UNIVERSITY OF THE 

WITWATERSRAND 

FACULTY OF HUMANITIES 

School of Literature, Language and Media 

Translation and Interpreting Studies 

THE IMPACT OF BLACK SOUTH AFRICAN 

ENGLISH SPEAKER’S ACCENT ON THE QUALITY 

OF INTERPRETATION 

By 

CONSTANTIN DÉSIRÉ LEBOGO 

Supervisor: Dr Natasha Parkins-Maliko 

Report submitted in partial fulfilment of the requirements for the award 

of Master of Arts in Translation by Course Work and Research Report  

(option: Interpreting) 

March 2023 
 
  



 ii 

DECLARATION 

 

I, the undersigned, hereby declare that the work contained in this 

dissertation is my own original work and that I have not previously, in its 

entirety or part, submitted it at any university for a degree. 

 

 

 

 

 

 

 

 

    10 August 2023 

Signature        Date 

 



 iii 

DEDICATION 

 

I dedicate this work to my late mother Mrs. Béatrice Ngabidi Bekada who 

left this world in June 2021. She was the wind beneath my wings and would 

have been proud to see me add this academic achievement to my record. 

 



 iv 

ACKNOWLEDGEMENTS 

I give thanks to God for gracing me with good health throughout the duration of my 

endeavours at the University of Witwatersrand. Secondly, I thank my children Godlove, 

Raphael, Jean Marie, Constantin and Reine for their constant support and patience. 

 

Profound gratitude goes to my supervisor, Dr Natasha Parkins-Maliko whose 

invaluable insight, patience and knowledge of the subject-matter steered me through 

this research and kept me believing in myself. I also want to thank Prof. Alice Leal, 

Head of the Translation and Interpreting Studies Department, for her tireless guidance.  

 

I thank the Chief of Staff of the Continental Secretariat of APRM, the African Peer 

Review Mechanism, Adv. Batlokoa Makong, who granted me access to the APRM 

official database of freelance interpreters for the purposes of this research project, other 

APRM colleagues, Sampson Osei and Caroline Akinyi, for their tremendous support at 

various stages of this project, as well as Mr. Kodi Barth for editing this work. 

 

I am particularly thankful to the interpreters who accepted to participate in the study, 

setting aside their commitments and submitting themselves to this tedious exercise. 

Without them, this project would not have materialised. In the same breath, I express 

sincere gratitude to Ms. Hilde Poulter who accepted to read and record the transcribed 

speech, and Mr. Antoine Nga Ossongo for dedicating his precious time for the 

evaluation of participants. 

 

In retrospect, I want to thank the lecturers at the Wits Language School, specifically 

Mr. Hans Mühler and Mr. Gene Matthey, who encouraged me to venture beyond the 

short courses and go for the full Master’s degree programme.  

 

Finally, I thank the staff in the Department of Translation and Interpreting Studies at 

the School of Literature, Language and Media (SLLM), who continually work 

tirelessly behind the scenes to ensure students complete their courses. To the readers 

and examiners involved in the continuous quality improvement of this study, I say 

thank you, too. 

 



 v 

ABSTRACT 

As one of the three cognitive activities of interpreting, listening comprehension is key 

for ensuring quality in interpretation (Kamel, 2015). Listening comprehension requires 

a deliberate effort of hearing and comprehending the produced utterance by a speaker 

(Gile, 1995), and can, therefore, be affected by several variables, including the 

speaker’s accent. 

 

Besides other difficulties involved in interpreting, this study focuses on challenges 

related to listening to and understanding an accented Black South African English 

(BSAE) speech. It is assumed that a heavily accented speech by a Black South African 

English speaker will pose a challenge to an interpreter with no or little exposure to the 

specific variation of BSAE which, consequently, might impact performance and 

quality.  

 

This study seeks to investigate if an accented BSAE, might affect the quality of 

interpretation, and if so, to what extent. The questions raised here are addressed by 

comparing the level of accuracy and completeness of interpretation provided by two 

groups of professional interpreters with French (first language) and English (second 

language) as their language combinations. Interpretations into French by 13 

participants of a video recorded speech, and the transcription of the same, read with a 

Standard British English accent were recorded. Thereafter, participants’ outputs were 

evaluated. This data including a questionnaire, was analysed using the mixed methods 

approach to test the hypothesis.  

 

The results suggest that the strong BSAE accent influenced the rendition of the 

message, as participants of the experimental group did not perform as well as those of 

the control group. The questionnaire elicited that accent and related factors, such as 

phonemics and prosody, represented the greatest challenge during the interpreting 

process.  The study finally presents the strategies proposed and used by participants to 

mitigate accent impact on the process of interpreting and the quality of the output. 

 

Key words: accent, Black South African English, quality of interpretation. 

 



 vi 

TABLE OF CONTENTS 

DECLARATION ....................................................................................................... ii 

DEDICATION ........................................................................................................ iii 

ACKNOWLEDGEMENTS ........................................................................................ iv 

ABSTRACT ............................................................................................................. v 

TABLE OF CONTENTS ............................................................................................ vi 

LIST OF TABLES ................................................................................................... viii 

LIST OF ABBREVIATIONS .......................................................................................ix 

CHAPTER 1: INTRODUCTION .................................................................................. 1 
Research problem ........................................................................................................ 1 
Research questions ...................................................................................................... 2 
Research aim ............................................................................................................... 2 
Research rationale ....................................................................................................... 3 

CHAPTER 2: LITERATURE REVIEW ....................................................................... 5 
Introduction ................................................................................................................ 5 
Definition, modes, and norms in interpreting ............................................................... 5 

2.1.1 Definition ................................................................................................................. 5 
2.1.2 Modes of interpreting .............................................................................................. 5 
2.1.3 Norms in interpreting .............................................................................................. 6 

Interpreting process ..................................................................................................... 9 
2.1.4 Listening comprehension ......................................................................................... 9 
2.1.5 Variables affecting listening comprehension ........................................................ 10 
2.1.6 Analysis or processing ............................................................................................ 13 
2.1.7 Production ............................................................................................................. 14 

South African linguistic landscape .............................................................................. 14 
2.1.8 English in South Africa ........................................................................................... 15 
2.1.9 BSAE roots and user profile ................................................................................... 15 
2.1.10 Is BSAE a distinct variety of English? ................................................................... 16 

Characteristics of BSAE accent .................................................................................... 18 
2.1.11 The origins of BSAE .............................................................................................. 18 
2.1.12 Variables of BSAE ................................................................................................. 19 

Quality in interpreting ................................................................................................ 25 
2.1.13 Assessing quality in interpreting: content vs form .............................................. 26 
2.1.14 Interpreting quality assessment criteria .............................................................. 28 

Conclusion ................................................................................................................. 30 

CHAPTER 3: METHODOLOGY AND THEORETICAL FRAMEWORK ........................... 31 
3.1 Introduction ......................................................................................................... 31 
3.2 Theoretical framework ......................................................................................... 31 
3.3 Methodology ....................................................................................................... 32 

3.3.1 Research method ................................................................................................... 33 
3.3.2 Participants ............................................................................................................ 34 
3.3.3 Sampling procedure ............................................................................................... 34 
3.3.4 Ethical considerations ............................................................................................ 35 
3.3.5 Data collection ....................................................................................................... 36 

3.4 Conclusion ........................................................................................................... 37 



 vii 

CHAPTER 4: DATA ANALYSIS AND FINDINGS ........................................................ 38 
4.1 Introduction ......................................................................................................... 38 
4.2 Interpretation output analysis .............................................................................. 38 
4.3 A comparative analysis of a transcribed excerpt .................................................... 43 
4.4 Analysis of responses to the questionnaire ........................................................... 59 

4.4.1 Participants’ profiles .............................................................................................. 59 
4.4.2 Participants’ perception of accents in the speeches ............................................. 60 

4.5 Strategies applied by participants ......................................................................... 64 
4.6 Findings and conclusions ...................................................................................... 65 

4.6.1 Findings .................................................................................................................. 65 
4.6.2 Conclusions ....................................................................................................... 67 

CHAPTER 5: CONCLUSION and recommendations ................................................ 68 
Introduction .............................................................................................................. 68 
5.1 Study limitations ............................................................................................ 68 

Participants sample size.................................................................................................. 68 
Data collection ................................................................................................................ 68 

5.2 Study replicability and recommendations ............................................................. 69 
5.3 Conclusion ........................................................................................................... 70 

REFERENCES ....................................................................................................... 71 

APPENDICES ....................................................................................................... 75 
Appendix 1: Ethics Clearance Certificate ..................................................................... 76 
Appendix 2: Survey questionnaire .............................................................................. 77 
Appendix 3: information Sheet ................................................................................... 80 
Appendix 4: Participant’s consent form ...................................................................... 82 
Appendix 5: Permission letter to use APRM official database of interpreters ............... 83 
Appendix 6: Transcription of source speech ................................................................ 84 
Appendix 7: Evaluation rubric..................................................................................... 88 

 



 viii 

LIST OF TABLES 

Table 1: Group A evaluation results ____________________________________________________ 40 
Table 2: Evaluator's breakdown and comments of Group A performance ______________________ 41 
Table 3: Group B evaluation results ____________________________________________________ 42 
Table 4: Evaluator's breakdown and comments of Group B performance ______________________ 43 
Table 5: Years of experience as a professional interpreter ___________________________________ 60 
Table 6: Level of education of the respondents ___________________________________________ 60 
Table 7: Years of experience in interpreting BSAE _________________________________________ 61 
Table 8: Extent of familiarity with the speaker's accent _____________________________________ 61 
Table 9: Level of difficulty understanding the speech _______________________________________ 61 
Table 10: The extent to which the speaker's accent affected respondents' output ________________ 62 
Table 11: General difficulties encountered while interpreting the speaker ______________________ 63 
Table 12: The extent to which PHONEMICS affected the speaker's accent ______________________ 63 
Table 13: The extent to which PROSODY affected the speaker's accent ________________________ 63 

 



 ix 

LIST OF ABBREVIATIONS 

AIIC  International Association of Conference Interpreters 

APRM  African Peer Review Mechanism 

BSAE   Black South African English 

CI  Consecutive Interpretation 

COGTA  Cooperative Governance and Traditional Affairs 

ESL  English as a second language 

EU  European Union 

RSI  Remote simultaneous interpreting 

SATI South African Translators’ Institute 

SI  Simultaneous Interpreting 

SPSS  Statistical Package of Social Sciences 

ST  Source text 

TT  Target text 

USA  United States of America 

 
 



 1 

CHAPTER 1: INTRODUCTION 

One of the consequences of the rise and expansion of the British Empire (1815-1914), 

and later, of globalisation is the English language spreading and becoming a global 

lingua franca. English is used among speakers whose first language is not English, but 

who have English as the only medium of communication (Tieber, 2017, p. 41). Beside 

Great Britain, USA, Australia, Canada, and New Zealand, where English – with other 

languages – is spoken as the mother tongue, where English is learned, studied, and used 

as a foreign language in a several countries, including India, China, etc. However, for 

most of the Commonwealth, English is considered a second language, which in most 

cases enjoys the status of an official language (Tieber, 2017). This is the case for 

African English-speaking countries, including South Africa (de Klerk, 1999, p. 311). 

 

The 2011 Census statistics presented by Brand South Africa, the “Official Custodian 

of South Africa Brand,” indicate that the black population in South Africa represents 

the majority of the country’s total population with 79.2%. South Africa has 11 official 

languages, including English, which is considered the lingua franca of the country, as 

it is spoken and understood almost by all at different levels (Brand South Africa, 2002). 

The 2011 Census also indicates that despite being a common language in the country, 

only 9.6% of the population in South Africa use English as their home language.  

 

Research has shown that as a foreign language or a second language, English in South 

Africa is impacted by the individual speaker’s mother tongue (e.g. IsiNdebele, 

IsiXhosa, IsiZulu, among others), and the phonetic system of the mother tongue often 

influences the speech of most speakers, including Black South Africans (Reis Esteves, 

2009, p. 4). This study assumes that when interpreted, this variation of English, also 

called “Black South African English” (BSAE) accent, constitutes a challenge to the 

interpreter and negatively affects interpretation quality, especially if it is not familiar to 

the interpreter.  

Research problem 

The interpreting process includes listening comprehension, analysis, and production 

phases. These three cognitive, complex, and intertwined activities contribute to quality 

interpretation (Kamel, 2015). However, the listening comprehension phase determines 



 2 

the success of the other two phases and of the entire interpreting process. Indeed, 

listening comprehension requires a deliberate effort of hearing and comprehending the 

produced utterance by a speaker (Gile, 1995). Therefore, listening comprehension can 

be affected by several variables, such as the speed at which the speech is delivered, its’ 

level of technicality, the density of the content, the quality of the sound, and the accent 

of the speaker. In this study, only the speaker’s accent will be investigated. 

 

Since English is a global lingua franca, various English accents exist due to linguistic 

transfers from the speakers’ mother tongue, and “a number of proficiency levels and 

intercultural differences.” Therefore, it becomes impossible for the interpreter using 

English only as a working language to prepare for all eventualities to the point of 

becoming familiar with all English variations and safe from any unpleasant surprises 

(Albl-Mikasa, 2014, pp. 811-812). Excluding any other difficulties in interpreting, this 

study will focus on challenges related to listening to and understanding a speaker with 

a strong BSAE accent. It is, therefore, assumed that a heavily accented speech by a 

BSAE speaker will pose a challenge to any interpreter with no or little exposure to that 

specific variation of English, which, consequently, may impact the quality of 

interpretation. 

Research questions 

“How does BSAE accent affect the performance of an interpreter with little or no 

exposure to such accent?” is the main question investigated in this study. 

The secondary research question pertains to the extent to which an accented BSAE 

speech affects the quality of interpretation. 

Research aim  

The main purpose of this study is to investigate a possible impact of BSAE speaker’s 

accent and the extent of such impact on interpretation quality. Therefore, an experiment 

is carried out to compare the levels of accuracy and completeness of the rendition of 

two groups – control and experimental – of English-to-French professional interpreters 

interpreting a speaker of BSAE to determine the impact the speaker’s accent has on the 

quality of their interpretation.  



 3 

Research rationale 

As a verbal communication process, interpreting involves a sender and a receiver of a 

message, the interpreter being the mediator or channel of communication. All these 

three actors can hail from distinct geographical, linguistic and cultural backgrounds. In 

a conference setting where interpretation services are provided, the speaker’s accent 

may be of no or minor importance to members of the audience who have little or no 

knowledge of the language spoken, their only source of information being the 

interpreter.  

On the other hand, the language produced by a speaker, especially its accent, is of 

paramount importance to the interpreter, especially if such accent is new to the 

interpreter, as he or she needs to understand and decode the message before recoding 

and transmitting it to the target audience. As the process of interpreting relies on three 

stages (listening comprehension, processing, and production), the interpreter must 

understand the speaker’s message to convey it appropriately to the target audience via 

the target language.  

 

In the first stage, listening comprehension can be affected by factors like speed of 

speech, sound quality of speech, and speaker accent, which can be mild or strong. 

Indeed, speaker accent has been shown to affect the listening comprehension stage of 

the interpreting process, and thus undermines the successful interpretation of the 

intended message. Lin, Chang and Kuo (2013) identify phonemics and prosody 

(including accent) as two important linguistic elements that constitute the message of a 

spoken text. They conclude that 

  

“phonemic deviations may […] give rise to misinterpretation at word level, 

[while] deviated prosody may […] fail to reflect the underlying syntactic 

structures and semantic functions within the source text (ST), resulting in 

miscommunication on the receiving end, i.e., misinterpretation by the 

interpreter” (Lin, et al., 2013, p. 31). 

 

Where miscommunication occurs due to accent challenges, undesirable consequences 

can also occur; for example, tension among speakers, improper decision-making, 



 4 

among others (Lin, et al., 2013). Scholars, including (Grabbi, 2010), (Durbán, 2018), 

(Mazzetti, 1999) (Lin, et al., 2013) have widely discussed the impact of accent on 

professional interpreting from both the client’s and the interpreter’s perspective. Others 

like (Klerk, 1999), (Lanham, 1996), and (Meierkord, 2005) have established the 

existence of a variety of South African “Englishes,” including BSAE. However, there 

seems to be an important gap: the lack of a comparative studies on how professional 

interpreters, for whom English is the second language and French the first working 

language, perceive and cope with BSAE accent. Filling this gap is what this study aims 

to achieve, through an empirical comparative study of French interpretation of an 

English speech by a Black South African speaker with a strong accent. 

  



 5 

CHAPTER 2: LITERATURE REVIEW 

Introduction 

Some definitions, modes and norms of interpretation are explored in this chapter, with 

a focus on the interpreting process and factors likely to influence the quality of 

interpretation, including accent. The chapter also examines the linguistic landscape in 

South Africa, and a few characteristics of BSAE accent. 

Definition, modes, and norms in interpreting 

2.1.1 Definition 

Nolan (2005) defines interpretation as the art of a language professional, called 

interpreter, who “helps speakers to discharge their duty to make themselves understood 

and helps listeners to satisfy their need to understand what is being said” (Nolan, 2005, 

p. 2).  The interpreter is called to listen to a spoken message delivered in the original 

language and to render it orally into another language, either simultaneously as the 

speaker proceeds or consecutively. 

According to Pöchhacker (2004), interpreting is a type of translational activity, which 

is characterized by its “immediacy” and occasionally performed for persons willing to 

undertake communication from one language to another and from one culture to a 

different culture. Based on the criteria of ephemeral utterance of the original message 

and its immediate rendition in the target language, “interpreting is a form of Translation 

in which a first and final rendition in another language is produced on the basis of a 

one-time presentation of an utterance in a source language” (Pöchhacker, 2004, p. 11). 

Besides the presentation and rendition dimensions, Pöchhacker (2004) completes his 

definition by adding the elements of meaning, effect, and culture. Interpreting, 

therefore, becomes an exercise consisting of producing utterances considered to be of 

a similar meaning or to have similar effects as a previous utterance made in another 

language and culture. 

Because it is about verbal or textual and interpersonal communication, interpreting is 

considered as an “interdiscipline” whose study requires a variety of approaches such as 

linguistics, psycholinguistics, sociocultural communication, etc. (Wang, 2018).  

2.1.2 Modes of interpreting 

The practice of interpretation is done following two main patterns referred to as 

interpreting modes: the consecutive and the simultaneous interpreting. 



 6 

Consecutive interpreting 

The consecutive mode of interpretation follows a two-step process where interpretation 

comes after the original speech by the speaker. This involves understanding the source 

speech and rendering it in a different language (Pöchhacker, 2011, p. 191). In 

consecutive interpreting, the interpreter must support his/her memory with notetaking 

when the statements are too long to memorize, which somehow facilitates the rendition 

of the message in the language of the audience. (Albl-Mikasa, 2002, p. 257). 

Simultaneous interpreting 

In his Dictionary of translation and interpreting, Mason (2018) defines simultaneous 

interpreting as a continuous verbal translation of a speech served to a group of people 

at a conference setting by a professional called interpreter speaking from a booth, while 

the audience is listening through headphones (Mason, 2018). From a cognitive 

perspective, the simultaneous interpreter “has to listen to and comprehend the input 

utterance in one language, keep it in working memory until it has been recoded and can 

be produced in the other language…, all of this at the same time” (Christoffels & de 

Groot, 2005, p. 3). The interpreting process, be it consecutive or simultaneous, involves 

three complex and overlapping cognitive activities (Kamel, 2015).  

 

Whether done simultaneously or consecutively, interpreting can be carried out in 

different settings, conditions and circumstances. This justifies the existence of different 

types of interpreting, including conference, court, dialogue, liaison, media, medical, 

community, public service, sign-language interpreting, and more. Each type of 

interpreting is done following a minimum set of rules, which can be referred to as 

norms. 

2.1.3 Norms in interpreting 

To determine or appreciate the quality of interpretation, it is important to take into 

consideration the concept of “norms,” which is a useful element in investigating quality 

in simultaneous interpreting (Garzone, 2002). No matter the language combination, 

certain norms are to be observed by the interpreter for their output to be of acceptable 

quality. Norms of translation and interpreting have been a subject of extensive studies.  

In general, norms are “the translation of general values or ideas shared by a community 

– as to what is right and wrong, adequate and inadequate – into performance 

instructions appropriate for and applicable to particular situations” (Garzone, 2015, p. 



 7 

281), citing Toury (2012: 63). It can, therefore, be understood that norms will vary 

depending on the context and the profession (Prunk & Setton, 2015, p. 274). In the field 

of interpreting, Antonini (2015) makes a distinction between professional and “non-

professional” interpreting, the latter referring to “interpreting and linguistic mediation 

activities performed by people who have had no formal training and who are often not 

remunerated for their work as interpreters” (Antonini, 2015, p. 277). Such activity is 

out of the scope of this study as all participating interpreters have had some formal 

training in the field of interpreting and live of their trade. 

Gile (1999) notes that in professional interpreting, “norms can be a tool for explaining 

the strategies interpreters deploy to address cognitive constraints and overload” 

(Garzone, 2015, p. 282). 

Examples of such norms are given by Harris (1989), including the “true interpreter” 

norm, or norm of the “honest spokesperson.” It is the equivalent of the Garzone’s 

fidelity or linguistic output norm, where the interpreter “reproduces the source speech 

accurately and completely, with no personal alterations” (Garzone, 2015, p. 282). 

Expanding on Jones’ (1998) view on the notion of fidelity, Koumba (2014, 18) states 

that: 

 

The conference interpreter must be able to provide an exact and faithful 

reproduction of the original discourse. Deviation from the letter of the original 

is permissible only if it enhances the audience’s understanding of the speaker’s 

meaning. Additional information should be provided only if it is indispensable 

to bridge cultural gaps referred above: it should in no way involve the 

interpreter’s adding their own point of view to that of the speaker (Koumba, 

2014, p. 18). 

 

In this study, fidelity is relevant as the rendition by participants of the original speech 

in the target text will be assessed to see if and how the speaker’s accent would have 

affected their listening comprehension.  

 

Another norm relating to output is that the interpreter is expected to use the first person 



 8 

as if they were the original speaker. Garzone (2015) adds that “users accept the 

inevitable “voice/personality dislocation” between interpreter and orator, together with 

the uncertainties, faults, and infelicities, which are tolerated in interpreted texts but 

would be less acceptable in written translation)” (Garzone, 2002, p. 282). 

 

Directionality is another norm which is prescribed and widely observed by some 

scholars and interpreting practitioners, especially in conference interpreting (Koumba, 

2014). By virtue of the norm of directionality, interpreters should “translate” into their 

first or main working language. It is worth noting that interpretation scholars and 

lecturers, and interpreting professionals have divided opinions as to the best direction 

to interpret into, one’s A or B language (Koumba, 2014). On the one hand, supporters 

of the traditional ideology believe that an interpreter must only interpret into his or her 

native language. Bartłomiejczyk (2004), Donovan (2002), and Chang & Schallert 

(2007) are of this view, which is also advocated by the International Association of 

Conference Interpreters (AIIC), where practitioners believe satisfactory target language 

interpretation can only be produced by an interpreter working in their first language.  

 

On the other hand, a new generation of interpreting practitioners are of the view that it 

has become difficult to observe the directionality norm in certain settings. According 

to Harris (1989) cited by (Garzone, 2002, p. 282), this precept is nowadays only applied 

in international organizations “and no longer systematically, due to the high number of 

language combinations to be covered, e.g., in EU institutions.” Koumba (2014) holds 

that interpreting into one’s B language is becoming more and more accepted, though 

“reluctantly,” not necessarily because of quality, but out of necessity as an option 

imposed by circumstances depending on the market situation, and also because of 

increased demand from conferences organizers to employ interpreters capable of 

working in both their A and B languages. 

 

The translational norms categorized by Toury in “preliminary norms,” “initial norms” 

and “operational norms,” and integrated by Garzone (2015) are also applicable to 

interpreting. Preliminary norms have to do with the decision to provide interpreting 

services for a certain event or encounter, and the choice of mode. The initial norm is 

part of the norms “governing performance and the choice between an approach aiming 

at ‘adequacy’ (i.e., leaning heavily towards the original), and one aiming at 



 9 

‘acceptability’, more relating to the target language and culture norms (Garzone, 2015, 

p. 282).” Operational norms concern the set of decisions or strategies adopted by the 

interpreter during the interpreting process (Koumba, 2014). They determine the link 

existing between the source speech and the resulting the translated text, “in terms of 

omissions, additions, and other manipulations (“matricial norms”) and, above all, the 

actual formulation of the target text (“textual-linguistic norms”)” (Garzone, 2015, p. 

282). To a certain extent, norms are considered a tool used to explain strategies 

deployed by interpreters to overcome cognitive constraints and saturation (Garzone, 

2015) citing Gile (1999). They are useful in investigating quality in simultaneous 

interpreting according to Garzone (2002). 

 

Investigating the concept of norms into institutional interpretation, (Ndirangu, 2016) 

concluded that interpreters working for institutions must apply the same rules used to 

guide, control and regulate the practice of interpretation called norms of interpretation 

as promoted by AIIC. Sy (2018) notes, following a study of interpreting norms within 

the context of the Islamic Development Bank, that interpreters were not subjected to 

specific norms in interpreting Islamic finance. Therefore, she concludes that “The 

concept of norms and institutional interpreting norms is obviously a vague concept to 

the different stakeholders, although it is embedded in their expectancy of quality and 

accuracy of the output” (Sy, 2018, p. 105). 

 

No matter the type of interpreting or the norms applied, the process of interpreting 

includes listening comprehension, analysis (with note taking in the case of consecutive 

interpreting) or storing in short-term memory (in simultaneous interpreting), and 

production phases. 

Interpreting process 

2.1.4 Listening comprehension 

Also referred to as “listening and analysis effort model” by Gile (1995) in his Effort 

Model, listening comprehension is one of the cognitive efforts required in all modes of 

interpreting (consecutive and simultaneous), and in sight translation. It involves 

understanding the speaker’s original message and immediately analysing its content.  

The listening phase includes the conscious effort of hearing and comprehending the 

produced utterance by a speaker. For example, if a speaker is fast in presenting a 



 10 

specialized or technical topic with a lot of data, or if the sound quality is poor or they 

are using an unusual accent, then the interpreter will have to put more effort in listening 

and understanding the message. Moreover, in the case of consecutive interpreting, the 

interpreter, in this same phase, must take notes to help him or her remember the 

message (Kamel, 2015). 

2.1.5 Variables affecting listening comprehension 

In his Dictionary of translation and interpretation Mason (2018) presents the three 

modes of listening, which an interpreter may apply from time to time. These include 

listening as a reporter (to memorize and repeat what one has heard), listening as a 

recapitulatory (to subsequently give an account of the story in one’s words), and 

listening as a responder (to reply after listening using one’s own terms). To complete 

these functions, the interpreter should not only pay attention to the speaker, but also 

needs to demonstrate language fluency and proficiency, as well as a solid general 

knowledge, because the knowledge of two languages is the minimum prerequisite for 

the development of interpreting skills (Pöchhacker, 2004, p. 167). This means that the 

interpreter must master both the language used by the speaker and the target language.  

Listening comprehension may be impacted by several input variables, such as speech 

speed, information density of the message, sound quality, and accent. All these 

variables, which are external to and difficult to control by the interpreter can affect his 

or her performance and undermine the quality of interpretation. However, in the context 

of this study, after a brief overview of the other variables, accent (foreign accent in 

particular) will be discussed in more depth.  

Speech speed 

According to Pöchhacker (2011), delivery rate of the source speech is one of the factors 

that influence interpreter performance the most. “First and foremost among the input 

factors that may jeopardize professional performance are the rate and mode of delivery 

of the original speech” (Pöchhacker, 2011, p. 196). He recalls that 100 to 120 words 

per minute is the ideal speech delivery rate recommended by AIIC in the mid-1960’s 

and confirmed by subsequent empirical studies for simultaneous interpreting. Beyond 

that speed, the interpreter’s ear-voice span and pausing begin to increase while the level 

of accuracy of interpretation decreases (Pöchhacker, 2011). 

In an analysis of the nexus between speech delivery speed and strategy use, (Anyele, 

2014) found that the choice of coping strategies by an interpreter is always determined 



 11 

by the delivery speed of the speaker. She notes that most identified coping strategies 

are used for all delivery speeds (fast, average, slow) at various frequencies, and that 

speech speed is not the only variable that determines the use of such strategies. Citing 

Pöchhacker and Shlesinger 2002, Anyele (2014) further classifies interpreting 

strategies in two categories: meaning-based and form-based strategies, with a focus on 

strategies aimed at coping with speech delivery speed. 

Information density 

The density and complexity of information contained in the source message is one of 

the stressors in interpreting, and one that is likely to hinder comprehension and 

ultimately affect interpreter performance. Accounting for density and complexity are 

such linguistic parameters as lexical level, lexical density (sentence length and word 

length), numbers, proper names, cultural terms, creative language and humour 

(Pöchhacker, 2004, p. 139). Interpreting numbers, for example, is considered “a 

common and complex problem trigger” and one of the variables responsible for errors 

in simultaneous interpreting (Frittella, 2019, p. 79). 

Sound quality 

In consecutive and simultaneous interpreting, both of which deal with spoken language, 

sound quality appears to be a critical determining factor for the success of interpreting 

(Pöchhacker, 2004). While sound challenges can be easily mitigated in consecutive 

interpreting when performed face-to-face with no technical equipment involved, the 

issue may be very alarming for the interpreter in simultaneous mode where electro-

acoustic transmission systems are used. Quality in interpreting has become an acute 

issue in the recent past with the use of teleconferencing and remote simultaneous 

interpreting. 

Accent 

Simply put, accent is how an individual sounds when speaking a given language. In 

linguistics and sociolinguistics, accent is a way of pronunciation specific to a particular 

region, ethnic or social group (Cheung, 2015). Durbán (2018: 7) identifies “two types 

of accents: the foreign accent, which is predominant in second language learners (L2), 

and the native accent - the way a group of people or community speak their native 

language, whether it be because of geographic location or social status” (Durbán, 2018, 

p. 7). This position agrees with Cheung (2015) who argues that the risk posed by non-

native accents or unfamiliar native accent to interpreting is a consequence of the 



 12 

generalized use of English at international conferences. 

 

According to Grabbi (2010: 9), “linguistic accent” differs from foreign accent. She 

supports that “linguistic accent is the prosodic phenomenon that sets certain syllables 

apart from the rest of the word and certain words apart from the rest of the sentence 

using emphasis” (Grabbi, 2010, p. 9). This definition brings out the aspects of 

phonemes and prosody in determining accent. This is concurred by Lin, Chang & Kuo 

(2013), who assert that accent in general, and foreign accent, in particular, involves 

both phonemic deviations and prosodic deviations. 

 

While she agrees with Durbán (2018) that foreign accent is the way people pronounce 

sounds in a given language, Grabbi (2010) further holds that foreign accents of a 

language differ from the native version of that language due to the unique linguistic 

system (phonetics, phonology and prosody) they use, which differ from that of the 

native language. This means that a foreign accent does not necessarily refer to a 

different country, but to a different linguistic system (Grabbi, 2010). 

 

As speech sound recognition de pends on the prior linguistic competence of the listener, 

any deviation from acoustic and phonetic patters that are familiar to the interpreter is 

likely to make listening comprehension more difficult for them (Pöchhacker, 2004). 

Lin, et al. (2013) are more exhaustive when analysing the nexus between non-native 

English and listening comprehension. In their empirical study examining which 

between phonemics and prosody (both being a part of accent) hinders listening 

comprehension the most, they conclude that “both phonemics and prosody were 

significant in deteriorating accuracy, but prosody had a stronger effect…, and when 

prosody no longer reflects sentence structure and intended focuses, interpreters may 

need to dedicate more effort to parsing and finding out messages from context” (Lin, 

et al., 2013, p. 41) in reference to Daniel Gile’s effort model.  

  

This conclusion corroborates Mazzetti’s (1999) findings that both segmental 

(phonemics) and supra-segmental (prosody) elements of speech can pose significant 

challenges to interpreters if presented with deviations. In her empirical study, Mazzetti 

(1999) investigated if and the extent to which source-text deviations can possibly affect 

listening comprehension as well as the quality of the ensuing interpretation. Mazzeti 



 13 

indeed identified unusual speaker accent as the factor that stresses interpreters the most 

in source text presentation (Mazzetti, 1999). Her study was on a speech of a non-native 

speaker. The impacts of segmental and prosodic deviations were measured by assessing 

the semantic accuracy of the rendition by the interpreter in the target language after a 

simultaneous interpreting exercise. It is important, therefore, to investigate further the 

impact of deviated phonemics and prosody on listening comprehension. Foreign accent 

seems to be one of the interpreter’s nightmares when English is used as the lingua 

franca at international conferences. Albl-Mikasa (2013) notes that “challenges that 

interpreters face when they interpret non-native speakers include having to grasp 

foreign accents and recover unfamiliar expressions, […] resolve unorthodox syntactic 

structures and compensate for the lack of pragmatic fluency” (Tieber, 2017, p. 41). 

Durbán (2018) is persuaded that when non-native speakers use a language, they tend to 

transfer into it features of their native language, including phonemes, stress, rhythm, 

and intonation. This is what Durbán (2018) refers to as “interlinguistic influences”, 

whereby an “interlanguage” becomes an idiolect developed by a second language 

learner while maintaining features of their mother tongue, which they use to speak or 

to write (Durbán, 2018, p. 11). 

  

Speakers using an acquired (foreign) language may carry over the phonetic 

patterns of their native language, giving rise to a non-native or “foreign” accent, 

which is often understood to involve not only pronunciation (i.e. phonetic 

substitutions, deletions and distortions), but also non-native stress, rhythm and 

intonation (Cheung, 2015, p. 32). 

 

However, it takes some more learning and mastery of the acquired language to refine 

such interlanguage features (Durbán, 2018).  

2.1.6 Analysis or processing 

The analysis phase is also termed by Gile (1995) the “Memory effort” (short-term 

memory for SI and medium-term memory for CI). During this stage, the interpreter 

undertakes a deliberate mental effort to decode the source message, putting into play 

all linguistic and para-linguistic resources available to him or her. This is where 



 14 

interpreting becomes a communicative process. 

 

At the stage of listening and analysis, the interpreter needs to not only absorb 

what the speaker has said (verbal information), but also perceive how the 

speaker has said it when making sense of the source speech. Through aural 

perception, the interpreter needs to make sense of the pauses, stress, intonation, 

speed, prosody, articulation, fluency, or hesitation in the speech (paraverbal 

information) (Wang, 2018, p. 155). 

2.1.7 Production 

Production or rendition of the received message into the target language is the ultimate 

phase of the interpreting process. This is the mental effort made by an interpreter to re-

express the content and intent of the message received in one language (source 

language) into another language (target language). Obviously, this last stage is very 

much dependent on the first two phases, mostly the first phase involving hearing and 

understanding the source message. In fact, in the event that the initial message is only 

partially perceived due to one of the listening conditions presented as the determining 

factors of the quality of interpretation – density of information, speed of speech, quality 

of sound (Christoffels & de Groot, 2005, p. 23), and accent –, one can expect a poor 

quality of interpreting output. 

 

According to Daniel Gile’s “tightrope hypothesis” or effort model, the total amount of 

the capacity required to process a message must be equal to, or less than the cognitive 

capacity available to the interpreter (Pöchhacker, 2011, p. 192). In other words, the 

available processing capacity or energy, which is 100% for each interpreter, should be 

distributed among the three efforts (four if the “coordination effort” is taken into 

account); i.e. 30% for listening comprehension, 30% for analysis or processing, 30% 

for production and 10% for output control or coordination (Gile, 2002).  

South African linguistic landscape 

South Africa Constitution of 1997 recognizes 12 official languages in the country. 

These are home languages to the population and include Afrikaans (13.5%), English 

(9.6%), Ndebele (2.1%), Sotho (7.6%), Sotho sa Leboa (9.1%), Swati (2.5%), Tsonga 



 15 

(4.5%), Tswana (8.0%), Venda (2.4%), Xhosa (16%), and Zulu (22.7%) (Brand South 

Africa, 2002). On 3rd May 2023, the National Assembly of South Africa approved an 

amendment of the Constitution of 1996 to include South African Sign Language 

(SASL) as the 12th official language. 

 

In fact, most South African nationals can speak more than one official language, 

including English, which is commonly used in official and commercial public life. Zulu 

appears to be the country’s other lingua franca. In addition to the official languages, 

other languages mentioned in the Constitution and spoken in South Africa include, 

Arabic, German, Greek, Gujarati, Hebrews, Hindi, Khoi, Nama, Portuguese, San, 

Sanskrit, Sign-language, Tamil, Telegu, and Urdu. These languages are spread across 

the country following linguistic groupings and clans. 

2.1.8 English in South Africa 

Meierkord (2005) describes English as a national lingua franca in South Africa where 

it is used as a home language, a second language, a language-shift variety, and a foreign 

language. Meierkord holds that though English is spoken as a mother tongue mainly by 

the descendants of British settlers, it has become the first language for other local 

communities, such as Black and coloured South Africans, for whom English is now 

considered a second language. English in South Africa is either learned at school or 

acquired through informal contacts with different speakers.  

The 2011 census revealed that English is spoken by approximately 4,892,623 South 

African citizens, that is, 9.6% of the population. These figures, however, do not 

represent the number of English mother tongue speakers, but include the Blacks and 

Coloured families whose languages have shifted over time from their native African 

languages and from Afrikaans (Meierkord, 2005). 

2.1.9 BSAE roots and user profile 

De Klerk (1999) traces the roots of English, and BSAE in particular, as far back as the 

early 1800s as a result of colonialism, when the language was imposed on the local 

populations who already had their indigenous languages. English thus became desirable 

because of the powers (military, economic and cultural) it conferred to the colonial 

masters, and because it gave access to social and economic advancement to the 

colonized local people. Later, the Bantu Education Act of 1953 under the Apartheid 

system made matters even worse. By virtue of this system, Black people were to receive 



 16 

education in their mother tongues, and did not have access to native English, save in a 

few missionary schools (de Klerk, 1999). So, despite the undeniable benefits of 

acquiring early literacy in one’s mother tongue, the forbidden status of English made it 

even more desirable as it was considered by many to guaranty social and economic 

success in society, unlike indigenous languages that were regarded as useless. 

Unfortunately, the change in the education policy later after the 1976 Soweto uprising 

and the ensuing increased access to English did not prevent BSAE from suffering a 

negative impact in its development. Thus, de Klerk (1999: 312) notes that: 

 

[T]he long-term effects of under-funding, overcrowding and teacher 

incompetence, combined with limited contact with native English speakers, led 

to characteristic patterns of pronunciation and syntax becoming entrenched as 

norms of spoken BSAE, with resultant reduced levels of comprehensibility (de 

Klerk, 1999, p. 312). 

2.1.10 Is BSAE a distinct variety of English? 

Meierkord (2005) considers BSAE as the variety of English used by Black South 

Africans. She holds that this variety of English, however, is heterogeneous and varies 

based on the mother tongue and competence of the speaker. It is “the variety of English 

commonly used by mother-tongue speakers of South Africa’s indigenous African 

languages in areas where English is not the language of the majority” (de Klerk, 1999, 

p. 311). Examining whether BSAE is a distinct English language variety or just a 

dialect, among other issues, de Klerk (1999) makes a clear difference between BSAE 

and “standard English,” the latter being the English mastered by “the privileged few,” 

and the former the language of the masses who have not been able to improve their 

English by reason of the “national decline in educational infrastructure.” This, however, 

is to be distinguished from pronunciation patterns used even by the most educated 

Black South Africans, and which are tinted with local, African languages and confer 

BSAE a typical accent, according to Lanham (1996), which is of interest in this study 

regarding interpreting. Meierkord (2005), however, notes that some features that were 

previously claimed to characterize BSAE at large are not specific to BSAE, but are also 

found in “standard English.” 

 



 17 

Nwaila (1996) argued earlier that it is not appropriate to use the term “Black English” 

in South Africa because such a variety of English has not been established in the 

country. This conclusion resulted from the finding that 64% of his respondents 

(teachers, inspectors, subject advisors, college students and native pupils) indicated that 

they had never heard of a variety of English called “Black English” in South Africa. 

The researcher claims that the opinion that there is a BSAE is a “powerful” one that has 

become “conventionally listed,” and which is found in “influential circles.” 

  

If Nwaila (1996) doubts the existence of varieties of English in South Africa he, 

however, admits “English used by most South African blacks is characterised by the 

persistence of peculiar forms and usages that are basically found at the basilectal level. 

However, pronunciation is probably the most prominent and remarkable feature of 

ESL” (Nwaila, 1996, p. 111), in other words accent. According to him, if this so-called 

variety of English is understood by other English users, then there is no need to dwell 

on such “negligible regional or community peculiarities” as accent, pronunciation, etc. 

How well this variety of English is understood by other users of English is the issue, 

especially for the interpreter who may not be familiar with such deviations. 

 

Instead of the concept of “Black English”, Nwaila advocates for the use of the term 

“Educated South African black English,” which is nothing but Standard English, 

excluding accent, and cannot be considered (yet) as a new variety of English, unless it 

is institutionalized as is British and American English. This opinion is, however, 

opposed by most of the studies carried out on English as a second language (ESL) in 

relation to interpreting, which show English spoken by ESL users as a lingua franca 

poses a major problem to interpreters and the quality of interpretation. 

 

Lanham (1996) is of the view that South African English accent is not a specific accent 

per say, as one would talk of an Australian English accent. Instead, there are as many 

accents or “variables of English” as there are groups of native speakers of English who 

settled in the country in the course of history, and ethnic groups where English is used 

as a second or a foreign language. He, therefore, identifies Black South Africans as “the 

largest body of English users in South Africa with their own distinctive accent” 

(Lanham, 1996, p. 2). Lanham also argues that even the pronunciation of BSAE varies 

in pronunciation, ranging from an almost native accent used by a small minority to a 



 18 

pattern that is shadowed by indigenous languages. This latter end of the continuum is 

where challenges arise in interpreting. 

Characteristics of BSAE accent 

As a national lingua franca, English in South Africa is used as a second language, a 

language-shift variety, and as a foreign language (Meierkord, 2005). BSAE is mainly 

characterized by its accent, which is the result of direct transfer from local language 

pronunciation into a pronunciation that is common to almost all African English 

languages, and which is used by most black South Africans (Lanham, 1996, p. 2). As 

such, it may not be surprising that conference interpreters would face challenges in 

interpreting BSAE speakers who have a strong accent. In such circumstances, 

interpreters would need additional cognitive capacities to understand the unusual accent 

and cope with new language structures and expressions, which Tieber (2017) believes 

are difficult to process because they are derived and strongly influenced by the 

speaker’s first language. In addition to pronunciation, which forms part of a deviated 

accent, the interpreter of BSAE may also face challenges in speed of speech and 

references to local realities, such as proper nouns of places, persons, among others. 

Also remarkable is the phenomenon of code switching where a speaker (including high 

profile personalities) mixes English with one of the many vernacular languages (de 

Klerk, 1999), assuming that most if not all South Africans would understand. In each 

case, the non-South African interpreter will have to find and apply the right strategy to 

cope with the situation. 

2.1.11 The origins of BSAE 

BSAE, a colonial heritage that dates to the early 1800s, was introduced by the ruling 

classes in a country that already had its own numerous indigenous languages (de Klerk, 

1999). Unfortunately, people did not freely choose to adopt English, but were forced to 

learn it as the language that conferred military, economic and cultural power, and 

guarantee social advancement to those who used it (de Klerk 1999). 

 

As English became the must-have evil for the majority of the Black community in 

South Africa around the early twentieth century, according to de Klerk (1999), this 

group was not privileged enough, except for the few lucky ones, to receive the standard 

form of English through education. Still, according to de Klerk (1999), things became 

even worse when apartheid was instituted in 1948, and with the adoption of the Bantu 



 19 

Education Act of 1953, which “imposed mother-tongue instruction up to the highest 

possible level for black pupils” [and] “effectively denied black pupils access to native 

English speakers, except in the few remaining mission schools” (de Klerk, 1999, p. 

312). The situation was only reversed after the 1976 Soweto tragic uprising, as access 

to English language increased. However, this change seems to have occurred too late 

and only had little impact on the development of a BSAE language. 

2.1.12 Variables of BSAE 

Several elements characterize accent and help to differentiate various accents. A 

variable is the variation associated with speech characteristics (lexicon, phonetics, 

syntax) “present in different degrees of prominence in different idiolects (the speech 

patterns of individuals)” (Lanham, 1996, pp. 2-3). It is worth mentioning from the onset 

that in South Africa, there is more than one South African English accent, but several 

accents characterize the English spoken in the country, which in part is due to the 

diversity of “origins of groups of native English speakers who came to Africa at 

different times, and in part a consequence of the variety of mother tongues of the 

different ethnic groups who today use English so extensively that they must be included 

in the English-using community” (Lanham, 1996, p. 1). 

  

In this regard, one can rightly agree with Lanham (1996) who categorizes the South 

African varieties of English in terms of variables. He, therefore, distinguishes the Cape 

English of British origin, the Cape English of Dutch origin, the Natal English, the South 

African Indian English, the General South African English, the Afrikaans English, and 

the African English. He cites English spoken by the Coloured community as a version 

of Afrikaans English, which forms its basis, though its pronunciation pattern stands out 

and is readily recognizable in South Africa. 

 

Reis Esteves (2009) identifies social variation as one of the characteristics present in 

South African English. She has divided this variation in three different groups: the 

“Cultivated” (reserved for the higher class and close to Received Pronunciation); the 

“General” (for the middle class); and finally, the “Broad English” (reserved for the 

working class). These three linguistic variation groups are found in South Africa at 

large and among the Black South African Community in particular. So, the variables 

applicable to “standard” British English are relevant for South African English, and can 



 20 

also characterize BSAE, while recognizing, however, that such a variety of English 

cannot be homogenous given the many native languages that are likely to have 

influenced the development of what is referred to as BSAE. In this regard, de Klerk 

(2003a: 465), cited by Meierkord (2005: 4-5), emphatically observes that: 

 

while BSAE includes all South Africans who speak a Bantu language as first 

language and who probably learned English from BSAE-speaking teachers, 

there are in fact 9 different official indigenous Bantu languages in South Africa, 

and although these form 4 clusters or groups (Sotho, Nguni, etc.) whose 

members share some characteristics, the language groups themselves differ 

quite significantly. [...] To lump all these Englishes together as BSAE from the 

start would, I believe, be unwise, since linguists would run the risk of 

overlooking any salient differences which might exist between them 

(Meierkord, 2005, pp. 4-5). 

However, for the purpose of this study, variables that characterize BSAE accent and 

that differentiate this accent from others will be discussed as applying to one linguistic 

entity. One of such variables is the lexicon. 

Lexicon 

According to Reis Esteves (2009), BSAE, like other varieties of English in South 

Africa, has borrowed words from indigenous languages existing in the country as this 

has always been the primary way of formation and enrichment of a language 

vocabulary. “South African English has adopted numerous influences from many of 

the local indigenous African languages, such as the Khoi which contributed the click 

sounds to the Xhosa language, and to English such words as “gnu” or “eina” (Reis 

Esteves, 2009, p. 4). 

 

There are words that have been adopted in South African English and that are used by 

all linguistic communities, including Black South Africans. These include: “Biltong” 

(dried meat); “braai” (barbecue) to grill spiced meat; “mealies” (an ear of maize), which 

is processed in maize flour and can be subsequently turned into “pap” (a porridge made 



 21 

from “mealie”). Some words with political connotations have entered the international 

“standard variety” of English and used worldwide. Such include “Apartheid”, 

“Madiba”, Nelson Mandela’s clan name or nickname, and “Townships,” the suburbs 

where black South Africans were forced to live during the apartheid era (Reis Esteves, 

2009). Garage in South Africa is not only the mechanics’ workshop or where one would 

park a car at home, but also a petrol filling station. Robots are not really the products 

of robotics, but traffic lights. 

  

There is also “code-mixing,” whereby a speaker uses several linguistic resources to 

express themselves, mixing local words with Standard English words, which in an 

indication and acceptance of a new variety of English (Reis Esteves, 2009, p. 5) citing 

Bragg (2003: 308).  

 

The same goes with the use of question tags. If standard English uses “isn’t”, “don’t”, 

“aren’t”, “didn’t”, “weren’t” to express a question tag, in South Africa this has been 

reduced to “isn’t?”, usually sounding as “is it?”. So, one would say: “Joan is coming, 

isn’t?” to mean, “Joan is coming, isn’t she?” (Reis Esteves, 2009, p. 6). Worse still, in 

informal settings and in spoken language, the expression “ne?” is used to obtain 

acquiescence from one’s interlocutor. So, it is not surprising that one would frequently 

hear such constructions like, “Get ready, we are leaving in 20 minutes, “ne?” where 

“OK?” would be commonly used in English. 

Accent 

As discussed above, accent is the specific way a person will sound when using a 

language or a way of pronunciation specific to a particular region, ethnic or social 

group. In fact, accent involves phonemics and prosody, two important, but different 

components that help to understand a message in a spoken text. 

  

• Phonemics 

Phonemics, including pronunciation, which is situated at the segmental level, may 

mean phonemes or language sounds that can serve to identify and differentiate words 

as units of meaning. Therefore, any deviation at phonemic level may lead to 

misinterpretation at word level (Lin, et al., 2013). Psycholinguistic theories assume that 

“one’s knowledge of a word includes “phonological, syntactic, morphological, and 



 22 

semantic’ dimensions” (Lin, et al., 2013, p. 35). They further suggest that retrieval of 

words stored in long-term memory may be difficult in case of any deviations from the 

phonological clues previously stored in the spoken source text, as it is assumed that 

when a word is stored in long-term memory, so is the phonological representation of 

the word, making possible the retrieval of the word from the memory. Furthermore, as 

speakers of English as a second language tend to “substitute English phonemes that do 

not exist in their native language with ones that exist, the resulting deviation may thus 

hinder word recognition (i.e. meaning retrieval) by the listener” (Lin, et al., 2013, p. 

35) and the interpreter, too. 

 

Lanham (1996) observes that English pronunciation among Black South Africans in 

the 1990s was diverse, ranging from nearly native English pronunciation to a pattern 

characterized by almost all African English norms. Lanham, however, indicates that 

although “Many intermediate versions exist, [...] the majority of idiolects present the 

characterising variables [...] which are drawn from a central core of African English 

pronunciation norms, being the variables most likely to mark even highly competent 

African English” (Lanham, 1996, p. 7). This view is shared by Meierkord (2005) who 

categorizes the BSAE English variations under “basilect,” “mesolect” and “acrolect” - 

terms known to have been developed in creole studies. Meierkord (2005: 5) contents 

that “The basilect is taken to describe the form of English spoken by people who have 

little contact with L1-English and who have received no or only little formal education. 

The acrolect, in contrast, is the form of English used by educated speakers, which, 

however, shows slight differences from the L1-English spoken in the region” 

(Meierkord, 2005, p. 5). The mesolectal variety of BSAE comprises the majority of 

intermediate idiolects presenting the main African English pronunciation norms 

(Lanham, 1996). 

 

Reis Esteves (2009) identifies several distinctive features among vowels, which 

distinguish South African English pronunciation. The most predominant of such 

features is what she calls the “kit-bit split,” where kit [kɪt] does not rhyme with bit [bət] 

(Reis Esteves, 2009, p. 6). Reis Esteves also observes the very common deviation in 

pronunciation of the sound /æ/, which is slightly raised in the ‘General’ and ‘Cultivated’ 

South African English pronunciation (for example in trap). In Broad varieties of 



 23 

English, it usually becomes [ɛ] (Reis Esteves, 2009, p. 6).” A typical example for this 

is the pronunciation of “South Africa”, which becomes “South Efrica.” 

Consonants also have their distinctive features, especially for the plosives /p, b, t, d, k, 

g/. Speakers, for example, tend to pronounce the /t/ and /d/ sounds by bringing the 

tongue against the upper teeth (Reis Esteves, 2009, p. 7). So, the word “to” will sound 

more like “tsou”. 

 

• Prosody 

While phonemics determines the recognition of words, prosody helps in conveying the 

intended message. Prosody, which occurs at the “suprasegmental” level, has to do with 

syllables, and more specifically with intonation, stress, and rhythm. Therefore, when 

prosody is deviated, it may “fail to reflect the underlying syntactic structures and 

semantic functions within the source text, resulting in miscommunication on the 

receiving end, i.e., misinterpretation by the interpreter” (Lin, et al., 2013, pp. 30-31). 

Listening comprehension is determined by the structure and importance of the message 

expressed in the source speech. These are determined by prosody, which is composed 

of three elements: intonation; stress; and rhythm (Lin, et al., 2013). 

 

• Intonation 

From the listener’s perspective, intonation serves to determine the structure of an 

utterance. It reflects the grammatical functions of parts of speech, and conveys the 

speaker’s emotions and attitudes, for example indifference, enthusiasm, irony, humour, 

and so forth. It also helps to determine sentence type (whether declarative, 

interrogative, or other) and sentence completion, that is, whether the speaker has come 

to the end of their point or idea. Speaking of the BSAE accent, Lanham (1996) points 

out that “intonation remains the main characterizing feature which is not derived from 

English or Afrikaans. It is, in sound, the sustained high pitch of terminal syllables in 

assertions as well as questions” (Lanham, 1996, p. 7). 

 

In any event, pitch variation should be neither “overly narrow” (which may make the 

speaker appear “perfunctory”), nor exaggerated, lest the speaker sounds “pretentious” 

(Lin, et al., 2013). Pignataro & Velardi (2015: 131) argue that voice intonation is also 

important to maintain the cohesion of the intended message as the result of the 



 24 

relationship between the way the message is expressed (wording), and the way in which 

it is uttered (sounding). “It is the intonation, the relationship between wordings and 

soundings, that sometimes makes it possible to infer the meaning of a message in which 

the meaning of each single word has not been fully understood” Straniero Sergio (2007: 

300) cited by (Pignataro & Velardi, 2015, p. 131). 

 

• Word stress 

Stress appears both at word and sentence level. It serves the same function at the 

sentence level as rhythm (discussed in the next paragraph). At word level, stress carries 

the syntactic function and meaning of the word. In English, word stress is expressed in 

three degrees: it can be strong, medial or weak, “the difference between stress and 

unstressed syllables being greater, and the rules for assigning word stress more complex 

than in most other languages” (Lin, et al., 2013, p. 35). Common stress factors 

encountered by non-native English speakers include overgeneralizing their first 

language stress pattern, and especially reversing the stressing order of some words, 

resulting in a change in the grammatical function. For example, “record” (to reˊcord 

and a ˊrecord); “insult” (to inˊsult and an ˊinsult); “rebel” to reˊbel and a ˊrebel. 

 

• Sentence stress and rhythm 

Stress at sentence level has the same crucial function as sentence rhythm, that of 

avoiding misunderstanding. In fact, stresses together with pauses give rhythm to spoken 

language. English is a stress-timed language, which has an influence on the rhythm 

(Lin, et al., 2013). This means, the length of an utterance is not determined by the 

number of syllables – as it would in a syllable-timed language – but by the number of 

stresses, which implies a variation in syllable length. Individuals who use English as a 

second language are less likely to observe variation in stress and in syllable duration 

between stressed and unstressed elements, than native users of the language. While 

appropriate pausing is necessary to reflect sentence structure, non-native speakers of 

English may pause more frequently and within grammatical boundaries (Lin, et al., 

2013). 

 

(Pöchhacker, 2011, p. 196) believes that “The issue of prosodic and other deviations 

has become particularly acute in international conference settings where English is used 



 25 

by many participants as a lingua franca, more often than not with interference from 

their first language”. If the term ‘foreign accent’ is generally considered by interpreters 

as a major source of stress in their profession, the phenomenon goes beyond deviation 

from the standard way of pronouncing individual words and extends to the syntactic 

and lexical levels of discourse Pöchhacker (2011). It is, therefore, assumed that as a 

second language used by the majority of South Africans, BSAE is no exception. 

Quality in interpreting 

Quality is a major concern and a challenge in the training of interpreters, the 

professional practice of interpreting, and in interpreting research (Pöchhacker, 2015). 

Yet, it is considered such a complex and multifaceted concept that some scholars find 

it difficult, if not impossible, to define. According to Zwischenberger (2010: 128), 

 

The concept of quality involves many different variables and perspectives so 

that it may be very difficult and maybe even impossible to ever find one uniform 

working definition of interpreting quality applicable to all kinds of interpreting 

situations and all the viewpoints involved. It always needs to be specified for 

whom, how and under which circumstances quality is investigated 

(Zwischenberger, 2010, p. 128). 

  

Quality mostly refers to the interpreter’s performance as an output in real time of a 

message uttered in an original text into a target language for the benefit of the intended 

audience. Rather than being elusive as it has been for decades, the concept of quality in 

interpreting by essence involves several dimensions that allow to have a diversity of 

views of the concept of quality (Pöchhacker, 2015). It means that quality in interpreting 

is considered from a variety of perspectives, including interpretation as an autonomous 

text and as a text that is in relation with the original, interpreters’ perception of quality, 

users’ expectations and perceptions, and employers’ expectations and perceptions of 

quality of interpretation (Dal Fovo, 2015). Citing Pöchhacker (1995: 234), Pignataro & 

Velardi (2015: 129) assume: 

 

that the purpose of interpreting is determined by the target culture of its 

recipients and that the target text must “make sense” in the specific 



 26 

communicative context and culture: the “skopos derives from the assignment, 

from the circumstances, from the requirements and ulterior motives […] of the 

client (Pignataro & Velardi, 2015, p. 129). 

 

Moreover, interpretation is said to be of good quality when it successfully achieves its 

communicative functions among the parties involved in the communication process in 

a specific context Diriker (2004) cited by Koumba (2014). At the cognitive level, Gile 

(1991:198), cited by Koumba (2014), is of the view that the interests as well as the 

intentions of the speaker should be fully represented by the interpreter. Interpretation 

must align with the original with regard to both content and form, even to the point of 

producing on the target audience the effects that the original speech has on its audience. 

In this regard, Kurz (2001) states that: 

 

What our listeners receive through their earphones should produce the same 

effect on them as the original speech does on the speaker’s audience. It should 

have the same cognitive content and be presented with equal clarity and 

precision in the same type of language. Its language and oratory quality should 

be at least on the same level as that of the original speech, if not better, given 

that we are professional communicators (Kurz, 2001, p. 395). 

  

2.1.13 Assessing quality in interpreting: content vs form 

Quality is a complex, multifaceted and sometimes contradictory concept as seen above. 

When it comes to interpreting, it may depend on the modes of interpreting, and on the 

aim and the theoretical approach of a given study (Grbic, 2015). Quality in interpreting 

can, therefore, apply to a material product, such as a recording or a transcription of an 

interpretation, to a “mental process, for instance when an interpreter’s output is 

analysed with regard to the use of certain strategies; or to social actions, as in the 

analysis of interpreter-mediated face-to-face interaction or of an entire conference 

event as hypertext” (Grbic, 2015, p. 334). As a service to a customer, quality can be 

investigated externally from the user expectation, or internally as what an organisation 

or institution normally considers its culture for quality (Grbic, 2015). 



 27 

 

Referring specifically to media interpretation, Pignataro & Velardi (2015) content that 

the norm of quality in interpreting has more to do with form than content because of 

the pragmatic nature of the context where it is performed, that is, the media. Here the 

entertainment function has pre-eminence on the information function, and interpreters 

are not for the accuracy of their interpretation, but for their capacity to convince their 

audience (Pignataro & Velardi, 2015).  

 

Many, on the other hand, including scholars and professional interpreters, believe that 

meaning consistency is the most important criterion to be considered when assessing 

the quality of a piece of interpretation (Guo, 2013). In considering the criteria to be 

used in such an assessment, Guo, echoing Pöchhacker (2004), suggests that fidelity, 

accuracy and completeness should be considered while assessing if the source language 

meaning has been conveyed into the target text while adopting a “language-bound and 

a concept-bound analysis” (Guo, 2013, p. 57). 

 

Another school of thought is of the view that quality interpreting must be accurate both 

in terms of content and form and be able to transfer all major types of meaning, namely 

ideational, textual and interpersonal meanings Guo (2013). This view is supported by 

Jean Herbert (1952) and Pöchhacker (2015) who believe that the task of interpreting 

amounts to “rendering the original speech as accurately as possible, retaining as far as 

possible the same style” (Pöchhacker, 2015, p. 34), while Gile (1983) considers quality 

as a “weighted sum of informational content and presentation (Pöchhacker, 2015, p. 

35).” As to whether pre-eminence should be given to one rather than to the other 

dimension, Pöchhacker advises that account should be taken of the context in which 

the interpreting exercise is carried out, participants’ professional background, as well 

as the purpose and the nature of the meeting. In fact, different listeners of the same 

speech in the same situation may have different expectations. That is why, while 

ensuring consistency with the original message is important, one should also bear in 

mind the context and the audience for which interpretation is provided (Kurz, 2001). 

Therefore, citing Jean Herbert (1952), Kurz (2001) holds that: 

 



 28 

It is quite clear that in a diplomatic conference the greatest attention should be 

paid to all the nuances of words, while in a gathering of scholars, technical 

accuracy will have greater importance; in a literary and artistic gathering, 

elegance of speech; and in a political assembly, forcefulness of expression. 

Similarly, the style and tone cannot be the same in a small group of three or four 

sitting around a table, in a committee room with a membership of twenty or 

fifty, and at a large public meeting where many thousands are gathered (Kurz, 

2001, p. 395). 

 

Grbic (2015: 335) recalls that “Pöchhacker (2001) proposes an ‘onion’ model of 

superimposed standards of quality, involving “accurate rendition of source, adequate 

target language expression, equivalent intended effect and successful communicative 

interaction” (Grbic, 2015, p. 335). This view is shared by Collados Ais and Gile 2002: 

312 cited by (Pignataro & Velardi, 2015, p. 130), who support that “Quality can be seen 

as the juxtaposition of a set of characteristics of several components of the interpreter's 

discourse, such as informational fidelity, linguistic correctness, quality of prosody, 

quality of voice” (the researcher’s translation). In essence, good interpretation is when 

the output message is consistent with the original utterance, while the duty of an 

interpreter is to ensure accuracy and faithfulness of the target text (Koumba, 2014). 

2.1.14 Interpreting quality assessment criteria 

To assess interpreter performance, the South African Translator’s Institute (SATI) uses 

a transfer competency and language competency rubric. This assessment is based on 

such criteria as “meaning transfer skills,” “application of interpreting mode,” 

“interaction management skill,” and “rhetorical skill” as far as transfer competency is 

concerned. For language competency, the assessment is based on the criterion of 

“language proficiency enabling meaning transfer.” SATI assessment rubric gives a 

description of the levels of performance for each criterion using a rating scale of five 

bands, Band 1 being the highest level and Band 5 the lowest level of performance. A 

band is assigned for each criterion depending on the interpreter’s performance. 

To earn Band 1 for the “meaning transfer skill” criterion (which is considered in this 

study) in a monologue simultaneous interpreting task, for example, an interpreter is 



 29 

required to interpret the intent and content of the message with accuracy and no 

unjustified distortions, omissions and insertions, and to prove his ability and skill to 

resolve all problems relating to meaning transfer. 

 

For Band 2 the interpreter must interpret the intent and content of the message with few 

minor unjustified insertions, omissions, and/or distortions. He or she must above all 

demonstrates ability and skill to resolve appropriately meaning transfer issues. 

Angelelli (2009) proposes the following criteria to assess translation ability, which can 

be applied in rubric assessment of interpreters: “effective meaning-based interpreting”, 

“situational appropriateness”, grammar and language use, style, register, goals, among 

others. Each criterion is assessed using a scale of 1-4 where the highest performance is 

“Proficient” with a score of 4, while the lowest performance is “Beginning” with a 

score of 1. The other intermediate performances are “Competent” with a score of 3 and 

“Developing” with a score of 3. The minimum total score required for accreditation 

being 12. 

 

Taking the example of “Effective meaning-based interpreting” criteria, which is 

applicable for study, the interpreter to be considered “proficient,” is expected to provide 

a meaning-based interpretation, with limited or no influence of the source language, 

and including most details and all major topics from the source speech. For an 

interpreter to be graded “competent,” their interpreting must be “primarily meaning 

based.” It may contain a few instances of influence from the source text, but this should 

not deter the message significantly. Moreover, the major topics must be reflected in the 

message with a significant number of details.  

 

In a study investigating the perception of quality in interpretation from the service 

provider’s perspective, that is, by other interpreters, Zwischenberger (2010) counts 

logical cohesion, sense consistency with original, and completeness among the 

“content-related criteria” used to assess interpreters. These, according to her, are 

distinct from the criteria that are related to the form, including correct grammar, correct 

terminology, and appropriate style (Zwischenberger, 2010, p. 135). They are also to be 

distinguished from “delivery-related criteria,” which include fluency, intonation, 

pleasant voice, synchronicity, and native accent (Zwischenberger, 2010). 



 30 

As indicated above, only content-related criteria will be considered in analysing the 

data collected for this research (see evaluation rubric in Appendix 6). 

Conclusion 

Scholars agree on the fact that the end function of the interpreting activity is to bridge 

gaps between languages and cultures, while different approaches may be adopted to 

study and understand the process of interpreting. While it can be done simultaneously 

or consecutively, there is a variety of types of interpreting depending on the setting and 

conditions, with several factors involved in the process, which are likely to influence 

the quality of the rendition, for example, the speaker’s accent, in this case BSAE accent. 

This chapter has examined the linguistic landscape in South Africa, and some 

characteristics of BSAE accent. The following chapter discusses the methodology and 

the theoretical framework adopted for this research. 

  



 31 

CHAPTER 3: METHODOLOGY AND THEORETICAL 

FRAMEWORK 

3.1 Introduction 

By its very nature, research in interpreting studies is a multidisciplinary and 

interdisciplinary social activity, which is carried out based on a very dynamic theory 

and a variety of approaches, concerning various aspects of spoken and sign language 

translation involving different modes and a variety of settings (Bartłomiejczyk, 2013). 

This chapter investigates the theories and the methodology applied to determine the 

impact that the accent of a BSAE speaker has on the quality of interpreter’s output. 

3.2 Theoretical framework 

A theoretical framework encompasses all approaches applied in a research process and 

is concerned with what data is collected, where, why and how it is collected, as well as 

how the data collected is analysed. The current study is experimental research, which 

is one of the two components of empirical research theory, identified by Bartłomiejczyk 

(2013), the other component being observational research. Observational research 

analyses existing material, both source and target texts, from a real-life interpreted 

event. It has the advantage of providing authentic data with reliable interpreting 

processes (Bartłomiejczyk, 2013). The main challenge faced by researchers engaging 

in observation, is “access to authentic interpreting data” (Bartłomiejczyk, 2013, p. 3). 

 

Experimental research, on the other hand, is a scientific explanatory approach based on 

isolating one factor or construct referred to as an “independent variable” (accent in this 

study) and measuring its effect on another, which will be the dependent variable (in this 

case the quality of interpreter’s output), while controlling the conditions (Liu, 2015). 

Liu adds that experimental research in interpreting can be considered “pre-

experiments,” or “quasi-experiments’ as opposed to “true experiment” because the 

researcher is not always be in a position to manipulate independent variables or assign 

participants randomly as required in true experiment. 

 

So, considering Liu’s definition, the current study on the influence of BSAE speaker’s 

accent on the quality of interpretation, can be considered a quasi-experiment following 

the empirical research theory. A limited number of participants – professional French 



 32 

A-English B interpreters are requested to interpret a real-life recorded speech by a 

Black South African personality. The experiment is a simulation of remote 

simultaneous interpreting exercise, and the recorded target texts are evaluated and 

analysed. 

 

As pointed out by Bartłomiejczyk (2013), in an experimental interpreting study, data 

are collected through participants who are requested to interpret the same source text 

in very similar conditions, which is an advantage. The main criticism against this 

method is that experiments are not carried out in real-life situations, thus lacking 

“ecological validity” (Bartłomiejczyk, 2013, p. 2). However, with the advent of remote 

simultaneous interpreting (RSI) which is widely used nowadays, this design as applied 

in this study has become very relevant as participants are required to interpret a 

recorded video speech in conditions similar to reality.  

 

Another criticism against experimental research in interpreting is the use of 

interpretation students or bilingual non-interpreters instead of professional interpreters 

as participants, which poses the problem of “population validity” and makes difficult 

the generalization of research findings to the target population, that is, professional 

interpreters (Liu, 2015). Again, this other criticism is addressed in the current study as 

participants are professional interpreters. It is worth admitting, however, other common 

shortcomings proper to this method, which may constitute an obstacle to achieving 

population validity. These are the sample size and the randomness of its selection from 

the target population. Indeed, our sample is limited to 13 participants and convenience 

sampling was used to select them. Except for the language combination, which they 

have in common, that is French A-English B, participants’ profiles differ in age, 

professional experience, and training background, among others (see section 4.4.1 on 

participants’ profile). 

3.3 Methodology 

Given that interpreters interpret for speakers of different backgrounds and professions 

in different settings and using different modes, “this professional dimension of 

interpreting distinguishes the field of interpreting studies from other fields in the 

humanities.” It is a multifaceted activity that “inevitably leads to similarly multifaceted 



 33 

and multidisciplinary modes of enquiry, or research methodologies” (Napier & Hale, 

2015, p. 257). 

 

Approaches to perceive, describe, and explain interpreting processes and practices as 

an intertextual and intercultural transposition are numerous and varied. For several 

years now, more and more empirical studies are dedicated to interpreting and more 

focus is on the nature of the research methods used. According to Napier & Hale (2015), 

the term methodology refers to “the overall approaches to the research process as a 

whole and is concerned with why certain data are collected, what, where and how data 

are collected, and how these data are analysed” (Napier & Hale, 2015, p. 257).  

 

Citing the Routledge Encyclopedia of Interpreting Studies, Ricoy & Napier (2017) 

identified no less than 21 interpreting research methods. A methodology for a research 

project in interpreting can either follow a single approach or be flexible to combine 

different approaches in the form of a “mixed-method approach” (Napier & Hale, 2015). 

In other words, whether corpus-based, cognitive, ethnographic, linguistic, 

psycholinguistic, sociolinguistic or other, all research approaches to interpreting studies 

are either qualitative, quantitative or a combination of both. Such combination is 

referred to as “mixed methods research.” Napier & Hale (2015) observe that 

“investigations of a more theoretical, philosophical nature have not been prominent in 

interpreting studies; rather, most research on interpreting has been data-based, using 

both quantitative and qualitative approaches” (Napier & Hale, 2015, p. 258).  

 

In a nutshell, quantitative studies follow a paradigm where hypotheses are deduced 

from existing theories based on previous studies. On the other hand, qualitative studies 

are more inductive, allowing general questions to bring about more complex solutions 

giving a description and an interpretation of the problems, while data allow to induce 

new findings and theories (Napier & Hale, 2015). This study uses the mixed-methods 

methodology to address the complexity of the processes and practices involved in 

interpreting an accented BSAE speech into French.  

3.3.1 Research method 

The mixed-methods approach to research is described by Hild (2015: 262) as “an 

integrative form of inquiry in which the researcher collects and analyses data and 



 34 

integrates the findings using both qualitative and quantitative approaches or methods 

in a single study or project” (Hild, 2015, p. 262). This research method is, therefore, 

considered by its advocates as an alternative to pure qualitative or quantitative 

approaches, which enables researchers to explore processes and measure the outcomes 

in a way that would lead to a more comprehensive treatment of the issue under study 

and produce more credible and comprehensive results than a mono-method research 

can accomplish (Babaii, 2012). 

 

The one quantitative method adopted for this study is survey research, as part of the 

required information is gathered from a sample of a population – a number of French-

speaking professional interpreters – requested to inform a pre-designed questionnaire.  

For this research, the qualitative approach is used through a case study in which the 

interpreter performance is the focus. Participants are expected to interpret a recorded 

real-life speech by a BSAE speaker, and the quality of their output is subsequently 

assessed by an interpreting lecturer. 

3.3.2 Participants  

The sampling framework for this study is a population of professional interpreters with 

French as their mother tongue or first working language (French A), while English is 

their B-language. The participants are trained, practicing and experienced 

professionals. The sample comprises freelance interpreters from Africa. The study uses 

a total sample of 13 participants picked from the official database of freelance 

interpreters available at the Continental Secretariat of the African Peer Review 

Mechanism (APRM).  

3.3.3 Sampling procedure 

The APRM official database of freelance interpreters was obtained from the Head of 

the Conference Management Unit of the APRM, who is the author of this study. Then, 

convenient sampling was used where participants were contacted personally, and their 

individual consent requested to participate in the study. As the database provides 

information about the permanent address of each interpreter in addition to their country 

of origin, the sample was constituted in such a way that it contains interpreters based 

in South Africa or in the Southern Africa region, irrespective of their countries of 

origin, and others from other regions of the continent. 



 35 

For practical reasons relating to collecting and analysing data, a sample of 13 

participating interpreters was targeted, of which eight are based out of South Africa and 

the other five based and working locally. 

 

APRM is an entity of the African Union based in Midrand, Johannesburg, South Africa. 

Its primary mandate to promote policies, practices and standards that contribute to 

economic growth, political stability, economic regional and continental integration, and 

sustainable development through experience and best practices sharing. APRM deals 

with economic governance and management, political governance and democracy, 

socio-economic governance, corporate governance, and State resilience to shocks and 

disasters. APRM membership is constituted of almost all African Union member states. 

The Mechanism has a very database of translators and interpreters for all AU working 

languages, including more than 30 French A-English B freelance interpreters from all 

over Africa. Using this database made it possible to have interpreters based in South 

Africa and those operating from other parts of the continent. It is assumed that the 

former group enjoys greater exposure to, and experience in interpreting BSAE accent, 

while the latter are less likely to interpret BSAE accent. 

In addition to interpreters, another individual is participating in the study to read, and 

audio record the transcription of the original speech using Standard British English 

accent. Two interpreting lecturers from two university institutions are participating in 

the study as evaluators of the interpreters’ output. One lecturer is requested to evaluate 

interpretation of the original source text, while the second evaluator evaluates 

participants who have interpreted the transcribed source text, read with a neutral or 

British Received Pronunciation (RP) accent. 

 

3.3.4 Ethical considerations 

 

3.3.4.1 Participants’ consent and anonymity 

All necessary measures were taken to obtain participants’ consent, respect their rights 

and preserve their anonymity by keeping all private information confidential. An 

information form was provided to all participants, including the evaluator and the 

reader of the transcription, explaining the nature of the study, thus allowing them to 

give their informed, free consent (see appendix 3). A consent form was also sent to all 



 36 

participants, inviting them to participate voluntarily in the study (see appendix 4). They 

were informed of the possibility to freely terminate their participation at any given point 

of the project. A link to the survey questionnaire (see Appendix 2), and the video or the 

audio recordings of the original or the transcribed speech were sent to participants. The 

evaluator was also provided with the same recording in addition to the audio files of 

participants’ interpretations. An anonymous code name was assigned to each 

participant for the sake of confidentiality. 

 

3.3.4.2 Citations and plagiarism 

A list of references of all consulted sources is provided and any source other than the 

author of this study has been recognised either, as a citation or otherwise indicated.  

 

3.3.5 Data collection 

 

The main data collected for the study is a recording of interpretation by participants of 

a speech in English delivered real time by a Black South African with an accent. The 

speech is a ten-minute excerpt of a briefing to the media by former Kwazulu Natal 

Premier, Mr. Sihle Zikalala, following the 2021 Local Government Elections (see 

appendix 6). At the time of the event where the speech was delivered, the speaker was 

a high-profile public figure; he is a highly educated personality, whose accent cannot, 

therefore, be attributed to lack of education or inadequate level of English. As a high-

level Cabinet member, he regularly gave public speeches with good chances of being 

interpreted.  

 

Two main groups of participants were formed: the experimental group (Group A) was 

requested to interpret the source speech with the original BSAE accent, while Group B 

(or control group) had to interpret the transcribed source speech read with a neutral or 

British RP accent. Each group comprised some interpreters based in South Africa and 

others working from other countries. A YouTube link to the excerpt of the speech, and 

the audio file of the transcribed speech were sent to individual participants, depending 

on their respective groups.  

 

Given the difficulty to gather all participants for an online working session – these are 



 37 

all freelance professional interpreters who are frequently engaged in assignments – each 

participant was requested to interpret the video or the audio-recording at their earliest 

convenience, to record themselves and to return the audio file of their interpretation to 

the researcher. They were, however, requested to do it at the first go, as would be the 

case in a real-life situation. Unfortunately, there is no guarantee that this instruction 

was respected, which constitutes a limitation to this study. 

 

A meaning-based rubric eva