))
157. 00:08:22.7 DOC PAIN?
158. 00:08:50.9 PAT there is a little shock ((Urdu))
Annex 7 250
(.)((implies sometimes))
160. 00:08:56.9 COM = hmmm
(1)
161. 00:09:12.6 PAT ((incomprehensible))
162. 00:09:14.8 More
163. 00:09:16.9 Ok
164. COM HHMMM
165. 00:09:20.8 DOC ((Physical examination))
pain?
166. 00:09:28.0 DOC ((Treatment Diagnosis)
We’re gonna do a blood sample and scanner of the tummy
(.)
167. COM OK
(1)
168. 00:09:33.4 DOC scan, do you understand?
169. 00:09:35.0 COM = hmmhm
(.)
170. 00:09:35.0 DOC OK
(6)
171. 00:09:39.7 DOC Is he still in pain?
(1.5)
172. 00:09:24.2 COM are you feeling pain now?((in Urdu))
173. 00:09:43.5 PAT = no ((Urdu))
174. COM = ((incomprehensible))
(1)
175. 00:09:44.5 COM He does not have pain now.
(1)
176. 00:09:46.7 DOC No, it's ok?
(1)
177. 00:09:47.6 COM ((incomprehensible))
(1.5)
178. 00:09:49.5 DOC Did he ever have
(.)
euh kidney stones in the past?
(.)
Ask him.
(2)
179. 00:09:55.5 COM ((speaks Urdu))
(2)
180. PAT no ((not sure))
181. 00:09:58.4 COM = ((Urdu))(repeats question)
182. 00:09:59.8 PAT = ((Urdu))
183. 00:10:00.8 COM = NO before not
Annex 7 251
184. 00:10:01.9 DOC = OK

(1)
There we go
Part 2
185. COM ((Incomprehensible))

186. 00:00:08.0 DOC He didn't drink today?
187. 00:00:11.4 DOC Ask him
188. 00:00:13.1 COM Pani? Pani?
189. 00:00:16.4 PAT in Urdu or Punjabi
190. 00:00:16.4 COM No
191. 00:00:21.3 PAT in Urdu or Punjabi
192. 00:00:21.2 DOC Ok
193. 00:00:22.0 DOC does he have other illnesses in the past?
194. 00:00:29.7 COM Hm
195. 00:00:30.7 DOC = was he very ill before?
196. 00:00:32.7 COM in Urdu/Punjabi
197. 00:00:36.3 PAT in Urdu/Punjabi
198. 00:00:43.6 DOC eeuhm, did he have anything of the heart?
199. 00:00:47.9 COM No / ((and then in Urdu Punjabi)
200. 00:00:55.0 DOC he has a stone between the kidney and the bladder, ok?
201. 00:00:59.8 DOC There is a tract between the kidney and the bladder
202. 00:01:00.8 COM hm
203. 00:01:00.8 DOC and there is a stone blocking the tract.
204. 00:01:07.0 DOC And it is the stone who gives the heavy pain
205. 00:01:07.7 COM Hm
206. 00:01:07.7 DOC Ok?
207. 00:01:08.3 COM OK
208. 00:01:08.3 DOC we will do, to do an operation today or tomorrow
209. 00:01:14.6 COM Hm hm
210. 00:01:14.6 DOC To extract the stone OK?
211. 00:01:17.9 COM Ok
212. 00:01:19.5 DOC so, he will have to stay in the hospital (1) Ok?
213. 00:01:23.2 DOC Translate
214. 00:01:23.2 COM In Urdu/Punjabi
215. 00:01:45.7 PAT In Urdu/Punjabi
Annex 7 252
217. 00:01:49.2 DOC she he mustn’t eat now because we will make him sleep to do the operation. (1)
Ok?
218. 00:01:58.3 COM Ok, but one moment. No other solution without operation?
219. 00:02:01.6 DOC No
220. 00:02:12.0 COM incomp
221. 00:02:12.0 DOC the stone is blocking too much (1) ok? (1) if it is a little stone (1) who gives
some problems without at the kidneys ok?
222. 00:02:27.6 COM Hm hm
223. 00:02:29.5 DOC we don't care (1) Ok?
224. 00:02:23.9 DOC but the stone is too big
225. 00:02:26.4 COM Too big?
226. 00:02:26.4 DOC Yes (1) he can't pass spontaneously
228. 00:02:32.6 PAT Hm
230. 00:02:40.8 COM euh, doctor he is not fine ((incomprehensible)) (he said the operation is ok or
not?)
231. 00:02:57.0 COM Incomp
232. 00:02:57.0 COM He said me, he is ok or not
233. 00:03:04.1 DOC he is ok to make him sleep? To do operation? Ok?
239. 00:03:22.4 COM he said I call with my phone my parents in Pakistan. I told him. He said me
240. 00:03:25.2 DOC It's a little operation (1) Ok?
241. 00:03:28.9 COM Hh hm
242. 00:03:28.9 DOC it takes ten minutes
243. 00:03:31.1 COM ten minutes
244. 00:03:32.2 DOC we look in the in the Blatter - we look for the stone ((makes pinching gesture
with his fingers)) we pinch it - and we pull it out
245. 00:03:38.1 COM = Minor Minor op-
246. 00:03:39.6 DOC = minor operation. With// almost no risks (1) ok?
247. 00:03:44.5 COM Ok
248. 00:03:44.5 DOC very low risk, you can tell
249. 00:03:49.4 COM in Urdu / Punjabi
250. 00:03:56.8 PAT Low risk ((in Urdu/Punjabi))
Annex 7 253
251. 00:03:56.8 COM Aha

252. 00:03:56.8 DOC But a very little risk just to make him sleep
253. 00:04:03.6 COM Ha ((yes in Urdu))
254. 00:04:03.6 DOC but it/ there are very few complications OK?
255. 00:04:09.0 COM Ok
256. DOC It's a minor operation.
257. 00:04:10.2 COM minor operation
258. DOC If I would have it, I would not euhh I would not worry
259. 00:04:14.6 COM today or tomorrow?
260. 00:04:16.8 DOC we, we are discussing whether it would be today or tomorrow (1) it depends of
the programme of the surgeon
263. 00:04:33.5 PAT Hm hm
266. 00:04:38.5 COM now is (there) a problem or not?
267. 00:04:42.7 DOC Hm?
268. 00:04:42.7 COM Now? Without operation is problem?
269. 00:04:47.5 DOC We will keep him in the hospital (1) ok? For painkillers. and euh for general
management. (1) Ok? Hey stays anyway
270. 00:04:59.0 DOC The only question is or it this evening or tomorrow morning ok?
271. 00:05:06.3 DOC It is just to keep you informed I say, we are still discussing
272. 00:05:06.7 COM he stay here?
273. 00:05:09.1 DOC Yeah (2) he has to stay (1) ok?
274. 00:05:11.6 COM Ok
275. 00:05:15.1 COM in Urdu / Punjabi
276. PAT Huh?
Part 3
Context: The doctor has just made the necessary calls to organize an operation to remove the kidney stone. He
now meets the patient and his companion who have been waiting on a bench in the waiting hall of the ED.
277. 00:00:19.1 DOC We will do the operation tomorrow in the morning. So, he stays in the hospital.
So we can do we can do operation tomorrow morning. Ok?
278. 00:00:19.6 COM Tomorrow morning?
279. 00:00:20.5 DOC Yes.
280. 00:00:21.6 COM Euh what time?
Annex 7 254
281. 00:00:23.7 DOC I don't know, in the morning.

282. 00:00:24.6 COM In the morning.
283. 00:00:26.0 DOC Everything fine?
284. 00:00:26.6 COM ((in Urdu)) Euh. Is it ok ?
285. 00:00:28.8 COM ((in Pakistani Punjabi)) So, what should we say to him?
286. 00:00:30.7 COM ((in Urdu)) Qamar has to stay here.
287. 00:00:35.6 ((Patient says something incomprehensible in Urdu))
288. 00:00:38.2 COM He said euuh "tomorrow I come home?"
289. 00:00:40.6 DOC No no no, he stays in the hospital.
290. 00:00:50.1 COM ((in Urdu)) (4) that he (doctor) is saying that, no, today you have to stay here
but there is no risk. No risk.
291. 00:00:53.0 COM No risk? (2) Euh minus, he is afraid.
292. 00:00:57.6 DOC If it was me, ((snaps fingers)) they can do it now. No problem.
293. 00:01:00.3 PAT (in Pakistani Punjabi)) Yes, yes, I will tell in 10-15 minutes.
294. 00:01:00.1 COM Eh in Urdu
295. 00:01:01.6 PAT ((in Urdu: incomprehensible))
296. 00:01:04.1 COM Ok, he said "after ten minutes I tell - tell you".
297. 00:01:04.6 DOC Pardon?
298. 00:01:10.4 COM After ten minutes (…) he is. he is. is he (phoning?) Pakistan
((incomprehensible)) mobile
299. DOC Yes
300. 00:01:14.3 COM After ten minutes he tell you.
301. 00:01:21.9 DOC If he decides to home (..) this night he will come back. Because it will be too
painful. Ok?
302. 00:01:22.9 COM Yes, yes yes yes
303. 00:01:38.5 DOC It's useless to return home.(..) and if he would return home and he looks for
another he looks for another hospital afterwards. They will redo all the exams
and they will also say that he must stay in the hospital.
304. 00:01:39.0 COM //Yes//
305. 00:01:40.8 DOC //So// it's useless to refuse
306. 00:01:40.8 COM aaah
307. 00:01:42.3 DOC //you can discuss//
308. 00:01:53.6 COM ((in Urdu)) He (doctor) is saying if you go home you cannot come back here. If
you will go to another hospital they will examine again and you have to wait for
the reports and it will take more time.
309. 00:01:58.7 PAT ((in Urdu/Pakistani Punjabi)) Give me half an hour and I will call while sitting
here.
310. 00:02:05.3 COM If you. He said: "if you give me half an hour."
311. 00:02:06.4 DOC Ok
312. 00:02:07.2 COM He?
Annex 7 255
313. 00:02:13.4 DOC Yes, but don't understand he is hesitating so much because it's. It's the most
easy operation.
314. 00:02:14.4 COM ((incomprehensible)) no no no
315. 00:02:15.4 DOC Heh?
316. 00:02:19.5 COM Sir, listen to me. First time he is he stays. He is afraid.
317. 00:02:21.0 DOC Ok, ok. No problem.
318. 00:02:22.3 COM He is afraid.
319. DOC Tell him. It is the most easy operation that exists.
320. 00:02:29.2 DOC Ok? He shouldn't worry about it.
321. 00:02:34.4 COM ((in Urdu)) is very easy operation, very short operation, no problem at all.
322. 00:02:35.7 PAT ((in Urdu)) “here or here?” (left or right kidney)
323. 00:02:36.2 COM ((in Urdu)): Here and here
324. 00:02:39.5 DOC No no. (..) First he sleeps.
325. 00:02:39.9 COM Uhuh
326. 00:02:44.0 DOC Afterwards they put a little wire in the penis.
327. 00:02:44.0 COM Uh
328. 00:02:48.4 DOC And they go upward to seek for the little stone.
329. 00:02:48.4 COM Aaah
330. 00:02:49.9 DOC Ok?
331. 00:02:58.1 COM ((in Urdu)) First they will make unconscious then, from the urine way, they will
operate and through camera they will watch where the problem is and
afterwards they take out.
332. COM Ok I call you
Annex 8 256
Annex 8. Full transcript case study Tuberculosis and technology
Participants:
DOC: doctor
PAT: patient (woman with high voice)
COM: companion (woman with low voice)
PNE: pneumologist
UDR: multilingual software application “Universal Doctor”
Part A
1. 00:00:01 PAT C’est pas bienne. C'est comme on

marcher, c'est beaucoup fatiguée, c'est
pas bon respirer.
2. 00:00:10 DOC D'accord
3. 00:00:12 PAT Eh, c'est tousser avec sang. Ça c'est
tout rouge euh
4. 00:00:16 DOC Et ça fait longtemps?
5. 00:00:18 PAT C'est deux semaines.
6. 00:00:19 DOC Deux semaines.
7. 00:00:20 PAT Chaque chaque jour après [] et []
8. 00:00:25 DOC D'accord. Vous avez du mal pour
respirer?
9. 00:00:29 PAT Oui
10. DOC et pour souffler?/
11. PAT ça c'est ici mal // et derrière ici comme
ça
12. 00:00:33 DOC D'accord. D'accord. Et le douleur, c'est
quand vous respirez?
(.)
13. 00:00:40 PAT Le douleur?
14. 00:00:41 DOC Les douleurs, le mal.
15. 00:00:42 PAT Mal oui oui oui (patient and translator
speak at the same time)
16. 00:00:43 DOC C'est quand vous respirez?
17. 00:00:44 PAT Oui ((phone rings in background))
18. 00:00:45 DOC Quand vous respirez? ((phone rings in
background))
19. 00:00:46 PAT ((Polish: incomprehensible)) ((phone
rings in background))
20. 00:00:47 PAT Aussi aussi mal quand on pas respiré
((phone rings in background))
21. 00:00:51 DOC Excusez-moi ((DOC picks up phone))
Allô? (.) Uhu () Dis-moi oui. (.) Uhu.
Annex 8 257
(.) Oui. C'est pas euh. (.) Pas vraiment

oui. Si vous pouvez lui donner si tu
veux deux mesures de [name of
medication] comma ça peut soulager
mais euh je pense pas que ce soit eh
inquiétant quoi. (.) Oui franchement ça
va. Hmm.
22. 00:01:28 DOC Euh donc ça fait mal quand vous
respirez ?
23. 00:01:30 PAT Oui
24. 00:01:31 DOC Oui
25. 00:01:31 PAT Ah oui
26. 00:01:32 DOC Comme ça, ça fait mal?
27. 00:01:32 PAT Oui
28. 00:01:33 DOC D'accord
29. 00:01:33 PAT Oui et mal ici
30. DOC dans le dos
31. PAT Ici ici mal
32. DOC d'accord
33. 00:01:38 DOC et donc, vous toussez aussi beaucoup? And so you are also coughing a lot? Is
= = = C’est ça? that it ?
<1>
34. 00:01:40 PAT non pas beaucoup tousser/ No, not much coughing.
35. 00:01:42 DOC = = d'accord/ Okay
36. 00:01:42 PAT = = et tousser tout sec* And coughing all dry
37. 00:01:44 DOC = = sec Dry
38. 00:01:45 PAT = = Sec* ((confirming tone)) Dry
39. 00:01:46 DOC == quand vous toussez c'est sec ? When you cough, it is dry?
40. 00:01:46 PAT = = ouais Yes
41. 00:01:48 DOC = = d'accord/… Donc vous crachez* Ok. So you do not spit? There is no
pas? [il] n'y a pas de crachats? sputum? To spit? Do you understand
Cracher? [Vous] comprenez “cracher”? “to spit”?
42. 00:01:52 PAT Cacher*? To hide ?
43. 00:01:53 DOC CR**acher (with emphasis) To spit ?
<1>
44. 00:01:54 PAT …cracher To spit
45. DOC = = comme ça ((demonstrates how to Like this
spit))
46. PAT = =non No
47. 00:01:55 DOC Non, No
48. 00:01:56 PAT = = eh oui un peu, un peu des petits* Euh yes, a little, some small red pieces.
morceaux .. rouges/
Annex 8 258
49. 00:02:01 COM = ((Polish :))((incomprehensible))

50. 00:02:05 PAT = ah c'est aujourd'hui aussi c'est ave[c]- Ah it is today as well it is with
51. 00:02:08 COM = avec sang/ with blood
<1>
52. 00:02:05 DOC c'est que du sang?... ou c'est salive/ Is it only blood ? or is with blood with
avec du sang? saliva?
53. 00:02:08 COM = avec… =avec des .. With… with…
54. 00:02:08 PAT = avec des petits morceaux ((shows a With little pieces
napkin in which she has coughed up))
55. 00:02:10 DOC …d'accord/ Okay
56. 00:02:11 PAT = = ça c'est/ That’s…
57. DOC = o-
58. COM = très très rouge Very very red
59. 00:02:13 DOC ok. Est-ce que vous avez perdu du Ok. Have you lost weight, madam?
poids madame?
60. 00:02:15 DOC Vous avez perdu du poids?
61. 00:02:17 PAT eh perdu ..
62. 00:02:19 DOC est-ce que vous avez MAIGRI?
63. 00:02:21 DOC ça c'est moins?
64. 00:02:22 PAT Oui c'est un peu moins
65. 00:02:24 PAT sept kilos
66. 00:02:25 DOC vous avez perdu sept kilos?!
67. 00:02:26 PAT Oui
68. 00:02:27 PAT allez allez, d'abord
69. 00:02:29 DOC En combien de temps?
70. 00:02:30 PAT Euh .. ah non d'abord 97, maintenant
90. ((incomprehensible))
71. 00:02:36 DOC d'accord en un ans ?
((Incomprehensible))
72. 00:02:39 PAT Oui.
73. 00:02:41 DOC d'accord et euhm .. qu'est-ce que je
voulais dire? oui. Vous transpirez la
nuit? Vous comprenez ça, ou pas?
74. 00:02:45 PAT Euh eh Oui oui oui
75. DOC Vous avez chaud? Chaud?
76. PAT oui oui J'ai comment un peu
77. 00:02:48 DOC dans la nuit. Quand vous dormez.
78. 00:02:51 PAT Oui c'est très transpirer.
79. 00:02:53 DOC Ah bon
80. 00:02:53 PAT J'ai changé changé
Annex 8 259
81. 00:02:54 DOC changé les draps

82. 00:02:55 PAT Oui
83. 00:02:56 DOC D'accord. OK. Euhm.
Hmhmhmhmmm. Et l’appétit?
Manger? C'est normal?
84. 00:03:04 PAT Euhm… Non c'est un peu moins.
85. 00:03:07 DOC d'accord
86. 00:03:08 PAT Ouais
87. 00:03:09 DOC vous fumez? vous fumez beaucoup?
88. 00:03:10 PAT Non fume pas.
89. 00:03:11 DOC Rien?
90. 00:03:12 PAT Avant oui avant ou (with rising
intonation in last part “avANT”)
91. 00:03:14 DOC c'est fini. Vous avez pas de problème de
santé euh?
92. 00:03:17 PAT Euh oui c'est ici 2005 ça c'est opérer
pour reins
93. 00:03:24 DOC pour les reins?
94. PAT Si
95. DOC Oui, d'accord, oké
96. 00:03:27 PAT Reins seulement une pièce maintenant.
97. DOC Uhu (confirming)
98. 00:03:29 PAT Année passée eh pour eh pierres, deux
((sounds as: doux)) pierres
99. 00:03:35 DOC Uhu (DOC seems to start losing
patience)
100. 00:03:36 PAT Comme ça, ici eh
101. 00:03:38 DOC la vésicule? la vésicule? Ok d'accord,
un instant (PAT and COM mumbling in
background) on va, on va quand même,
on va quand même mettre des masques.
Euhn, d'accord? (sound of door..)
102. 00:03:47 DOC on va mettre des masques
103. 00:03:52 DOC eh facile
[Mumbling, sound of door]
104. 00:04:03 DOC Il est ? sentiment
105. PAT Voilà ((in background))
106. 00:04:09 RES He is going to put on masks
107. DO ((Incomprehensible mumbling))
tuberculose ((noise))
108. 00:04:32 NUR tu penses que c'est possible une
tuberculose? Ou?
Annex 8 260
109. 00:04:34 DOC Non mais il faut pas euh

110. 00:04:35 NUR Ou alors en cas ou ?
111. 00:04:36 DOC elle a tous les critères quoi
112. 00:04:38 DOC elle transpire la nuit
113. ((Incomprehensible)) ((mumbling))
((music ? phone ringing))
114. 00:05:33 COM/ On range pour les passants ?
PAT/
NUR
(?)
115. 00:05:35 DOC Donc ça c'est pour les euh..
116. 00:05:37 COM/ Oui ((noise)) d’accord ((noise))
PAT/
NUR
(?)
117. 00:05:40 RES ((everyone puts on a mask))
((incomprehensible mumbling))
Part B
116. 00:00:05 DOC on peut vous mettre un masque eh ?
117. 00:00:07 PAT/ Uhmm ((noise))
COM
118. DOC ça va?
119. 00:00:13 ((incomprehensible)) ((door slams))
120. 00:00:14 DOC Oui
121. 00:00:15 PAT/ On peut voir les radios après?
COM
122. 00:00:17 DOC Les radios après. D'accord? On va
continuer. Je vais l'examiner, d'abord
l 'examiner. Ça va?
123. 00:00:22 PAT/ Comme ça? (in the background: ça va,
COM ça va, ça va)
124. 00:00:22 PAT/ ((incomprehensible)) égale?
COM
125. 00:00:26 DOC Comme ça. Ça va?
((silence)) ((putting on the masks))
126. 00:00:37 DOC OK. asseyez-vous, hein. Donc elle ne
fume pas? Elle a arrêté de fumer?
127. 00:00:41 COM Non
128. DOC Elle a arreté ? (cross talk))
129. 00:00:42 COM Non non non.
(incomprehensible) elle a nerviaux ?
Annex 8 261
130. 00:00:45 DOC OK et euh est-ce que je voulais dire:

elle boit de l’alcool?
131. 00:00:50 COM Non
132. DOC Pas de l'alcool?
133. PAT Noooooooon
134. DOC OK
135. PAT Nooon
136. 00:00:53 DOC Vous habitez ici en Belgique?
137. 00:00:54 COM/ Oui
PAT
138. 00:00:55 DOC Vous travaillez?
139. 00:00:56 COM/ Oui
PAT
140. 00:00:57 DOC Quelle sorte de travaille?
141. 00:00:58 COM/ eh [kik/Quick]? service
PAT
142. 00:00:59 DOC D'accord. ((PAT and COM :
incomprehensible)) OK. ((PAT and
COM : incomprehensible)) ça va.
143. 00:01:04 COM Chaque jour travaille. C'est pas pour
rien hehehe
144. 00:01:06 DOC Ouais
145. PAT Hehehehe
146. 00:01:07 DOC Vous avez des allergies?
147. 00:01:09 DOC Vous comprenez? Allergies?
148. 00:01:10 PAT Allergies. Incomprehensible
Oui, j'ai allergies.
149. 00:01:15 COM/ Oui un peu ouais pour euh produit
PAT
150. 00:01:18 DOC produit d'accord
151. 00:01:20 COM/ Ouais
PAT
152. DOC OK et vous prenez des médicaments?
153. 00:01:23 COM/ Eh oui pour tension
PAT
154. 00:01:26 DOC OK. Vous avez les noms? Vous
connaisez les noms des médicaments?
155. 00:01:29 COM/ Eh ouais. ((incomprehensible))
PAT
156. 00:01:31 DOC ((incomprehensible)) Ah vous avez
déjà donné?
157. 00:01:32 COM/ Oui j'ai hehehe
PAT
Annex 8 262
158. 00:01:34 DOC OK. Je vais voir ça. Ça va?

159. 00:01:35 COM/ Uhu ça va. ((incomprehensible))
PAT
160. 00:01:39 DOC Mais je pense qu’elle l’a déjà donné.
Je vais aller voir, hein.
161. 00:01:41 COM Ah ouais.
162. 00:01:42 DOC ça va? Et là, vous avez mal
maintenant?
163. 00:01:44 PAT Oui mal eh
164. 00:01:45 DOC Où ça?
165. 00:01:45 PAT derrière mal comme ça, derrière
(wheezing)
166. 00:01:49 DOC OK, ça va. Derrière. Je vais voir ça.
Elle a déjà eu des opérations? Ça c’est
167. 00:01:56 COM Oui, ça c’est avant aux reins et année
passée aux pierres.
168. 00:02:03 DOC Et ça c’est ici en Belgique, ça?
169. 00:02:04 COM Oui
170. 00:02:04 DOC Oh la la. ((looks at scar))
171. 00:02:05 COM donc la Belgique
172. 00:02:06 DOC d’accord ((quietly))
173. 00:02:08 ((incomprehensible))
174. 00:02:11 DOC … compris
175. 00:02:12 COM elle a perdu une rein.
176. 00:02:15 DOC elle a plus des reins ?
177. 00:02:16 COM Oui
178. DOC hehe (laughter)
179. 00:02:17 COM seulement un
180. 00:02:18 COM normalement deux
181. COM oui
182. 00:02:19 DOC OK, ils ont enlevé un rein. D’accord.
OK. Je comprends. ((long silence))
183. 00:02:41 COM Beaucoup desproblèmes pour respirer
184. 00:02:45 PAT ((gasps))
185. 00:02:51 DOC ça va ? OK ? Ça brule ? Pas mal ici ?
186. 00:02:57 PAT Non
187. 00:02:57 DOC Non ? OK. Ça va. Et vous pouvez vous
ASseoir (high pitch) si vous voulez
188. 00:03:03 DOC Vous n’avez pas mail ? Non ?
Annex 8 263
189. 00:03:06 PAT EH []

190. 00:03:07 DOC Ici ?
191. 00:03:07 PAT Oui
192. 00:03:08 DOC Beaucoup ?
193. 00:03:08 PAT Oui un peux
194. 00:03:10 DOC Ici oui ?
195. 00:03:11 PAT Oui
196. 00:03:12 DOC Faire pipi, pour faire pipi ça fait pas
mal ? Ça brule pas ?
197. 00:03:15 PAT Non non non. Pipi c’est pas mal.
198. 00:03:17 DOC OK. Vous avez pas eu de diarrhée ?
199. 00:03:19 PAT pas d’infection non
200. 00:03:20 DOC Non. Vous avez pas vomit ?
201. 00:03:21 PAT Non
202. 00:03:23 DOC Vous avez pas eu de diarrhée ? Vous
comprenez ? Diarrhée ?
203. 00:03:25 PAT Diarrhée ? Non
204. 00:03:26 DOC Non ? OK.
205. 00:03:29 COM Polish: Are you puking ? (vulgar:
puking)
206. 00:03:32 PAT Ah c'était vomi. Non pas vomi.
207. 00:03:33 DOC OK? respirer ? (patient coughs)
208. 00:03:38 ((Silence))
209. 00:03:43 Polish: incomprehensible
210. 00:03:46 ((Silence))
211. 00:03:58 DOC OK. ça va. C'est bien. C'est bien.
((Silence))
212. 00:04:09 ((Phone ringing))
213. 00:04:15 DOC [picks up phone] : Allô ? Oui, oui.
Oui. Oui. OK. Euh. OK, ça va. Je vais
aller voir une radio.
214. PAT Oui.
215. DOC ça va ? Vous avez vu un médecin
ehh... à la maison ? C'est le médecin
qui vous a envoyé ici ?
216. 00:04:30 PAT //Non// non non non le médecin pas à
la maison. (incomp 'Je ne... ) médecin.
217. 00:04:41 DOC Vous l'avez vu ?
218. 00:04:41 COM Polish: Are you taking medication? /
Do you normally take medication ?
Annex 8 264
219. 00:04:48 COM pour euh elle a toujours, chaque jour

prendre quelque chose euh pour
respirer.
220. DOC Oui
221. PAT Pour respirer c'est deux fois inhalation
pour masque. Produit pour masque.
222. DOC Oui
223. COM Euh
224. 00:05:04 DOC Les aérosol ? Aérosol ? Non ?
225. 00:05:08 PAT J'ai oublié ça.
226. 00:05:09 PAT ((Polish)): I will do it myself.
227. 00:05:10 PAT Oui aerosol oui oui.
228. 00:05:11 DOC Aérosol ?
229. 00:05:12 PAT Oui. J'ai me acheté petit appareil.
230. 00:05:17 DOC Ah bon.
231. 00:05:18 PAT Ouais
232. 00:05:18 DOC D'accord. () Et vous avez déjà fait des
radios eh en dehors de l'hôpital ?
233. 00:05:24 PAT Mais non, seulement.
234. COM maintenant c'est
235. 00:05:26 DOC Et j.. Ah bon, jamais.
236. 00:05:27 PAT Jamais
237. COM Euh
238. 00:05:28 DOC Jamais fait de radio ?
239. 00:05:30 PAT ((coughs)) Avant .. //
240. COM ((Polish: incomprehensible))
241. PAT Non non. Je … 5 années … c'est peut-
être avant ou…
242. 00:05:35 DOC Mais. OK.
243. 00:05:36 PAT Comment comment ici
244. 00:05:38 DOC Mais il y a deux semaines vous avez
pas fait de radio eh ?
245. 00:05:39 PAT Non non non non non non
246. COM Non non non non non
247. 00:05:42 DOC ça va. ((background noise : chair
moving))
248. 00:05:44 PAT Euh.. Douze ((make sign with fingers))
ans avant ça c'est euh () scanner eh
249. 00:05:49 DOC Ah il y a deux ans
250. 00:05:50 PAT Deux ans oui oui
Annex 8 265
251. COM Deux ans incomprehensible

252. 00:05:52 PAT C'est maintenant date rendez-vous 21
253. 00:05:56 DOC Et scanner quoi ? Du venture ?
254. 00:05:58 PAT pour ici
255. 00:06:00 DOC Et pour quoi vous avez fait un
scanner ?
256. 00:06:02 PAT Euh () Docteur euh médecin euh
257. 00:06:04 DOC OK. On va voir ça. //D'accord.//
258. 00:06:06 PAT //Il dit:// 'toujours voir'. Il dit : 'que je
ne sais pas qu'est-ce qui se passe'.
259. 00:06:09 DOC C'est cela. Faites 'A'.
260. 00:06:11 PAT 'A'
261. 00:06:12 DOC OK. C'est rouge. Euh. OK.() Vous n'
avez plus de dents ?
262. 00:06:17 PAT Oui.
263. 00:06:18 DOC Ah. OK. Patient couhgs
264. 00:06:21 DOC ça va… Ok?. ça va… On va aller voir Right… okay? Right… We are
la radio et on va peut-être faire une going to look into your scan and we
prise de sang aussi eh// may take a blood sample as well.
265. 00:06:26 PAT = = oui Yes
266. DOC = = d'accord ? ok?
267. PAT = = d'accord. Ok
268. 00:06:28 DOC = =ça va ? …Donc vous allez cracher Are you okay? So you are going to
dans un petit pot* aussi eh//... ça va ? spit in a small cup too. Ok?
269. 00:06:31 PAT = =oui* ça va.// Yes, okay.
((DOC takes a specimen cup out of the
closet))
270. 00:06:32 DOC je vais vous donner un petit pot/ I’m going to give you a small cup.
((Rustle of paper + chair))
<2>
271. 00:06:36 DOC vous allez cracher* ? ((PAT says You are going to spit?
something incomprehensible in the
background, DOC mumbles
something))
272. 00:06:38 DOC Vous comprenez cracher* ? To spit?
273. 00:06:39 PAT …((Polish:)) Kaszel? Cough?
Non. Eh.
274. 00:06:41 DOC = =eh bien. Well
275. PAT = = pipi? To pee?
276. DOC = = Non/ No
277. COM = =Non/ No
278. 00:06:43 PAT = =kaka? Poo?
Annex 8 266
279. COM = = non No

280. 00:06:44 DOC = = on peut utiliser Universal Doctor ? Can we use Universal doctor?
Euh… Euh…
281. 00:06:49 COM = ((Polish:)) I think you have to spit.
282. 00:06:50 PAT = Ah ! C'est ... ((PAT makes a spiting Ah, it is…
gesture))
283. 00:06:52 COM = = Oui. Yes
284. 00:06:53 (.)((DOC and RES type out search
terms in Universal doctor))
<16>
285. 00:07:11 RES Cracher ? To spit?
286. 00:07:11 DOC Cracher. Oui. (.) ( mumbling op To spit. Yes.
achtergrond)
<10>
287. 00:07:29 DOC Tuberculose, ((gazes at tablet screen))
288. 00:07:31 DOC = = ah non (.) ( mumbling) Non
<24>
289. 00:07:57 DOC ((incomprehensible)) par système ?
290. 00:07:59 UDR = = ((Polish:)) tuberculosis
291. 00:08:00 PAT = = oh!*
292. 00:08:01 DOC = = non non No no
293. 00:08:03 PAT/COM = = (( anxious laughter ))
294. 00:08:05 DOC/ …Euh ça ne peut pas Euh, this cannot
RES ((incomprehensible)) … l'appareil …
Vas-y, vas-y... Go ahead, go ahead
295. 00:08:14 DOC ((DOC is mumbling))
296. 00:08:40 DOC Ah… oui… oui… Ah …yes….yes
((incomprehensible)) No, here, that one, this one maybe
Non… Ici… ça… Ici… Peut-être
que…
297. 00:08:45 UDR ((Polish:)) is your cough a dry cough?
<1,5>
298. 00:08:48 PAT Euh... Oui*. Euh… yes
299. 00:08:48 DOC = == oui ?.. Et alors… le suivant là/ Yes ? And so… the next one there
((PAT in the background -
incomprehensible))
300. 00:08:51 PAT = sec*… Tout sec Sec// Dry. It is all dry. Dry.
301. 00:08:53 DOC = = no,c elui là, celui-là*, celui-là*.. No, that one there, that one there,
ok… that one there
(( points at questions) on the tablet’s
screen while talking to RES while
<5>
302. 00:09:00 DOC Ça ne va pas? ((talking to RES while That doesn’t work?
Annex 8 267
<2>
303. 00:09:03 UDR Polish: Do you cough up phlegm when
you cough?
304. 00:09:07 PAT … euh. Oui./ Euh… yes
305. DOC = = ((Polish:)) Tak ? Yes ?
306. PAT = =Oui/.. Oui/. Yes, yes
307. 00:09:10 DOC = =Oui** ? Yes?
308. PAT = = oui/. Yes?
309. DOC Ah*! Ah!
310. 00:09:11 COM = = et avec sang// And with blood
311. 00:09:13 PAT = = Avec* sang* With blood
312. DOC = = Ouais, c'est ça// . Yes, that’s it
<17> ((talks to RES while looking at

tablet – incomp))
313. 00:09:33 UDR ((Polish:)) What is the phlegm like?
<1>
314. 00:09:37 PAT euh. C'est <1> comment,/ Euh… it is like…
<2>
315. 00:09:43 PAT ((Polish:)) it is solid It is solid
316. COM = ((Polish:)) which colour ? Which colour?
317. DOC = ((talks to RES while looking at tablet
– incomp))
318. 00:09:48 PAT ((Polish:)) it is solid. It is solid
319. 00:09:51 PAT = liquide avec... avec sang. Liquid with… with blood
320. 00:09:53 DOC = = avec sang, ok. With blood, okay.
321. 00:09:55 PAT = = liquid, liquid// Liquid, liquid
322. DOC = = oui, c'est ça. Ok.((talks to RES – Yes, that’s it. Ok.
incomp))
323. PAT Liquid, liquid// <3> Liquid, liquid
325. 00:10:02 PAT = =oui* Yes
326. COM = oui* Yes
<1>
<2>
328. 00:10:07 UDR ((Polish:)) whitish?
<1>
329. 00:10:09 PAT Non*, c'est rose// No, it is pink
330. 00:10:11 DOC = = rosé Pink
Annex 8 268
331. 00:10:11 PAT = = rosé Pink

332. 00:10:13 DOC = = ok Ok
333. Patiënt coughs () patient lauhgs
334. 00:10:17 DOC incomprehensible salive cracher
incomprehensible
335. 00:10:24 COM Polish: affixiated
336. 00:10:25 DOC incomprehensible évaporation
(incomprehensible) ça ce n'est pas
(incomprehensible) Vous voulez dire
(incomprehensible)
337. 00:10:34 PAT Coughs
338. 00:10:36 DOC Incomprehensible
339. 00:10:38 DOC OK. Alors, si on dit ça.
340. 00:10:42 UDR Polish: Is your cough a dry cough??
341. 00:10:45 DOC Tak ?
342. 00:10:45 PAT Tak.
343. 00:10:46 DOC OK. Mais sèche mais avec des
mucosités
344. 00:10:51 DOC OK, ça va.
345. 00:10:53 UDR Polish: Do you have more phlegm than
normal?
346. 00:10:57 PAT Tak.
347. 00:10:58 DOC Tak. OK. () ça va, eh ?

348. 00:11:10 DOC Eh, c'est bien.
349. 00:11:13 UDR Polish: Do you get short of breath
350. 00:11:16 PAT Tak((Polish)/ beaucoup//
351. DOC Oh oh. ((Incomprehensible))
352. 00:11:22 UDR Polish: Do you get short of breath
when you exert yourself??
353. 00:11:25 PAT Euh. Pas chaque fois. C'est // aussi eh//
354. 00:11:27 RES Si tu veux, je veux dire, eh.
355. 00:11:28 DOC OK. Non, ça je vois. J'ai pas besoin.
356. 00:11:30 PAT ça ça ne va pas
((incomprehensible))((coughs))
when you climb a few stairs or walk
100 metres?
358. 00:11:39 PAT Ah oui
359. 00:11:41 PAT 50 metres.
Annex 8 269
360. 00:11:42 DOC 50 metres.

361. 00:11:43 PAT Oui. Pas 200. Laughter
362. 00:11:45 DOC 200 ? ça c'est 100. Il a dit 100 en fait.
363. 00:11:48 PAT Ah 100 ?
364. 00:11:49 DOC Oui
365. PAT 100 metres
366. 00:11:50 UDR ((Polish:)) Has your shortness of
breath become worse in the last few
days??
367. 00:11:52 PAT Oui. Chaque jour plus.
368. 00:11:57 DOC Super
369. 00:11:58 UDR ((Polish:)) Did your shortness of
breath come …. ((truncated utterence))
370. 00:11:59 DOC Oh.
371. 00:12:00 UDR ((Polish:)) Did your shortness of
breath come on suddenly??
372. 00:12:01 RES C'est l'internet, eh
373. DOC Eh.
374. 00:12:03 COM Euh.
375. 00:12:04 DOC Non. Oui ?
376. 00:12:05 PAT Non, un peu un peu un peu.
377. 00:12:07 DOC OK. Progressive quoi. D'accord.
Super. Super. Super. ()
378. 00:12:12 DOC Oui ça on a, ça on a déjà dit, huh. ça
c'est oui, eh. Tak. D'accord.
379. 00:12:16 UDR ((Polish:)) Does your shortness of
breath get worse after exerting
yourself??
380. 00:12:20 DOC Tak. Tak.
381. PAT Oui. // Tak.
382. 00:12:23 ((Polish: incomprehensible))
383. 00:12:24 UDR ((Polish:)) Have you had wheezing in
your chest?
384. 00:12:25 PAT Euh. Oui un peu.
385. 00:12:27 DOC Un peu ?
386. 00:12:29 PAT Un peu. ((Polish: incomprehensible))
((Cough)) Oui. Oui.
387. 00:12:33 UDR ((Polish:)) When you are short of
breath, do you wheeze?
388. 00:12:38 PAT and COM talking in background
Annex 8 270
389. 00:12:39 DOC Siffle. Pas vraiment souffler ça.

Souffler c'est: ((whistles)) Non.
Comme ça.
390. 00:12:44 PAT Non non c'est ((incomprehensible)) eh
391. 00:12:48 DOC Je crois que ((incomprehensible))
traduction eh. Non, je crois que
((incomprehensible)) OK, ça va. C'est
super. ((incomprehensible mumbling)).
On peut jouer un peu eh
392. 00:13:02 UDR ((Polish:)) Do you have chest pain?
393. 00:13:05 ((silence))
394. 00:13:09 DOC Alors?
395. 00:13:10 UDR ((Polish:)) What is the pain like?
396. 00:13:12 DOC Euh. () ça va. ()
397. 00:13:18 UDR ((Polish:)) Tightening?
398. 00:13:21 PAT Oui oui. ()
399. 00:13:25 UDR ((Polish:)) Stabbing?
400. 00:13:27 PAT Euh
401. 00:13:28 DOC ((Polish:)) Nie? Tak? (Nee? Ja?)
402. 00:13:29 PAT Eh. Oui. Un peu.
403. 00:13:31 DOC Un peu?
404. 00:13:32 PAT Un peu. C'est pour euh à droite
405. DOC Huh.//
406. PAT et à gauche.
407. 00:13:36 ((incomprehensible))
408. 00:13:37 UDR ((Polish:)) Burning?
409. 00:13:38 PAT Oui. ((Polish:)) yes it is burning.
410. 00:13:40 DOC Ah bon. Tout tout le temps? cough
411. 00:13:45 UDR ((Polish:)) Have you felt nauseous or
vomited with the pain?
412. 00:13:49 PAT Non.
413. 00:13:49 DOC Non. OK. OK. ()
414. 00:13:53 UDR ((Polish:)) Have you been sweating a
lot with the pain?
415. 00:13:57 PAT Euh. Euh. dormir. Quand on dormir, ça
c'est
416. DOC OK.// transpirer.
417. 00:14:02 DOC Attends, ça c'est super important
encore.
418. 00:14:05 UDR ((Polish:)) Is your shortness of breath
worse when lying down than when
standing or sitting?
Annex 8 271
419. 00:14:12 PAT Euh. Oui. Jamais comme ça.

420. DOC Non.//
421. PAT comme ç[a] (last word is swallowed)
422. 00:14:16 DOC OK. ((mumbling in the background))
423. 00:14:22 COM magnifique tablet ((incomprehensible))
424. 00:14:28 RES Je pense que c'est la connection. Il faut
attendre un peu.
425. 00:14:30 DOC Ah il faut attendre?
426. 00:14:31 RES Oui, parce que c'est l'internet je pense.
427. 00:14:33 DOC Tu crois? Mais c'est rouge. En
fait,rouge je crois que ce n'est pas
encore fait.
428. 00:14:36 RES Ah.
429. 00:14:37 ((Incomprehensible))
430. DOC Oui./
431. RES Peut-être tu peux le mettre euh il faut
directe en demande
432. DOC Ouais/
433. RES ((incomprehensible)) comme
télélingue On peut le mettre // en ..
434. 00:14:42 UDR ((Polish:)) How many pillows do you
use?
435. 00:14:44 PAT Oh! Cinq?
436. 00:14:46 DOC Cinq
437. 00:14:46 PAT ((laughter)) Quatre. ((laughter)) //
((incomprehensible))//
438. 00:14:50 UDR ((Polish:)) // Have you started using
more pillows recently? //
439. 00:14:52 PAT Non.
440. DOC Nie? // OK. ((coughs))
441. 00:14:57 UDR ((Polish:)) Do your legs often swell
up?
442. 00:14:59 PAT Oui. ()
443. 00:15:02 UDR ((Polish:)) Is the swelling more than
usual?
444. 00:15:05 PAT Euh. Maintenant non.
445. 00:15:06 DOC OK. Ça va
446. 00:15:08 PAT C'est comment chaud et voyager pour
eh pour Pologne
447. DOC Oui. 2000 kilometres. Ça //c'est
complète comme ça. //
Annex 8 272
448. 00:15:14 DOC Ah oui oui oui. OK. Ça va. Oui, ça va.
D'accord. On peut recommencer, eh.
449. 00:15:19 PAT ((coughs)) C'est maintenant un peu.
450. 00:15:21 UDR ((Polish)) Have you got a fever?
451. 00:15:23 PAT Eh oui. Aujourd'hui c'est
((whispering))
452. 00:15:28 DOC Vous a pris quelques (le
thermomètre)?
453. 00:15:29 PAT Euh. Non. J'ai.
454. 00:15:31 COM Elle a euh
455. PAT ((Incomprehensible))
elle a ((incomprehensible)) une
madame faire controle. ((coughs))
456. 00:15:35 DOC ((incomprehensible)) Donc, ça c'est ça.
Ah oui ça va. On va le demander.
457. 00:15:44 UDR ((Polish:)) Have you been very tired or
fatigued lately?
458. 00:15:49 PAT Oui.
459. COM Oui.
460. 00:15:50 UDR ((Polish:)) Have you lost any weight?
461. 00:15:52 DOC Ah, on a déjà demander.
462. 00:15:53 PAT Oui. Un peu. ((incomprehensible))
((laughter))
463. 00:15:56 DOC Oui? //Oui.//
464. 00:15:57 PAT Oui. Oui. ((incomprehensible))//
465. 00:15:58 UDR ((Polish:)): How many kilos?
466. 00:15:59 DOC OK. Ça va.
467. 00:16:00 PAT Deux kilos.
468. 00:16:00 DOC Deux kilos.
469. 00:16:02 UDR ((Polish:)) Have you lost your
appetite?
470. 00:16:05 PAT Eh. Manger un peu moins.
471. 00:16:07 DOC Un peu moins, oui. D'accord. () OK.
Ça on a demandé. () OK. () Super.
C'est bien.
472. 00:16:16 PAT Oui.
473. 00:16:17 DOC Vous avez tout compris?
474. 00:16:18 PAT Oui. (laughter) Oui. //C'est… (DOC,
PAT, COM chatting) (laughter)//
475. 00:16:19 DOC // C'est bien. //
476. 00:16:20 COM ((incomprehensible)) super.
Annex 8 273
477. 00:16:22 DOC Vous comprenez pour le polonais.

C'est bien, eh.
478. 00:16:23 Laughter
479. 00:16:26 DOC 'Cracher' on s'est pas
((incomprehensible)) eh. Ça il faut
peut-être eh bien ajouter.
480. 00:16:29 RES Ah.
481. 00:16:30 DOC Donc traduction 'mucosité' et comment
dire comment 'Vous devez cracher' ou
je ne sais pas eh. ()
482. 00:16:34 DOC ((louder)) Donc 'cracher' euh.
483. 00:16:35 PAT Cracher ça c'est: coughs ici. AH! Ah!
Ça va.
484. 00:16:38 DOC OK?
485. 00:16:39 PAT OK.
486. 00:16:39 DOC Ça va? Quand vous pouvez eh. Quand
vous sentez que //ça vient de loin eh
Dcoughs//
487. 00:16:41 PAT // Oui. Oui. Ça va.//
488. 00:16:44 DOC Ce n'est pas la salive eh. Ça doit venir
d'ici eh un peu. //
((incomprehensible))//
489. 00:16:47 PAT ici uhu uhu // ((noise; COM talking in
background - incomprehensible))
490. 00:16:49 DOC Ça va?
491. 00:16:50 PAT Ça va.
492. DOC OK.
493. 00:16:51 COM Elle a ((incomprehensible)) et elle a
attendre ((incomprehensible))
494. 00:16:55 DOC Oui.
495. 00:16:55 COM Comme ((incomprehensible)) ici pour
le ((incomprehensible))
496. 00:16:58 DOC Oui.
497. 00:16:58 COM Et ici.
498. DOC C'est pas ((incomprehensible)).
499. COM ((incomprehensible)) elle a, elle a
500. DOC Ouais.
501. COM Elle a ((incomprehensible)) elle a pas
possible ((incomprehensible))
502. 00:17:06 DOC Oui. OK. Pas de problème. Ça va.
503. 00:17:09 COM Ça va.
504. 00:17:10 ((Cross-talk, coughing))
Annex 8 274
505. 00:17:15 DOC Quoi? Oui, oui. Restez là eh.

506. PAT Ah ça va.
507. DOC Ça va? Oui.
508. PAT Merci.
509. 00:17:19 DOC Super, eh?
510. 00:17:20 RES Oui. C'est euh. Et on a les deux eh.
Sans et avec le ((incomprehensible))
511. 00:17:23 DOC ((Incomprehensible))
512. 00:17:24 RES Oui. J'ai enregistré tout. () Euh. Je vais
voir si ensuite en anglais. (.) Euh.
Incomprehensible Tu pense que tu as
tout, toutes les infos que tu voulais ou
il y a encore des choses eh // de la
((incomprehensible)) pas obtenir? //
513. 00:17:39 DOC Non. // Je pense que avec le, l'outil
après, c'était beaucoup plus facile, en
fait. () Pour poser des questions plus
précises euh. Allez, vraiment au début
de, de l'anamnèse j'avais des trucs que
je n'ai même pas poser parce que je
savais pas, je savais pas si elle me
comprend pas, tu vois.
514. 00:17:54 RES Ce sont des choses que normalement
tu demander sans
515. DOC Oui, oui.
une barrière linguistique
516. 00:17:57 DOC Oui je pense que normalement
((mumbling))
517. 00:18:00 RES Donc la deuxième partie //
518. DOC ((Incomprehensible))
519. RES la deuxième euh la réanamnèse
520. DOC Oui.
521. RES avec l'outil tu as plus posé des
522. 00:18:05 DOC la chance// quand même eh non je je
((incomprehensible)) Oui. C'est plus
précis.
523. 00:18:13 RES Et eh en fait, le mot, la traduction
c'était quoi? ((Incomprehensible)) je le
note.
524. 00:18:16 DOC Quelle traduction? Le //eh ?//
525. 00:18:17 RES ((Incomprehensible))
526. 00:18:18 DOC Ah oui. 'Mucosité' eh
527. 00:18:21 RES Ça en était.
Annex 8 275
528. 00:18:21 DOC Enfin, ((incompréhensible)) dans le

dictionnaire, mais moi eh on n'utilise
pas ça couramment, quoi.
529. 00:18:25 RES Qu'est-ce que vous dites alors?
530. 00:18:26 DOC Expectoration.
531. 00:18:27 RES Expectoration. () OK.
532. 00:18:30 DOC Oui.
533. 00:18:31 RES Euh.
534. 00:18:32 DOC Avec ça?
535. 00:18:32 RES Oui oui oui. ()
536. 00:18:33 ((noise))
537. 00:18:41 RES OK. 'Mucosité' () Voilà.
Part C
538. 00:00:00 DOC ((machines beeping in the background))

On va faire le scanner eh.
539. 00:00:02 PAT Oui.
540. 00:00:03 DOC Vous comprenez?
541. 00:00:04 PAT Oui, scanner, oui.
542. 00:00:05 DOC D'accord. Donc eh il se passe quelques
chose aun niveau des poumons vous
devez répondre mais on ne sait pas
encore qu'est-ce que c'est, eh. D'accord
vous devez rester à l'hôpital.
543. 00:00:14 PAT Oui.
544. 00:00:14 DOC Vous allez aller chez docteur euh
545. 00:00:16 PAT Oui.
546. 00:00:16 DOC ((name of the doctor)) Ça va?
547. 00:00:18 PAT Ça va, oui.
548. 00:00:19 DOC Vous avez compris? Have you understood?
549. 00:00:20 PAT = = Compris// Understood.
<1>
550. 00:00:21 DOC D'accord. .. Je veux dire… euh/ Ok (.) I want to say… euh… (.)
<7>
551. 00:00:30 UDR ((Polish:)) Do you understand what
illness you have?
552. 00:00:33 PAT = = Ah* Ah.
553. 00:00:36 UDR ((Polish:)) We need to carry out different
diagnostic tests
<1>
Annex 8 276
554. 00:00:40 DOC Demain* Tomorrow

555. PAT = = D'accord. Ok
556. DOC = = D'accord? Ok?
557. PAT = = Oui*. Yes
558. 00:00:42 DOC = = Demain on fait l’echographie/ Tomorrow an echography will be taken.
559. 00:00:43 PAT = = Oui. Yes
560. 00:00:44 DOC <16> ((discussing with pneumologist in
background))
561. 00:00:59 DOC Ça va? ((incomprehensible discussion Ok?
between DOC and PNE)) ((to PNE)) Tu
as des questions à demander avant qu’il Do you have further questions before the
y a la nuit ? night falls?
562. 00:01:10 PNE = = ((to DOC) Oui, donc … sur les Well, yes… About any precedents of TB,
antécédents de tuberculose, des contacts any contact with TB…
avec tuberculose/
563. DOC = = (to PAT:) Dans la famille quelqu'un Is there anyone in your family who has
a eu la tuberculose ? had TB?
564. 00:01:17 PAT = = Euh. Oui. Oui. Ici. Oui// Euh yes, yes, here, yes
565. 00:01:19 DOC = = Elle n'a pas compris/ She has not understood it
566. 00:01:20 PNE Est-ce que, est-ce que vous avez de Have you, have you got family members
famille* qui a eu* la tuberculose ? Est- who have had TB? Do you know people
ce que vous connaissez des gens qui ont how have had TB?
eu la tuberculose ?
<2>
567. 00:01:30 DOC Attends, elle n'a pas compris// Wait, she has not understood.
<4>
568. 00:01:35 RES ((incomprehensible)) si tu tapes famille Maybe if you click “family” because…
peut-être parce que -
569. 00:01:39 UDR = ((Polish:)) Tuberculosis
570. 00:01:41 PAT = = Eh? Eh…
571. 00:01:42 DOC = = Vous comprenez ça? Do you understand this ?
572. PAT = = Oui. Yes
573. 00:01:43 DOC = = La famille? Your family?
<1>
574. 00:01:44 PAT Eh oui avant mon mother mother/// Euh yes, before, my mother, mother
575. 00:01:47 DOC …C'est ça*? That’s it?
576. PAT = = Euh/ Euh…
577. 00:01:49 PNE = Votre mère tuberculose ? Your mother TB?
578. 00:01:51 PAT …Ouais. Yes
579. 00:01:51 PNE = = mère* tuberculose. Yes? TB?
580. 00:01:52 DOC *UXĨOLFD ((Polish: reads from tablet)) TB
581. 00:01:53 PAT *UXĨOLFD* ma* mère// TB my mother
<1>
Annex 8 277
582. 00:01:55 DOC Ta mère/ Your mother

583. 00:01:56 PNE = = Sa mère/ Her mother
584. PAT = = Maman. Mom
585. PNE = = Elle a eu tuberculose ? Has she had TB?
<1,5>
586. 00:02:01 PAT Euh c’est avant ma mère, c'est moi, c'est Euh it is before my mother, it’s me, it’s
petit bébé… little baby
587. 00:02:06 PNE Elle a eu la tuberculose quand vous étiez She has had TB while you were a little
un petit bébé? baby?
588. 00:02:08 PAT = = Oui* oui//. Yes yes
589. 00:02:09 PNE = = D'accord… OK… Et vous? Vous* Ok. Ok. And you? Have you never had
n'avez jamais* eu la tuberculose? TB?
590. PAT = = jamais* Never
591. 00:02:14 PNE = = D'accord… OK.<3> Vous avez Ok. Ok. Have you quit smoking?
stoppé de fumer?
592. 00:02:20 PAT Non. Cinq ans incomprehensible
593. 00:02:22 PNE Cinq anées, oui. Ça c'est très bien.
594. 00:02:38 UDR Polish: For how many years were you a
regular smoker?
595. 00:02:40 PAT Oh. Trente ans.
596. 00:02:42 DOC Vous avez fumé trente ans. –
597. PAT Uhu. laughs
598. 00:02:45 PNE (Vous habitez avec des fumeurs?)
599. 00:02:46 PAT Oui.
Part D
597. 00:00:02 PAT cough

598. 00:00:03 DOC Ça va?
599. 00:00:27 PNE Bien respirer?
600. 00:00:28 PAT Oui. C'est maintenant, oui.
Incomprehensible
601. 00:00:31 door elkaar gepraat /
602. PNE De plus. De plus.//
603. 00:00:34 PNE Bien respirer Comme ça.
(.)
604. 00:00:55 DOC Oui très bien
605. 00:01:01 PAT Si. Si ((incomprehensible))
606. 00:01:02 DOC //C'est //la tète?
607. 00:01:03 PAT ((incomprehensible))
Annex 8 278
608. 00:01:04 RES Respirer avec la bouche ouverte eh.

609. DOC Oui.
610. RES Ça on, ça on n'a pas. ((PNE and PAT in
the background))
611. 00:01:06 PAT Ça c'est ((incomprehensible))
612. 00:01:07 RES ((incomprehensible))
613. 00:01:08 PAT Ici. ((incomprehensible))
614. 00:01:09 RES C'est bizarre, eh? Parce que
615. 00:01:11 ((cross-talk))
616. 00:01:12 DOC Tu veux dire ((incomprehensible cross-
talk))
617. 00:01:16 RES Respirer avec la bouche ouverte, eh? Sss
618. 00:01:20 DOC Si non tu sais quand même bien
619. 00:01:23 PAT ((coughs))
620. 00:01:29 PNE ((Incomprehensible)) respirer de
nouveau?
621. 00:01:30 ((sound of moving papers. Noise))
622. 00:01:52 DOC and PAT in background
623. 00:02:03 DOC Non, moi je, j'ai vu sur le scan eh (.)
624. 00:02:13 PNE trés bien. () On va vous euh transférer à
mon service pour vous observer
625. 00:02:23 PAT Oui. Ça va.
626. 00:02:24 PNE C’est important de mettre à le monitor.
627. 00:02:26 PAT Oui.
628. 00:02:27 PNE Pour bien suivre votre coeur.
629. 00:02:29 PAT Ah oui.
630. 00:02:30 PNE Parce qu’il me semble (qu’il) se trouve
au niveau de votre cœur
631. 00:02:33 PAT Ah, oui
632. 00:02:34 PNE Ça peut être dangereux. Maintenant ça
va encore.mais on va quand même te
suivre
633. PAT Uhu.
634. PNE et d'abord, et demain on va faire des
examens en plus de tout euh regarder un
peu de plus proche.
635. 00:02:48 PAT Ah oui.
636. 00:02:49 PNE Vous comprenez?
637. 00:02:50 PAT Eh oui.
Annex 8 279
638. 00:02:51 PNE D'accord. Vous pouvez vous alonger de

nouveau, si vous voulez. ()
639. 00:02:59 PAT ((coughs))
640. 00:03:02 DOC OK. ((incomprehensible)) Ça va?
641. 00:03:08 PAT Oui, ça va.
642. 00:03:09 DOC ((incomprehensible))
643. 00:03:12 PNE Oui. Très bien pour suivre
644. 00:03:14 DOC ((laughs))
645. 00:03:15 PNE On a //
646. DOC Qu'est-ce que ça veut dire?
647. PNE On a, on a deux qu'on a besoin.
648. 00:03:19 DOC Oui.
649. 00:03:19 PNE D'accord.
651. 00:03:23 PNE Oui oui.
653. 00:03:29 PNE Si moi on voit plutôt les passeurs pour
euh sans raison eh
654. 00:03:31 DOC ((laughter)) Ça va? C'est bon?
655. 00:03:37 PAT Bon
656. 00:03:38 DOC Au revoir eh.
657. 00:03:39 PAT //Au revoir //
658. PNE Au revoir//
659. PAT et merci pour,
660. PNE ((incomprehensible in background)) //
661. PAT merci pour m'aider.
View publication stats

The Dynamics of Miscommunication in

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Dynamics of Miscommunication in

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

What if the diagnosis was tongue-tied? Linguistic ethnography in a

Conference Paper · September 2014

Emergency Department Communication View project

The user has requested enhancement of the downloaded file.

Please cite this dissertation as: Cox, A. (2017) The dynamics of

To my beloved wife Anneleen

Table of Contents ........................................................................................................................ i

4.4.7 Misalignment of frames ..................................................................................... 38

6.2.3 Setting of the pain .............................................................................................. 78

9 Conclusion ..................................................................................................................... 159

Table 4.1: Integrative model of levels of analysis of miscommunication ............................... 31

Excerpt 6.1: Presenting complaint ........................................................................................... 71

In 1998, I first visited the Department of Applied Linguistics at ErasmushogeschoolBrussel

1.1 Motivation for this PhD

“Emergency medicine is largely a communicative activity, and medical

International migration is at an all-time high. The number of international migrants residing in

In general, the literature recommends the use of professional interpreters to overcome

1.2 The Emergency Department as a research setting

1.3 Aim and research questions

The study’s broader objective is to contribute to addressing problems of unequal access to

More specifically, my analysis considers multiparty interactions. So far, multiparty medical

This PhD research looks at a unique dataset of audio-recorded transcripts of medical

1.4 Structure of this PhD

2 Miscommunication in medical settings

2.1 A brief historic overview of research on doctor-patient communication

The consideration that miscommunication could be traced back to individual characteristics of

2.2 Language and social interaction research on doctor-patient communication

2.3 LSI research on language-discordant medical interactions

Other aspects of professional interpreter-mediated consultations have been studied by

2.4 Research on language barriers in the ED

Research that explores the microstructures of physician–patient ED interactions involving

2.5 Addressing language discordance in a medical setting

Appropriate organisational arrangements are crucial for an adequate provision of professional

Nevertheless, if a language-discordant interaction is mediated by a non-professional

According to Ho (2008), family interpreters do not encounter linguistic problems in every

3.1 Worldview and strategies of inquiry

“I suppose it is tempting, if the only tool you have is a hammer, to treat

A qualitative research approach is chosen. According to Creswell (2014, 2), “qualitative

Figure 3.1: The interconnection of worldviews, design, and research methods

Research Methods Research Designs

Source: Creswell (2014, 5)

3.2 Linguistic ethnography

3.3 Linking discourse to medical practice: an interdisciplinary approach

“Nothing underscores the importance of the [medical] history more than

information and displaying emotions such as affection (Ainsworth-Vaughn 2001). In this

“We can, however, make a beginning; we can adopt a more open

The medical perspective is strengthened by co-constructing the analysis (Candling and

3.4 A case study approach

A first prerequisite for the development of a framework to analyse miscommunication, is

4.1 Defining miscommunication

A misunderstanding is a specific type of miscommunication. It may emerge from the difficulty

Like miscommunication, misunderstandings may have different causes, outcomes, as well as

4.2 Developing a conceptual framework

4.3 Levels of miscommunication

Level II miscommunication refers to minor misunderstandings that are recognisable, either by

in the same, or in a later interaction. In the context of the analysis of misunderstandings in

Level V miscommunication focuses on those instances of miscommunication which relate to

Finally, Level VI miscommunication considers the potential of interactions to reinforce or

Table 4.1: Integrative model of levels of analysis of miscommunication

Level Characteristics Problem status Awareness level Repairability

6 Ideological framing of Participants Participants are Only through

4.4 Sources of miscommunication