Professional Documents
Culture Documents
John Hodgetts
Pronunciation
Instruction
in English
for Academic
Purposes
An Investigation of Attitudes, Beliefs
and Practices
Second Language Learning and Teaching
Series Editor
Mirosław Pawlak, Faculty of Pedagogy and Fine Arts, Adam Mickiewicz
University, Kalisz, Poland
The series brings together volumes dealing with different aspects of learning and
teaching second and foreign languages. The titles included are both monographs
and edited collections focusing on a variety of topics ranging from the processes
underlying second language acquisition, through various aspects of language
learning in instructed and non-instructed settings, to different facets of the teaching
process, including syllabus choice, materials design, classroom practices and
evaluation. The publications reflect state-of-the-art developments in those areas,
they adopt a wide range of theoretical perspectives and follow diverse research
paradigms. The intended audience are all those who are interested in naturalistic
and classroom second language acquisition, including researchers, methodologists,
curriculum and materials designers, teachers and undergraduate and graduate
students undertaking empirical investigations of how second languages are learnt
and taught.
Pronunciation Instruction
in English for Academic
Purposes
An Investigation of Attitudes, Beliefs
and Practices
123
John Hodgetts
Łódź, Poland
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Acknowledgements
The process of devising and implementing the research project upon which this
book is based, although challenging, was a rewarding and enjoyable experience.
I would like to thank all of those involved in the research: the teachers, the course
leader, the learners, and the course mentor. I would particularly like to express my
gratitude to the three teacher participants who agreed to be observed and took part
in interviews on such a demanding course. I truly appreciate the involvement and
participation of all those concerned at University A.
I would also like to thank Prof. Ewa Waniek-Klimczak for her guidance and
support as a supervisor during the doctoral research that forms the basis of this
book. Both her encouragement and knowledge of the subject area were invaluable
and enabled me to defend my thesis in a timely manner.
I would like to express my gratitude to the two reviewers of the original Ph.D.
thesis: Profs. Andrzej Porzuczek and Romuald Gozdawa-Gołębiowski. Their
advice and comments were extremely helpful in assisting the realization of this
book.
Finally, I would like to thank my family and friends for their steadfast support,
both during the research process itself and also during the writing of this book.
v
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Background to the Current Research . . . . . . . . . . . . . . . . . . . . . 1
1.2 General Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Contents of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Pronunciation Instruction: Background, Techniques and Relevant
Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 The First Focus on Speaking . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Audiolingualism: The Beginnings of an Analytical Approach . . . 12
2.3.1 The Theoretical Underpinnings of Audiolingualism . . . . 13
2.3.2 The Impact of Contrastive Analysis . . . . . . . . . . . . . . . . 14
2.3.3 The Critique of Behaviourism, Audiolingualism,
and Contrastive Analysis, and New Directions Sparked
by Universal Grammar and Mentalism . . . . . . . . . . . . . 15
2.4 Sociocultural Theory and Scaffolding . . . . . . . . . . . . . . . . . . . . . 18
2.5 The Natural Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.6 Communicative Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.6.1 The Broad Nature of Communicative Approaches . . . . . 22
2.6.2 Criticisms of the Communicative Approach
and Divergent Subdivisions . . . . . . . . . . . . . . . . . . . . . 23
2.7 The Postmethod Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.8 Humanistic Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.9 The Impact of Humanism, the Self, and Possible Limitations . . . 28
2.10 The Importance of Affect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.11 Pronunciation Techniques and Relevant Research . . . . . . . . . . . . 31
2.12 English as an International Language and the Status
of the Native Speaker Variety . . . . . . . . . . . . . . . . . . . . . . . . .. 37
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 41
vii
viii Contents
5.7.3
What Are the Teachers’ Goals, Attitudes, and Beliefs
on the Pre-sessional? . . . . . . . . . . . . . . . . . . . . . . . . . . 154
5.7.4 Assessment Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
5.7.5 Other Factors Concerning Pronunciation Assessment . . . 157
5.7.6 Opinions on the Syllabus . . . . . . . . . . . . . . . . . . . . . . . 158
5.7.7 Error Correction Feedback . . . . . . . . . . . . . . . . . . . . . . 161
5.7.8 Comments on Declared Practice and Priorities
of Instruction: Teacher Priorities . . . . . . . . . . . . . . . . . . 162
5.7.9 More General Suggestions for Course Improvement . . . . 163
5.8 Teacher Questionnaires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
5.8.1 Biographical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
5.8.2 Teacher Attitudes and Beliefs . . . . . . . . . . . . . . . . . . . . 164
5.8.3 The Most Important Aspects of Pronunciation
and Difficulties Encountered . . . . . . . . . . . . . . . . . . . . . 168
5.9 Learner Questionnaires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
5.9.1 Biographical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
5.9.2 Confidence, Motivation, Error Correction, Accent
Preference, Assessment of Instruction Received,
and Independent Learning . . . . . . . . . . . . . . . . . . . . . . . 172
5.9.3 Frequency of Instruction . . . . . . . . . . . . . . . . . . . . . . . . 173
5.9.4 Most Helpful Activity for Improving Pronunciation
According to Learners . . . . . . . . . . . . . . . . . . . . . . . . . 174
5.10 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
5.10.1 Research Question 1: What Are the Course Goals
in Terms of Suprasegmental Pronunciation Instruction
and Intelligibility-Based Instruction? . . . . . . . . . . . . . . . 175
5.10.2 Research Question 2: Does Assessment Reflect
the Course Pronunciation Goals? . . . . . . . . . . . . . . . . . . 177
5.10.3 Research Question 3: To What Extent Are These Goals
Reflected by Teacher Instruction, Attitudes,
and Beliefs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.10.4 Research Question 4: What Are the Learners’ Attitudes
Towards Pronunciation Instruction? . . . . . . . . . . . . . . . . 181
5.11 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
5.11.1 The Course Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
5.11.2 The Syllabus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
5.11.3 The Course Leader’s View . . . . . . . . . . . . . . . . . . . . . . 184
5.11.4 Possible Problems with Assessment . . . . . . . . . . . . . . . . 185
5.11.5 The Lack of Positive Washback in Pronunciation
Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
5.11.6 Guidance on How the Materials Can Be Used . . . . . . . . 187
5.11.7 A Lack of Support from Listening Materials . . . . . . . . . 188
5.11.8 Conclusions from Actual Practice . . . . . . . . . . . . . . . . . 190
Contents xi
xiii
List of Tables
xv
xvi List of Tables
Pronunciation instruction has long been an area that I have found fascinating both
as a practitioner and researcher. Reflections on my own experiences as a learner of
foreign languages, both in childhood and as an adult, made me aware of the peculiar,
and some would argue, almost unique nature of pronunciation when compared to
other areas of language study. Although learners may react negatively when they
discover that they have produced inaccurate language in writing, pronunciation has
an intensely personal quality that is absent in writing. Losing face by pronouncing
something incorrectly in an L2 can have a detrimental effect on learner motivation,
more so than errors in other aspects of language learning. Of course, this is culture
bound too, so in some learning contexts other errors (for example, grammatical errors)
may also induce a sense of failure or discouragement in learners. Nevertheless, the
act of speaking and articulating words, phrases, and sentences has a link with the self
that other aspects of language learning do not seem to have (Arnold, 2011; Cohen &
Norst, 1989). However, the consequences of not providing pronunciation instruction
and guidance may well mean that learners could complete their language learning
being able to perform the four skills, but not always being understood when speaking:
a highly undesirable outcome. The challenge is therefore to provide instruction that
is appropriate for the particular group of learners. This can depend on a range of
factors, including the age of the learners and the motivation for learning a particular
language.
My experiences as a practitioner have spanned over twenty years (since 1996) and
have been diverse in terms of the type of pronunciation instruction that was demanded
from me (if any). In certain teaching contexts, there was a lack of guidance in terms of
what appropriate instruction could entail. I recall several teacher observations where
the inclusion of pronunciation instruction seemed important for the teacher trainer,
but the type of instruction seemed immaterial, as long as there were elements of
2011). This neglect is perhaps surprising given the evidence that suprasegmental
instruction in particular can assist in improving learners’ intelligibility, even over
short courses of instruction (Derwing, Munro, & Wiebe, 1997, 1998; Derwing &
Rossiter, 2003; Hahn 2004). Pronunciation research itself, particularly in terms of
suprasegmental instruction, has often been viewed as an area of research that has
not had the attention that it should (Deng et al., 2009). In recent years, it seems
that this lack of research has been addressed to some extent (Baker, 2011; Baker &
Burri, 2016; Foote et al., 2011; Foote & McDonough, 2017; Gordon & Darcy, 2016;
Sonsaat, 2017). However, there still seems to be a lack of research in terms of the
application of suprasegmental instruction in the EAP sector (the Baker & Burri 2016
research is a notable exception). Furthermore, the area of pronunciation assessment
has also been noted as one area that has received little focus in the overall patchwork
of pronunciation research (Piccardo, 2016, p. 11).
University pre-sessional access courses in the UK are common. Most UK univer-
sities administer one or more similar courses in the summer months, and some also
run courses in the autumn. Because such courses are so widespread throughout the
UK, the current research will hopefully be helpful in making tentative suggestions
in terms of how they might be improved in terms of their provision of supraseg-
mental instruction based on intelligibility. It could also assist in shedding further light
on pronunciation assessment and learner attitudes towards instruction. Although, as
mentioned earlier, pronunciation research has become a little more common in recent
years, it should be recognized that the crossover between research knowledge and
actual practice is not always evident, and practices that are recommended in research
might not necessarily be assimilated into instruction (Piccardo, 2016, p. 12; Setter,
2017). Indeed, even if teachers are aware of some research recommendations, it
is unrealistic to expect practitioners to be cognizant of all new research develop-
ments in EAP instruction. Even if teachers are aware of pertinent research, there are
certain pressures, attitudes, beliefs, obstacles, and inertias that can shape instruction.
Observing how goals of instruction are set out and exactly how all of these pressures,
attitudes, beliefs, obstacles, and inertias combine to provide pronunciation instruc-
tion is of importance so that more knowledge can be gained in terms of how best to
incorporate intelligibility-driven suprasegmental instruction into such high pressure
courses where time is of the essence. I believe that this book can play a small part in
contributing to this knowledge.
In terms of the scope of the research, although it is clearly focused on supraseg-
mentals instruction and intelligibility-driven goals on a pre-sessional EAP course,
some findings of the research may also be applicable to other ELT contexts, such as
the ESL sector. Of course, due to the different overall aims of ESL courses, there are
certain areas of enquiry that may not be so applicable. For example, it could be argued
that ESL courses should focus on native-like production as a goal of instruction in
order to assist learners in assimilating in their new host country. ESL learner goals
may necessarily be different because integration into what will be a new permanent
home may obviously be a desired outcome for many ESL learners. In EAP contexts,
although it is possible for many learners to remain in their university city after they
have completed their pre-sessional course and degree course, this is generally the
4 1 Introduction
exception rather than the rule, particularly in light of the relatively strict UK visa
regulations. However, in spite of these differences, other aspects of the research
findings could also be applied to ESL, EFL, and ESP sectors.
The book is divided into six chapters: This introductory chapter provides some back-
ground to the current research, general objectives, and details of the book’s contents;
Chaps. 2 and 3 explore the relevant literature; Chap. 4 presents the methodology,
tools, and instruments employed; Chap. 5 presents the results, along with their anal-
yses, and concludes with a discussion of the findings; and Chap. 6 provides a conclu-
sion to the whole book, together with tentative recommendations for future practice
and research.
Chapter 2 is more general in nature than Chap. 3 and is divided into four
sections. The first section provides the general historical background to pronunciation
instruction, examining the approaches to pronunciation instruction taken in different
1.3 Contents of the Book 5
methodologies. The second section investigates the importance of affect upon instruc-
tion, examining how emotional and psychological factors can impact instruction. The
third section outlines the techniques that have been employed to provide instruction,
and the final section considers the issue of the status of the native speaker variety
and the critiques of native speaker-based instruction. Chapter 3 focuses more closely
on the relevant literature pertinent to the focus of the current study and is divided
into multiple sections. It defines suprasegmentals and outlines their communica-
tive role between L1 English speakers, examines the debate between nativeness and
intelligibility, and investigates the concepts of intelligibility, comprehensibility, and
accentedness as these are crucial for the current study. Specific reference is made
to Jenkins’ (2000) lingua franca core and its critique, before the literature review
proceeds to investigate other key issues connected with intelligibility and the research
showing the importance of suprasegmental instruction in improving the intelligibility
of English L2 users. The review concludes by focusing on the testing of English as
this is an important element of the research, followed by an examination of pronun-
ciation research involving Chinese learners of English and pronunciation research
in EAP contexts. The vast majority of learners at University A are Chinese, and the
context of the study is EAP. It is therefore beneficial to examine these two areas in
addition to the research carried out in other contexts.
Chapter 4, the Methodology Chapter, outlines the research context of the study,
presents the Research Questions, provides a research timetable and describes the
methods employed and instruments used, and describes the pilot study that gave
impetus to the current study. It provides a rationale for employing a mixed methods
approach and the particular methods chosen, together with their implementation
and the piloting of instruments employed where appropriate. A description of the
participants, along with a stakeholder analysis is also included.
Chapter 5 presents the results gleaned from the instruments employed, along with
an analysis and discussion of the findings. After the initial results section, the results
are analysed with reference to each Research Question. Finally, the chapter concludes
with a discussion of the key findings. Reference is made to the relevant literature
and research, and the extent to which the findings of my own research coincide with
previous research findings is also explored.
Finally, Chap. 6 presents a concise summary and makes recommendations
for pronunciation instruction on future pre-sessional EAP courses. Although the
recommendations specifically address pre-sessional EAP courses, certain recom-
mendations can also be applied to other areas of English language teaching
provision.
References
Baker, A. (2011). Pronunciation pedagogy: Second language teacher cognition and practice
(Doctoral dissertation, Georgia State University, Atlanta, US). Retrieved March 26, 2018, from
https://scholarworks.gsu.edu/cgi/viewcontent.cgi?article=1017&context=alesl_diss.
Baker, A., & Burri, M. (2016). Feedback on second language pronunciation: A case study of EAP
teachers’ beliefs and practices. Australian Journal of Teacher Education, 41(6), 1–19. https://doi.
org/10.14221/ajte.2016v41n6
Baker, A., & Murphy, J. (2011). Knowledge base of pronunciation teaching: Staking out the territory.
TESL Canada Journal, 28(2), 29–50. https://doi.org/10.18806
Cohen, Y., & Norst, M. J. (1989). Fear, dependence and loss of self-esteem: Affective barriers in
second language learning among adults. RELC Journal, 20(2), 61–77. https://doi.org/10.1177/
003368828902000206
Deng, J., Holtby, A., Howden-Weaver, L., Nessim, L., Nicholas, B., Nickle, K., & Sun, M. (2009).
English pronunciation research: The neglected orphan of second language acquisition studies?
Edmonton, AB: Prairie Metropolis Centre.
Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidence-based perspectives
for L2 teaching and research. Amsterdam: John Benjamins.
Derwing, T. M., Munro, M. J., & Wiebe, G. (1997). Pronunciation instruction for ‘fossilized
learners’: Can it help? Applied Language Learning, 8(2), 217–235. https://eric.ed.gov/?id=EJ5
67518.
Derwing, T. M., Munro, M. J., & Wiebe, G. (1998). Evidence in favour of a broad framework for
pronunciation instruction. Language Learning, 48(3), 393–410. https://doi.org/10.1111/0023-
8333.00047
Derwing, T. M., & Rossiter, M. J. (2003). The effects of pronunciation instruction on the accu-
racy, fluency and complexity of L2 accented speech. Applied Language Learning, 13(1), 1–
17. https://www.researchgate.net/publication/234570703_The_Effects_of_Pronunciation_Instru
ction_on_the_Accuracy_Fluency_and_Complexity_of_L2_Accented_Speech.
Foote, J. A., Holtby, A. K., & Derwing, T. M. (2011). Survey of the teaching of pronunciation in
adult ESL programs in Canada, 2010. TESL Canada Journal, 29(1), 1–22. https://doi.org/10.
18806/tesl.v29i1.1086
Foote, J. A., & McDonough, K. (2017). Using shadowing with mobile technology to improve L2
pronunciation. Journal of Second Language Pronunciation, 3(1), 34–56. https://doi.org/10.1075/
jslp.3.1.02foo
Gordon, J., & Darcy, I. (2016). The development of comprehensible speech in L2 learners: A class-
room study on the effects of short-term pronunciation instruction. Journal of Second Language
Pronunciation, 2(1), 56–92. https://doi.org/10.1075/jslp.2.1.03gor
Hahn, L. D. (2004). Primary stress and Intelligibility: Research to motivate the teaching of
suprasegmentals. TESOL Quarterly, 38(2), 201. https://doi.org/10.2307/3588378
Hattie, J. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to achievement.
Florence, US: Routledge.
Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1),
81–112. https://doi.org/10.3102/003465430298487
Hodgetts, J. (2019). A mixed methods study of institutional, teacher, and student pronunciation
priorities on a UK pre-sessional EAP course (Doctoral dissertation, University of Łódź, Poland).
https://hdl.handle.net/11089/28795.
Jenkins, J. (2000). The phonology of English as an international language. Oxford: Oxford
University Press.
Levis, J. M. (2018). Intelligibility, oral communication, and the teaching of pronunciation.
Cambridge, UK & New York, NY: Cambridge University Press.
Piccardo, E. (2016). Council of Europe: Common European framework of reference for languages:
Learning, teaching, assessment phonological scale revision process report. Retrieved from
Council of Europe: https://rm.coe.int/phonological-scale-revision-process-report-cefr/168073
fff9.
References 7
Setter, J. (2017, June 1). (Review of the book, Pronunciation fundamentals: Evidence-based
perspectives for L2 teaching and research, by T. M. Derwing and M. J. Munro). Applied
Linguistics, 38(3), 430–433. https://doi.org/10.1093/applin/amw041
Sonsaat, S. (2017). The influence of an online pronunciation teacher’s manual on teachers’ cogni-
tions. (Doctoral dissertation, Iowa State University, US). https://doi.org/10.31274/etd-180810-
5849
Chapter 2
Pronunciation Instruction: Background,
Techniques and Relevant Studies
2.1 Introduction
Before the mid-nineteenth century, the dominant teaching methodology was the
grammar translation approach, which saw written language and a concentration on
grammatical forms as goals of instruction. There was therefore very little focus
on speaking as such (Richards & Rodgers, 2001). During the second half of the
nineteenth century, a number of societal shifts meant that contact with speakers of
other languages became easier and more common. Factors such as increased travel
opportunities, a growth in the middle class, and a growth in universities and state
sector education led to a move away from the sole preoccupation with the classical
languages of Latin and Greek towards the European languages of French and German
(Lorch, 2016, p. 175). Early pioneers, such as Prendergast, Marcel, and Gouin began
to advocate the use of alternative methods, largely based on observations of how
children learn their first language.
Prendergast’s Mastery System, initially intended for adult learners’ self study,
emphasized that the primary stage of learning a foreign language should focus on the
accurate pronunciation of the L2 (Lorch, 2016). The publications involved learners
repeating sentences before any focus on grammatical structure. Importantly, Prender-
gast intentionally sought to choose words that were used frequently and sentences that
would illustrate as many of the rules of language as possible (Howatt, 1984, p. 158).
The treatment of vocabulary items as part of a sentence rather than stand-alone,
separate entities was also a significant break from the classical tradition. Repeating
sentences rather than reciting single words meant that the pronunciation of words
was in the context of a sentence. Learning was viewed as an unconscious process,
achieved through the imitation of sentences, not a product of logic, but rather a
product of intuitive memory (Howatt, 1984), and enabled through utterances, not
grammar (Lorch, 2016). It was therefore an inductive approach to language acquisi-
tion. Although some have viewed the impact of Prendergast’s publications as negli-
gible (Richards & Rodgers, 2001), recent appraisals of their influence view them
as being, not only popular at the time, but also influential for other authors and
linguists, such as Francois Gouin and Henry Sweet (Atherton, 2010; Lorch, 2016).
Indeed, Prendergast’s contribution in terms of observing how children learn language
is of note. In particular, the observations that contextual information is used by chil-
dren to assist comprehension and that they initially use memorized phrases were
important contributions (Howatt, 1984; Richards & Rodgers, 2001).
Like Prendergast, Marcel’s Rational Method was based on observations of how
children appear to learn their first language. Although Marcel used different termi-
nology to describe productive and receptive skills, receptive skills were viewed as
preceding productive skills (in a practical and psychological sense) and therefore
recommended as the starting point for learners (Howatt, 1984). Marcel outlined 20
truths of methodology, one of the most significant being that the meaning of utterances
is understood before the acquisition of the symbols and words used to communicate
(Howatt, 1984, p. 153). Although Marcel advocated both analytic methods (examples
and practice) and synthetic methods (imitation and intuitive learning), the latter were
2.2 The First Focus on Speaking 11
viewed as more appropriate and effective in some learner contexts and especially for
children. The method proposed involved the repetition of expressions and the use of
pictures to assist meaning, along with teacher gestures and actions (Howatt, 1984).
In some ways, this explanation of gestures and actions to assist comprehension was
echoed almost a hundred years later in Vygotsky’s work (1962,1978). One problem-
atic element of Marcel’s method is the primacy that is placed on reading: Learners are
expected to read language texts before hearing the language (Howatt, 1984, p. 154).
Nevertheless, the fact that imitation was seen as a crucial element of learning rather
than translation, with learners listening to utterances before producing them, was
an important addition to the inductive methodologist writers of the late nineteenth
Century.
Gouin’s publications, The Series, also presented an inductive method based upon
repetition (Howatt, 1984). The method entails the description of an action or event
by breaking it down into smaller events. For example, Handschin (1912, p. 174) cites
Gouin’s example of “the child washes his hands and face.” The action is broken down
into the actions “he takes up the soap” and “he dips the soap in water.” Aside from the
idea of focusing attention on the verb, the variety of the words chosen was intended
to facilitate the unconscious acquisition of as many words as possible. Also, unlike
Marcel, the first step involved imitating the teacher’s recitation of the sentences.
Aside from a number of drawbacks to the method, such as the lack of motivation that
may be caused by the uninspiring nature of the sentences (Howatt, 1984), Gouin’s
method proved widely popular at the end of the nineteenth century and the beginning
of the twentieth century, with authors such as Handschin (1912, p. 174) praising the
method for its psychological assumptions of learning (unconscious and inductive)
and pedagogical effectiveness. Gouin’s ideas of the sequencing of spoken language
items and presenting new spoken language items in a context were an important
advance at the time and a significant departure from the grammar translation method
(Handschin, 1912). Apart from viewing language learning as something which should
be approached in a similar way to how children learn language, the common thread
connecting all of the three linguistic pioneers was the move away from written
language to oral production as the prominent focus of language learning (Brown,
2006). The principles upon which Gouin and his contemporaries based their ideas
about language instruction became known as the direct method, with communication
restricted to the target language rather than relying upon translation, and the main
points of instruction introduced orally. Both speaking and listening comprehension
were viewed as important aspects of language learning, as was correct pronunciation.
Although these early linguists focused on the oral imitation of language, the practical
implementation of their ideas remained limited until the establishment of the reform
movement at the end of the nineteenth century (Brown, 2006; Richards & Rodgers,
2001).
12 2 Pronunciation Instruction: Background, Techniques …
Although the early pioneers of the nineteenth century had some impact on the
teaching of listening, speaking, and pronunciation, it was not until the advent of
the reform movement at the end of the century that large scale international cooper-
ation between phoneticians and practitioners began to have a greater impact on how
language was taught and how pronunciation instruction was provided. The forma-
tion of the International Phonetics Association by the phoneticians Henry Sweet,
Wilhelm Vietor, and Paul Passy, in 1886, led to the development of the International
Phonetics Alphabet. This accurate representation of sounds in the form of an alphabet
emanating from the new scientific discipline of phonetics meant that sounds could be
transcribed and recorded. The IPA was to become a fundamental tool in pronunciation
instruction. Although, like Gouin and the early linguistic pioneers, spoken language
was to be given prominence, the crucial difference was that phonetics was viewed
as a vital tool for language learning, with instructors expected to have some knowl-
edge and training in phonetics and learners expected to learn the IPA (Celce-Murcia,
Brinton, & Goodwin, 2010). It was the establishment of this reform movement and
the belief that language learning methods should be supported by solid theoretical
foundations that facilitated one of the most significant developments in pronunci-
ation instruction in the twentieth century: that of audiolingualism. The theoretical
foundations of the reform movement would be further strengthened by developments
in behaviourist psychology in the twentieth century (Howatt, 1984, pp. 168–208).
Audiolingualism became established in the 1940s and 1950s and was, at least
in part, a reaction to the necessity of training huge numbers of people to speak a
foreign language. The immediate necessity for such an approach was caused by
the second world war, with US soldiers posted throughout the world (Brown, 2006,
pp. 105–106). Subsequent advances in technology meant that spoken language could
be recorded for instruction purposes or used for self study, without the necessity of
having a native speaker teacher in the classroom: the native speaker model would be
represented on the cassette recording.
The focus of audiolingualism is on speaking and listening activities, and one
particularly distinctive technique is that of minimal pair drills (Celce-Murcia et al.,
2010). Minimal pairs: two words that only differ due to one contrasting phoneme,
were very much a feature of Bloomfieldian structural linguistics (Bloomfield, 1933)
and are a central tool of instruction in audiolingualism. Students are often guided
through a series of listen and repeat drills in order to improve listening skills and
particularly to assist in pronunciation so that they acquire the habit of pronouncing the
target sounds in the correct way (Larsen-Freeman, 2000, p. 35). Both syntagamatic
drills, where minimal pairs are contrasted within a sentence, and paradigmatic drills,
where the contrasting pairs appear across two different sentences, can be employed
(Celce-Murcia et al., 2010). A key departure from the direct method was the system-
atic analysis of the language system: The direct method is an intuitive-imitative
approach, relying on the pupil imitating the instructor, whereas audiolingualism is
2.3 Audiolingualism: The Beginnings of an Analytical Approach 13
of the language and a crucial element which should be learnt before mastering other
elements. Indeed, linguists who were persuaded by behaviourist ideas argued that
pronunciation instruction should take place at a very early age in order to prevent
the reinforcement of erroneous production (Stockwell & Bowen, 1965).
The idea that language learning was a result of learning speech habits also led to a
great deal of research in terms of contrastive analysis, and contrastive analysis lent
further legitimacy to audiolingualism. Contrastive analysis is primarily associated
with Lado (1957) and has had a significant impact on pronunciation instruction,
and particularly, but not entirely, on the method of audiolingualism. It proposes that
the level of similarity between the L1 of the learner and the target language will
necessarily influence the effectiveness of instruction: Put simply, if the target sounds
are very similar to the L1, then in theory, these sounds should be easier for the learner
to acquire, whereas if the sounds of the target language are much different to the L1,
or if they do not exist in the L1, then they will be much more difficult to learn.
Again, a key feature is that mastery of the sound system of the language should
be of primary importance (Fries, 1945). The early advocates of contrastive analysis
were also broadly aligned with the behaviourist ideas of habit formation and the use
of stimulus, response, and reinforcement to assist this habit formation in order to
facilitate language learning (Saville-Troike, 2006, p. 35).
A key concept of contrastive analysis is that of interference of the L1, and a
major aim of contrastive analysis is to maximize the impact of positive transfer (the
production of sounds that are similar or the same in the L1 and the target language)
and minimize the impact of negative transfer: the sounds that differ (Carroll, 1968).
This could be achieved by minimal pair drilling of the sounds that differ. For example,
an Italian speaker might be expected to have difficulties in pronouncing the /i / and
/I/ sounds in English because those two sounds are not different phonemes in Italian:
they are merely allophonic variations. Likewise, Polish and French speakers are likely
to have difficulty with the /θ/ sound in English because the phoneme does not exist in
their respective L1. In its strong form, proponents of contrastive analysis claim to be
able to predict where grammatical and phonological errors will occur by examining
the differences that exist between the target language and the L1. However, this
strong form has been criticized, particularly because learner errors cannot always
be explained by the distance between the L1 and the target language. Critics argue
that errors sometimes occur where contrastive analysis would seem to suggest they
should not, and vice-versa, and that this could be because these errors are caused
due to psycholinguistic factors rather than purely linguistic ones, i.e., the differences
between the target language and the L1 (Richards, 1971). It is therefore debateable
as to whether contrastive analysis can be used as a predictor of student errors.
Another approach, which is also relevant in terms of pronunciation instruction,
is that of functional typology and the concept of markedness. When comparing
2.3 Audiolingualism: The Beginnings of an Analytical Approach 15
linguistic items, the concepts of unmarked and marked are used to describe
those which are more frequently distributed (unmarked) and those less frequently
distributed (marked). Functional typology categorizes aspects of different languages
in order to discover the more marked/unmarked aspects and to compare different
languages in order to identify those aspects which may be more readily trans-
ferrable, and conversely, those marked aspects that may be more difficult to transfer.
These aspects could include phonology, as well as vocabulary, syntax and discourse.
Eckman’s markedness differential hypothesis (1977) attempts to predict the features
that will be easier to transfer between an L1 and an L2, asserting that those features
that are unmarked in the L1 will be more readily transferred and that marked L2
features will be more challenging. In terms of phonology, the syllabic structure of
the particular language will have a marked and unmarked sequence. Although there
are similarities between contrastive analysis and functional typology, particularly in
terms of the predictions of transferability, Saville-Troike (2006) points out that func-
tional typology “goes beyond the surface level constraints of CA to more abstract
patterns, principles and constraints” (p. 57). In particular, the concept of markedness
in Eckman’s hypothesis means that transfer does not always occur uniformly in both
directions.
structured. This innate knowledge was termed universal grammar. The existence of
this universal grammar was therefore seen as a solution to this logical problem of
language learning (Mitchell & Myles, 2004, p. 33). For Chomsky, although linguistic
input is required for successful language acquisition to take place, acquisition is only
possible due to the innate universal grammar that children possess (Saville-Troike,
2006, p. 46). However, the linguistic principles and parameters that children have
must be activated and parameters set to enable acquisition to take place (Aitchison,
2003).
Although Chomsky’s theory of language acquisition (1965) was originally
focused on how children acquire their native language, when applied to second
language acquisition, his work has been crucial in supporting an entirely different
approach to language learning to that of audiolingualism. Obviously, if language
acquisition is no longer viewed as the result of habit formation, it throws serious
doubt on the focus on error correction and minimal pair drilling that takes place in
audiolingualism.
The concept of interlanguage also served to cast doubt upon the assumptions of
behaviourism and contrastive analysis. Selinker’s (1972) term is used to describe the
stage that learners encounter on the way to learning the L2. The interlanguage stage
is viewed as a creative stage where the learner is influenced by their environment and
both their L1 and the L2. This development of language in this distinct stage between
the L1 and L2 is seen as different to the L1 (Cook, 2008, pp. 13–15). Furthermore, and
particularly important in the context of pronunciation instruction, it is argued that the
learner may stop developing certain aspects of their language on their journey from
interlanguage to L2, and if the learner has not developed these aspects in adulthood,
there could be the possibility that they could become fossilized, i.e., that they will
not progress to the L2 (Saville-Troike, 2006, pp. 41–42).
Due to the gradual shift from behaviourism to mentalism, and the fundamental
idea of the learner as a creative, active participant rather than a passive recipient of
surface linguistic structures, and of language as rule governed behaviour, contrastive
analysis gradually became less credible and was succeeded by error analysis (Saville-
Troike, 2006, p. 38). This new way of viewing language acquisition focused on the
description of acquisition and the general stages of acquisition rather than recording
deficient production (e.g., McNeil, 1966). The departure from contrastive analysis
lies in the assumption that learner errors can be helpful because they illustrate how
the learner is discovering the L2 and how they are testing the limits and the rules of
the language. Error analysis is generally associated with Corder (1967) and views
errors as a positive phenomenon and a natural part of the learning process, showing
that learners are experimenting with the new L2 rules. Corder argues that learners
of an L2 go through a stage of transitional competence before they become fully
competent in the L2.
The major aims of error analysis are to scientifically analyze how learners’
language changes and develops, and describe and categorize the different types of
errors made (for example, phonological errors). This is often achieved by conducting
longitudinal studies with L2 learners (Ellis, 1994). One important distinction made
in error analysis is between learner errors and learner mistakes: the former being a
2.3 Audiolingualism: The Beginnings of an Analytical Approach 17
result of gaps in knowledge of the L2 and the latter being slips or memory lapses. For
the purposes of error analysis, mistakes are not the focus of study (Corder, 1967). The
identification of whether errors are interlingual (caused by interference) or intralin-
gual (i.e., originating within the language) and establishing the seriousness of errors
depending on the effect they have on intelligibility or social norms are the major
goals of error analysis (Ellis, 1994). The assumption of the seriousness of an error in
terms of intelligibility is obviously of great importance for pronunciation instruction
because if the pronunciation error does not interfere with communication, it follows
that it is of little importance.
Although error analysis was useful in providing an insight into second language
acquisition, there are a number of criticisms worth noting. Firstly, the classification
of errors into different categories (e.g., morphological, syntactic, and phonological)
presents much difficulty and may be far from definitive. Secondly, it may, for example,
be difficult to identify the omission of tense inflections in English by Chinese L1
speakers as being caused by L1 interference or whether they are merely the result of
the universal development progress that also occurs in the L1 (Saville-Troike, 2006,
p. 40). Thirdly, the focus on errors is counter-productive in the sense that there is
little or no data provided in terms of which items of language have actually been
acquired (Bigelow & Enser-Kananen, 2015; Saville-Troike, 2006, p. 40). Finally,
error analysis cannot record all of the problematic developmental errors in learners’
language production because some learners may avoid making errors by simply
avoiding certain forms and difficult structures (Shachter, 1974).
Notwithstanding these valid criticisms, error analysis has provided valuable data,
not least because, unlike contrastive analysis, which relied on a surface comparison
of the L1 and L2, it focuses on practical English use. The fact that research is often
of a longitudinal nature also means that important stages of linguistic, phonological,
and discourse development can be observed over time, and tentative conclusions can
therefore be drawn in terms of the features that are difficult for learners at particular
levels (Bigelow & Enser-Kananen, 2015).
In terms of comparative analyses, and specifically in terms of speech perception,
it is also worth noting briefly the work of Flege (1995) and Best (1995). The speech
learning model proposed by Flege (1995) categorizes the new sounds that an L2
learner of a language will encounter when learning a language into new, similar, and
the same. It is argued that the most challenging category of sounds for learners is
the similar category, chiefly because sounds that are similar to their native language
may be recognized and perceived as their own L1 sounds and not different ones. This
differs from contrastive analysis because in Flege’s model the new sounds for the
learner do not pose the same difficulty; it is the crucial similar sounds that may be
problematic.
Best’s perceptual assimilation analysis (Best, 1995; Best & Tyler, 2007) argues
that the difficulties for learners in acquiring the new sounds of an L2 will be dependent
upon the phonological similarities and differences in terms of articulation, not only
between their own L1 and the target language but also of sounds within the target
language itself. This means that contrastive analysis, the idea that a comparison of
L1 an L2 sounds is the best way to identify potentially problematic areas for learners,
18 2 Pronunciation Instruction: Background, Techniques …
is insufficient as it does not account for the differences and similarities within the L2
phonological system.
In terms of acquisition, the logical issue to be addressed if one accepts the concept
of a universal grammar, is whether it continues into adulthood. If we assume that
this innate capacity to acquire language does not exist in adults, then pronuncia-
tion instruction that involves adult students might well have different goals to that
involving children. Chomsky’s assertion that we learn our L1 due to the possession of
some innate ability: a language acquisition device, was enhanced further by biolog-
ical research seeming to show a loss of brain plasticity which occurs during puberty
(Lenneberg, 1967). However, Krashen and Terrell (1983) argue that this atrophy of
the language acquisition device occurs from the age of 5, and Bialystok & Hakuta
(1999) see atrophy as a much more gradual process continuing into adulthood.
Although there is little consensus on exactly when this innate ability disappears
or is deactivated, there does seem to be a broad consensus that children possess a
significant advantage over adults in L2 acquisition, and particularly in ESL contexts
due to this innate ability (Cook, 2008, p. 148; Oyama, 1976). Indeed, it is particularly
in the area of pronunciation that some researchers and linguists believe this advantage
to be more significant (Singleton, 1989).
The advantage that children possess in learning a second language does not mean
that pronunciation instruction should be ignored with adult learners. There is much
research to suggest that instruction can be beneficial (Brazil, 1997; Derwing et al.,
1997, 1998; Derwing & Rossiter, 2003; Foote & McDonough, 2017). Furthermore,
there are some researchers who suggest that adult learners are capable of attaining a
native-like competence in their speech and pronunciation (Bialystok, 1997; Derwing
et al., 1997; Marinova-Todd, Marshall, & Snow, 2000; Neufield, 2001; Ricard,
1986). This specific issue will be addressed further in the context of suprasegmental
instruction in Chap. 3 of the book.
Another key feature of sociocultural theory is the concept of the zone of prox-
imal development. This concept is used to describe the potential area of development
learners may acquire if they receive outside assistance (Vygotsky, 1978). There are
certain mental functions which are beyond the current capacity of the learner that have
to be performed in collaboration with an interlocutor before learners can perform
them independently. This interaction and collaboration is therefore a process that
causes language to be acquired (Saville-Troike, 2006, pp. 111–112). This collabo-
rative assistance can be known as scaffolding, and it is important to note that it is a
collaboration, so the learner participates in a two-way interaction and is not merely a
passive recipient of guidance. Another important aspect of scaffolding is the idea that
both L1 and L2 language can be useful, and this means that classroom interaction
should not merely involve the teacher, but should also involve the learners’ peers.
Furthermore, although symbolic mediation is usually based on language and will
often include face to face communication, it can encompass areas of communication
beyond language itself (for example, non-verbal communication), and the interaction
itself need not necessarily be face to face: Learners can be said to be interacting with
authors when they read a text (Saville-Troike, 2006, p. 113). Recent applications
of scaffolding in EAP contexts have concentrated on how dialogic interaction can
be used to encourage student participation and maximize learner progress, partic-
ularly with regard to the ability to engage in purposeful academic talk. Research
has focused on how classroom conditions can be created to facilitate this dialogic
interaction (Engin, 2017). One example of how this might be achieved is the teacher
employing referential questions rather than polar questions.
Criticisms of sociocultural theory are centred around the fact that some learners
manage to achieve a good level in the L2 on their own, seemingly without the need for
interaction. Moreover, others may be able to interact in an L2 environment without
learning the language (Saville-Troike, 2006). Saville-Troike (2006, pp. 116–119)
counters these criticisms by arguing that, in the first case, interaction of some kind
will have taken place; for example, if a learner studies online then this can be viewed
as a form of interaction. In the second case, although it is possible for an L1 speaker
to live in an L2 country and manage to communicate basic needs and wants, this
communication is context specific and relies on contextual clues; the speaker is
unable to communicate abstract ideas and concepts.
Although the concept of scaffolding has made a valuable contribution to pronun-
ciation instruction, Johnson (2006) points to the broader issue of the importance
of considering other social aspects of learner context, for example, in terms of the
need to consider the cultural context of instruction and teacher cognition research
that can reveal the thought processes of teachers and their beliefs and attitudes when
providing pronunciation instruction (e.g., Baker & Murphy, 2011). There are also a
host of other macrosocial factors which can influence how instruction takes place, and
Saville-Troike (2006, pp. 120–127) outlines a number of these factors. The status
of the L1 and L2 is seen as important because the economic or social benefit of
learning a particular language will have some effect on learner motivation. Also, in
terms of external motivation, in ESL contexts, those learners who wish to become
integrated in the L2 society into which they have moved will be more likely to be
20 2 Pronunciation Instruction: Background, Techniques …
successful. Schumann’s acculturation model (1978) also delineates the factors that
could lead to a greater social distance between learners and their L2 target. The social
dimension is particularly relevant in ESL and EAP programmes which take place
in L2 target countries, and is of importance in light of research which has shown
the value of learners using the target language outside the classroom, particularly in
terms of improving fluency (Derwing, Thompson & Munro, 2006). Finally, there are
a number of other social factors, such as the age, sex, and ethnicity of the learner, that
could be important in informing instruction and how second language acquisition
takes place.
Apart from the fact that Chomsky’s critique of behaviourism was damaging to audi-
olingualism, one of the effects of his ideas was to give impetus to new approaches
to language teaching. One of these approaches was Krashen and Terrell’s natural
approach (1983). As the name suggests, in many ways, this approach was a return
to the idea of acquiring a second language in a natural way. The ideas of drilling and
listen and repeat exercises were viewed as unnatural means of learning the sound
system of the language. The major assertion of the natural approach is that learners
should not be expected to produce language until they are ready. The assumption is
that learners must initially absorb the sound system of the language, and this means
that instruction utilizing this method differs markedly from audiolingualism in the
way that listening exercises are used. Rather than viewing language as a system
based on the individual sounds that students should learn incrementally, the natural
approach advocates a great deal of listening exposure, with the sole purpose of
enabling learners to assimilate the sound system. No specific language production is
demanded in the initial stages of instruction.
The broad coincidence of the natural approach with the work of Purcell and Suter
(1980) gave rise to a growing belief for some that the impact of any pronunciation
instruction was, at best, negligible (Derwing & Munro, 2015, p. 48). The research of
Purcell and Suter (1980) found that most of the factors that determine pronunciation
progress in learners are ones that the instructor has no control over (e.g., the learner’s
native language and their natural ability to imitate the sounds). They generally found
pronunciation instruction had little effect on the language learners who were the
subject of their research. Although influential at the time, the research results have
been contradicted by subsequent research (Derwing et al., 1997; Derwing & Rossiter,
2003; Hahn, 2004). Also, the research has been heavily criticised for a number of
reasons. Firstly, it concentrated on accent rather than intelligibility as a yardstick
for improvement in pronunciation (Derwing & Munro, 2015, p. 48). Secondly, there
were a number of methodological concerns about the research. Pennington (1998,
p. 327) points out that the use of voice recognition was measured against a native
speaker standard and that native speaker raters were used. Also, it may have been
the case that the instruction that the learners received was ineffective because it
2.5 The Natural Approach 21
was poor due to a lack of teacher training (Piccardo, 2016, p. 14). Furthermore,
problematic sampling and other problems affecting the reliability of the participants
were raised by Derwing and Munro (2015, p. 49). Consequently, there is now broad
acceptance that Purcell and Suter (1980) can no longer be considered relevant and
that pronunciation instruction can be beneficial (Derwing & Munro, 2015, p. 53).
Aside from Krashen & Terrell’s natural approach, Chomsky’s concept of a universal
grammar was also of great importance in influencing communicative approaches to
language learning. Chomsky (1965) made a distinction between linguistic compe-
tence, that is, the knowledge of the linguistic system and its rules by native speakers,
and linguistic performance, which is how that language is used to communicate.
Hymes (1966; 1972) developed these two concepts further by introducing the concept
of communicative competence. He suggested that although linguistic competence is
vitally important, there are three other competences that are of importance: sociolin-
guistic competence is necessary to ensure that language is used appropriately and
that social taboos are not broken; discourse competence is necessary to deal with
types of discourse, both verbal and written, coherently and cohesively; and finally,
it is necessary to possess strategic competence, which is the ability to deal with
communication breakdowns, for example, by asking for clarification or repetition,
or by using non-verbal communication. For Hymes and the functional approaches
that were influenced by the Prague School of Linguistics, the focus of research and
analysis should be how language is used as a communication system rather than as a
system of rules (Saville-Troike, 2006, p. 52). Functional approaches are therefore also
concerned with aspects of communication that do not only involve language (Tomlin,
1990). This can clearly be seen in Hymes’ category of strategic competence, where
gestures can be an important strategy in maintaining effective communication. These
functional approaches laid the foundation for a range of communicative approaches
which shaped how pronunciation instruction would evolve towards the end of the
twentieth century and beyond.
One particularly influential linguistic school of thought which provided some
theoretical foundation for communicative approaches was that of systemic linguis-
tics, which is particularly associated with the work of Halliday (1973). Saville-Troike
(2006) describes systemic linguistics as a “model for analyzing language in terms of
the interrelated systems of choices that are available for expressing meaning” (p. 53).
In particular, systemic linguistics places a great emphasis on the circumstances and
contexts in which language is used, and therefore takes into account other factors
(e.g., the social circumstances of interaction) and not just linguistic ones. Halliday
viewed the process of language acquisition as one that involves the acquisition of
“meaning potential” rather than the acquisition of systemic rules (Halliday, 1973).
Basic language functions are learnt during childhood in the L1 and their meaning
potential is gradually developed (Halliday, 1975). A range of functions are delineated
22 2 Pronunciation Instruction: Background, Techniques …
by Halliday, including, for example, instrumental and heuristic functions. The instru-
mental function, being one of the first functions acquired, involves getting something
or getting something done, whereas the heuristic function involves language that is
used in order to learn about things (Halliday, 1975). According to Saville-Troike
(2006, p. 54), one of the main advantages of such an approach is the ability to utilize
the functions that have already been learnt in the speaker’s L1. It is argued that some
of these communicative functions can be performed with relatively limited linguistic
ability (for example, through the use of non-verbal communication). Therefore, the
learning of the L2 can be seen as a way of putting the flesh on the bones of these
functions that are readily transferrable by learning the new linguistic forms within a
different social context.
In terms of practice, the communicative approach, drawing upon the foundations laid
by Chomsky, is perhaps a little more difficult to describe as a single entity. Perhaps it
would be more appropriate to describe communicative approaches because the term
itself can be used to describe quite a range of approaches (Harmer, 2003, p. 70).
Indeed, Richards and Rogers (2001, p. 173) argue that the term is used in a very
general way to describe types of teaching that encourage students to communicate.
Any description of communicative language teaching is also further complicated by
whether distinctive elements such as task-based learning should also be included
within the broad definition (Littlewood, 2011, p. 541). In general terms, the commu-
nicative approach focuses very much on meaningful communicative activities with a
purpose. In terms of pronunciation, the emphasis in speaking activities is on fluency,
and classroom interaction between students themselves in the form of pair work
and group work is encouraged (Littlewood, 1981). Widdowson (1972, 1978) made
important contributions, particularly in terms of the ideas of signification and value,
that could be particularly pertinent when considering how pronunciation might be
taught. Widdowson (1972, p. 17) explains the limitations and artificial nature of
learners repeating a phrase or structure without meaning. The example of the teacher
modelling the phrase I am walking to the door is given, with learners completing the
drill with various substitutions (e.g., he is walking to the door). Widdowson points
out that in such a task learners are only performing the signification aspects of utter-
ances and not their communicative value; they are providing a commentary on an
action which would not usually require any such commentary (Widdowson, 1972,
p. 17). In terms of phonology, this could be an important consideration too because
the purpose of a communicative utterance may shape how it is pronounced.
Because providing a communicative context is of importance in the communica-
tive approach in order to ensure purposeful communication, a focus on discourse,
exemplified by the work of Brazil (1997), Lewis (1993, 2002), and Widdowson
2.6 Communicative Approaches 23
(Ellis, 1996; Wang, 2007; Yu, 2001). Some of the concerns raised are connected
with students from those countries being reluctant to engage in groupwork. However,
although it may be true that some communicative techniques may not be so successful
in some countries, the important factor should be that the activities designed should
encompass real communication; how this communication can be realized should
be flexible according to the context of instruction (Littlewood, 2011). It may well
be that an eclectic approach to instruction is required in certain teaching contexts,
choosing certain teaching techniques from different methods (Ho & Wong, 2003).
Although broadly in favour of a communicative approach, Littlewood (2011, p. 551)
acknowledges the need to adapt the teaching methods employed in certain contexts
in order to achieve the communicative aim. Another challenge mentioned is that the
communicative tasks often chosen by teachers may not be sufficiently demanding
in terms of the level of English expected from students. The practical application
of communicative methods has also been criticised in other ways: Research has
shown that in several supposedly communicative classrooms grammatical accuracy is
favoured over fluency, and form is favoured over function (Nunan, 1987; Thornbury,
1996). This seems inconsistent with the general principles of the communicative
approach.
It should also be noted that there is a general division within the communicative
approach itself in terms of whether a focus on linguistic forms is appropriate. The
strong form suggests that it is the experience of communication that causes learning
to take place, whereas the weak form encourages analytic strategies that can also be
based upon grammar as well as functions (Littlewood, 2011, p. 545). The choice of
strong or weak form will obviously play a role in determining what kind of focus and
prominence pronunciation is afforded. A focus on form, as well as function, could
provide more opportunities for instruction to take place.
As mentioned earlier, the task-based approach, although broadly classified as a
communicative one, does have a distinctive, but by no means uniform, approach to
instruction. Littlewood (2004) points out there are various definitions of the word task
and what constitutes task-based instruction. For example, there are some researchers
who tend to give a broad definition of the word, meaning that it does not necessarily
require a communicative purpose (Williams & Burden, 1997, p. 168). Estaire and
Zanon (1994) delineate communication tasks, where there is a focus on meaning
rather than form, and enabling tasks, where there is a linguistic focus which can
include grammar, vocabulary, functions, discourse, and pronunciation. If task-based
instruction is provided that includes these enabling tasks, then it is clear there will
be the possibility to focus on pronunciation at some stage in a lesson. At the other
end of the spectrum, there are those who argue that tasks can only be those that
are used with a communicative purpose or goal (Willis, 1996). Viewed in this way,
pronunciation will always be within a general communicative context, and a focus
on linguistic forms would, generally speaking, be less likely to be included.
A more recent development that falls within the umbrella of communicative
approaches is the introduction of the DOGME approach to language instruction.
2.6 Communicative Approaches 25
I made the point earlier that it would be more correct to refer to communicative
approaches rather than the communicative approach. The very fact that the word
approach rather than method has become more widely used perhaps signifies a looser,
less prescriptive set of ideas and principles for teachers to utilize. The postmethod
approach moves one step further, challenging the idea of a transmission model: the
idea that there is a best way to teach.
The postmethod approach, associated with the work of Kumaravadivelu (2001,
2003, 2006), asserts that previous methods were adhered to with little attention paid
to their efficacy. Instead, the postmethod approach advocates the empowerment of
teachers by providing teacher training that encourages them to think critically about
their own personal and professional knowledge rather than adhering to a transmission
approach. It is also suggested that teachers, as well as being equipped with class-
room discourse analysis skills to facilitate classroom investigation, are encouraged
to engage with the literature and reflect on the global power inequalities that may
be present in the local teaching context. Teachers are expected to employ research
methods, such as questionnaires, to develop an awareness of their learners so that
knowledge of preferred learner styles, strategies, attitudes to learning, and learner
anxieties can be gleaned. Teachers can then explore how to utilize and extend the skills
that learners already possess (sometimes through collaboration with other teacher
researchers). Systematic self-evaluation of their attempts to provide instruction is
crucial, with changes in their own practice implemented after reflection on the results
of their classroom research. Collaboration with peers is also a prominent feature of
the postmethod approach (Kumaravadivelu, 2001, 2006).
For learners, the idea of reflection is also crucial (Kumaravadivelu, 2001, 2006).
They are encouraged to keep a learning journal and write a biographical history of
their own language learning experiences. Also, apart from their preferred learning
styles and strategies, they should engage with new styles and strategies. Classroom
collaboration with other learners should take place in small groups, with each group
researching projects and then presenting their results in class. The stress is very
much on learner autonomy, and the learner should be encouraged to get feedback,
both from the teacher inside the classroom and from competent speakers outside the
classroom.
The postmethod approach does serve to highlight some of the shortcomings
of the communicative approach and challenges the idea that teaching methods
have enjoyed a steady march towards progress, culminating in the communica-
tive approach (Brown, 2002; Pennycook, 1989). Although in some ways the same
observation has been made with regards to the postmethod approach (Bell, 2003),
that is, that those who support this approach view it as the best method to employ,
there is merit in Kumaravadivelu’s (2006) criticisms of the transmission model of
teacher training. However, significant obstacles to its implementation could lie in
the unwillingness of teachers to engage in classroom research. Those in charge of
teacher training programmes would also need to be convinced of its effectiveness,
2.7 The Postmethod Approach 27
There have been a number of humanistic approaches to language learning that draw
broadly upon the work of the psychologist Carl Rogers. Although their popularity has
been on the ebb for some time and their approaches differ widely, a brief examination
of some of their methods may be pertinent. The silent way, introduced by Gattegno
(1972), is an intuitive approach to pronunciation instruction, and it focuses very much
on the early stages of language learning. As the name suggests, little teacher language
is used, with the teacher eliciting sounds from the students to form words by pointing
to different sounds represented on a Fidel wall chart. Various coloured cuisenaire rods
are used for a number of different purposes, but particularly to demonstrate intonation
patterns and morphological endings, and to create whole sentences (Celce-Murcia
et al., 2010, p. 6). Although the method is rather uncommon, the concept of colour-
coding to enable students to recognize certain phonological features is one that has
been utilized by other researchers (Nikolaidis & Mattheoudakis, 2012).
Suggestopedia, associated with Lozanov, attempts to ensure that optimal condi-
tions for learners are created by organizing the learning environment into one that
is conducive to creating the most open and receptive conscious state. This learning
environment should include comfortable surroundings and soothing, relaxing music.
In particular, proponents of suggestopedia claim that it can vastly improve the
memorization of words. Also, in terms of pronunciation, intonation and rhythm are
presented against the backdrop of soothing music. The aim is to remove or minimize
anxiety from the learning situation (Richards & Rogers, 2001).
Perhaps the most significant aspect of the community language learning approach
is the role of the teacher as counsellor. Associated with Charles Curran, the approach
focuses on the whole person, and the learning of language is viewed as a social
process (Richards & Rogers, 2001, pp. 91–93). The result is a language learning
environment where learners talk about what they want to, and a syllabus that is topic-
based. The teacher translates what the learners want to say into the target language,
28 2 Pronunciation Instruction: Background, Techniques …
and the learners record their own output (Larsen-Freeman, 2000). There is a good
deal of group work, self reflection, and free conversation (Richards & Rogers, 2001).
Although the approach is no longer common, in part due to its lack of structure and
the need for the teacher to know the students’ L1, the ideas of self-reflection and
students recording their own language production have been incorporated into many
teaching contexts (Baker & Burri, 2016; Scarcella & Oxford 1994, p. 227).
Although these humanistic approaches are less popular now than in the 1970s, it can
be seen that certain elements of the approaches have been adopted and have informed
other approaches (particularly communicative approaches). Stevick’s (1980) descrip-
tion of humanism includes the ideas that what students think and feel is important and
that the teacher should be interested in students and how they view their learning.
Furthermore, learners should have a good deal of control and input in lessons. In
terms of pronunciation, the idea that the teacher should pay regard to what students
feel and be interested in how they view their learning is obviously important, partic-
ularly when we consider error correction. The need for sensitivity is important in
pronunciation instruction because speaking, and pronunciation in particular, is very
personal, and losing face could lead to a loss of motivation. This is particularly true
where students have to give a formal presentation (for example, in EAP contexts).
In this circumstance, the possibility of loss of face is even greater. It is therefore
desirable to know how students feel and to try to make them feel safe and relaxed.
The extent to which the teacher can cater for the individual self of the student
is a more complex issue. Gadd (1996) draws a distinction between romantic and
pragmatic humanism, and argues that romantic humanism: the idea that the major
task of teachers is to “encourage and advance the development of the students’ inner
selves” (p. 225) is flawed because, according to Gadd, the purpose of the language
teacher is to facilitate the learning of a language and not to make students emotionally
fulfilled. Fundamentally, Gadd objects to the role of the teacher as a confessor or
therapist. There is also the risk that if this type of humanistic approach is applied,
instruction will lack structure (Stevick, 1980) and that the student will be ill-equipped
to deal with a variety of registers (Gadd, 1996, p. 233). Nonetheless, the idea of a
pragmatic humanism, which takes account of students’ feelings, attitude to learning,
and learner needs can be important in enhancing motivation and promoting student-
centred learning (Gadd, 1996; Stevick, 1980). This can be particularly important in
the area of pronunciation instruction where the potential for loss of face (which could
lead to de-motivation) is greater (Cohen & Norst, 1989).
2.10 The Importance of Affect 29
One of the consequences of humanism was the focus on the psychological aspects
of language learning. Affective factors can include the inhibitions, self-esteem, and
motivation of the learner, and also the relations between learners in the classroom
(Stevick, 1980). Arnold and Brown (1999, p. 1) define affective factors as “aspects of
emotion, feeling, mood, or attitude which condition behaviour.” Research seems to
show that, alongside cognitive factors, affective factors can influence the effective-
ness of language instruction (Scovel, 1978). In second language acquisition, affective
factors, depending on their nature, could have a positive or negative effect on learner
participation and achievement (Arnold, 2011). For example, a learner’s embarrass-
ment could cause them to participate less in class (Cook, 2008, p. 80). If learners
feel less anxiety, this embarrassment is minimized. There is evidence to suggest that
language teachers should not only consider learner cognition but also learner affect.
Furthermore, it would be erroneous to view cognition and affect as separate entities
that instructors might choose between (Arnold, 2011; LeDoux, 1996). The interde-
pendence of the two seems clear because emotional factors have an impact on how
we think (Bless & Fiedler, 2006). Indeed, MacIntyre and Gardner (1994) found that
negative affective factors led to poorer input, processing, and output from learners.
Although it may seem appropriate to consider affective factors in learning in
general, it has been argued that language learning is an area where affect is of partic-
ular importance. There appears to be an inextricable link between language and iden-
tity, and learners who are not able to express themselves in the target language may
experience feelings of vulnerability (Arnold, 2011). This is quite different from other
academic subjects. Indeed, Cohen and Norst’s (1989) diary study of language learners
confirms that anxiety seems more marked in language learners precisely because of
the link between language and identity. Because of the seemingly increased impor-
tance of affect in language learning, some researchers have sought to develop different
constructs to categorize the causes of affect (e.g., Gardner, 1985). Some have argued
that the relationship between affective factors and learner performance is actually
more complicated, and that learner anxiety might actually be caused by learners
possessing poor language skills (Sparks & Ganschow, 1995). However, the bulk of
the research suggests that anxiety does have a significant role to play in terms of
affecting learner performance and motivation (Arnold & Brown, 1999; Horwitz &
Young, 1991; Young, 1990, 1999). Young’s (1991) review of the literature identifies
personal factors, learner beliefs, instructor beliefs, instructor and learner interaction,
the events that occur in the classroom, and testing as key determinants of affect. Chief
causes of anxiety that have been identified in language learners include comparison
with other learners, the perception of the relationship between teacher and learner,
and testing. Testing was also found to be a key cause of anxiety in Bailey’s (1983)
qualitative study of learner diaries (see also Horwitz, 1986).
For foreign language instruction, one of the most serious consequences that can
arise due to the negative affective factors of stress and anxiety is an unwillingness
for learners to contribute. It has been observed that learners who experience stress
30 2 Pronunciation Instruction: Background, Techniques …
and anxiety are less willing to engage in genuine communication in the classroom
(Steinberg & Horwitz, 1986). This could be a significant obstacle for pronuncia-
tion instruction; if learners participate less, there will be fewer opportunities for
them to receive constructive feedback. There is also a danger that a cycle of non-
communication could begin, where learners feel less and less confident about their
pronunciation because they are not producing the language. For my own study, the
findings of Liu and Jackson (2008) are also significant because their research focused
on the effect of the anxiety experienced by Chinese learners. They found that there
was a significant correlation between the learners’ anxiety and their willingness to
contribute. It therefore appears that negative affective factors could lead to a lack
of progress in learners’ pronunciation of the target language. Teacher strategies and
actions to reduce these negative affective factors could therefore be beneficial in
optimizing pronunciation instruction.
There are a number of important recommendations and observations from the liter-
ature and research on how positive affective factors can be optimized and how nega-
tive affective factors can be minimized. Firstly, metacognition, that is, the learners’
knowledge of how their own feelings and emotions might affect their learning, is
certainly an important consideration (Arnold, 2011). Learners are often required to
give presentations on pre-sessional courses (and also on their destination courses).
If they have a greater awareness of how stress and anxiety might affect their deliv-
ering presentations, they might be able to develop their own ways of coping with
stress and reducing anxiety. Secondly, with reference to Gardner’s (1983) work on
multiple intelligences, it is important for instructors to incorporate activities that
involve several intelligences so that the individual needs of learners are catered for
in order to facilitate positive affect in the classroom (see also Reid, 1995). Thirdly,
learner choice has been shown to reduce anxiety and increase motivation, and this
can be facilitated in a number of ways, for example, by learners participating in
class projects (Arnold, 2011). Encouraging student-centred activities in pronuncia-
tion instruction could be challenging, but not impossible (see Dlaska & Krekeler,
2008, for an evaluation of self/peer assessment of pronunciation). Fourthly, it has
been shown that activities that do not rely heavily on coursebooks seem to increase
motivation and decrease negative affect (Wajnryb, as cited in Arnold, 2011, p. 15).
This may be particularly true of pronunciation instruction in the light of research that
is critical of coursebook content (Derwing, Diepenbrook, & Foote, 2012; Levis &
Cortez, 2008). Finally, and perhaps most importantly, simply encouraging learners is
insufficient. Activities should be chosen that show learners that they can be successful
in real communication (Arnold, 2011).
Certain teacher behaviours can facilitate positive affect and enhance learner moti-
vation. In particular, student interview research shows that learners respond posi-
tively when instructors transmit confidence, when they provide constructive feed-
back, when they have good eye contact, show a personal interest in learners and are
interested in learner responses, and regularly check understanding (Leon, as cited
in Arnold, 2011, p. 18). There is also some evidence that foreign language learners
prefer to participate in speaking activities in small groups rather than in whole class
mode (see Young’s, 1990 survey of 200 Spanish university and high school learners).
2.10 The Importance of Affect 31
It is important to note that affective factors can also influence teachers. Arnold
(2011) warns of the danger of teacher burnout if they are not supported. There is also
evidence to suggest that non-native teacher anxiety can lead to them choosing class-
room activities that do not provide sufficient opportunities for real communicative
practice (Horwitz, 1996). This is significant because, as was noted earlier, for learners,
achievement in using language in real communication is crucial in promoting positive
affect and enhancing motivation (Arnold, 2011). If few opportunities are afforded,
learners are likely to be less motivated and less willing to contribute, which may in
turn also make teachers feel less confident and less willing to engage in commu-
nicative activities. To prevent this cycle occurring, it is important that teachers are
supported through teacher development and guidance.
This section outlines some of the pronunciation techniques and tools that are avail-
able to teachers. It is important to take an inventory of some of the more common
techniques suggested by the literature and evaluate how they might be effective,
particularly in light of research showing the potential benefits of instruction (Baker
& Murphy, 2011; Derwing & Munro, 2015; Derwing & Rossiter, 2003; Derwing
et al., 1997, 1998; Gordon & Darcy, 2016; Hahn, 2004; Morley, 1991). The lack
of training in pronunciation skills that teachers receive (Baker & Murphy, 2011;
Derwing & Munro, 2015; Macdonald, 2002) and the paucity of effective pronuncia-
tion materials in coursebooks (Derwing et al., 2012; Levis & Cortez, 2008; Piccardo,
2016) mean that teachers might not be aware of the available techniques and tools
and their effectiveness. Earlier sections of this chapter briefly touched on some of
the techniques that were associated with particular methods. What follows is a more
thorough examination of the techniques and tools, and how they can be used by
language instructors to provide pronunciation instruction.
Error correction techniques have been a prominent focus of research in pronun-
ciation instruction. It is important to consider what type of error should be corrected
and also how correction could take place. In terms of the type of error corrected, the
rationale behind the correction of low functional load segmentals would seem flawed
as these segmentals do not significantly interfere with intelligibility (Brown, 1991;
Derwing & Munro, 2015; Jenkins, 2000). In terms of the type of correction that takes
place, much research has focused on the effectiveness of recasts (Carpenter, Jeon,
MacGregor, & Mackey, 2006; Lyster, 2007; Saito, 2015). Recasts can be described
as teacher reformulations of learner errors and repetition of the utterance with the
error corrected in order to make learners aware of the error without significantly
hindering communication (Lyster & Ranta, 1997). There is evidence to suggest that
recasts with the aim of correcting pronunciation features might be more effective
than recasts with the aim of correcting syntactic features (Sheen, 2006). Carpenter
et al. (2006, p. 228) argue that this is because learners see that pronunciation is
32 2 Pronunciation Instruction: Background, Techniques …
to use those features that facilitated understanding and eradicate those that obscured
understanding.
Kjellin’s (1999) approach is one which relies on a technique which has remained
familiar since the Reform Movement: that of repetition. Alongside listening exer-
cises centred around suprasegmental features, learners identify the particular features
they wish to emulate. These features are modelled by the instructor and then repeated
chorally by learners. Instant feedback is then provided by the instructor in an encour-
aging manner so as to enhance motivation. The choral repetition is designed to train
speech organs in the new soundscape of the target language. The final stage of the
approach involves learners applying the pronunciation skills acquired to new original
utterances.
Neufield’s (1987) approach is one which is, at least to some extent, influenced
by Krashen and Terrell’s (1983) natural approach in that learners are not expected
to produce the target pronunciation before they have been exposed to it through a
sustained period of listening. This suggested listening training entails 15 h of listening
to phrases no longer than eight syllables long and, reminiscent of Gouin’s Series,
contains listening extracts with repeated words. It is also accompanied by listening
discrimination activities and visual presentations of suprasegmental features, such
as intonation contours. The 15 h training is then followed by 3 h of learner produc-
tion of these target features. The method is based on Neufield’s (1987) research
which seemed to show that after relatively short courses of instruction, the adult
research subjects improved their pronunciation, and some participants were even
rated as having attained native-like production. However, although positive results
were obtained, there were doubts raised regarding the methodology used in the study
(Long, 1990, p. 267).
The concept of imitation has been employed in various pronunciation techniques.
As outlined earlier, minimal pairs were employed extensively as part of the audiolin-
gual method, and although the method is no longer as popular as it once was, others
have sought to adapt minimal pair techniques in order to make them more effective.
Celce-Murcia et al. (2010, p. 346) describe an activity where two different phonemes
are written on the whiteboard, learners are divided into teams, the teacher says a word
containing the target phoneme, and one member from each team races to the board
to touch the corresponding phoneme on the board. I have used a similar activity to
the one described with Italian teenage learners. I wrote a number of different words
on the board containing the target phonemes (in this case /i / and /I/). I also extended
the activity by reciting a story containing those words. Learners then had to run to
the board when they heard the particular word in the same way outlined by Celce-
Murcia et al. (2010, p. 346). This was an attempt to contextualize the language so
that learners heard these sounds in the word as they would in connected speech, and
therefore minimize the major weakness of minimal pair techniques: their failure to
take into account suprasegmental features and the whole of the sound system (Cook,
2008, pp. 81–82). However, minimal pair techniques generally cannot accommodate
the whole of the sound system and the different meanings indicated by, for example,
pauses, intonation, sentence stress, and rhythm. There is also the problem that if
the pairs are derived from coursebooks, they may include words which are rarely
34 2 Pronunciation Instruction: Background, Techniques …
used, and therefore could be of little importance for communication (Cook, 2008,
p. 82; Levis & Cortez, 2008). Also, these activities may be successful with younger
learners, but are far less suitable for adult learners.
Although listen and repeat techniques can be of use, those involving little more
than a copy the rhythm instruction may be of little value, as is repetition without
feedback (Cook, 2008, pp. 81–82). More recently, the idea of providing pronunci-
ation through repetition has undergone a revival due to the growth of task-based
learning. Some have pointed to the problematic nature of incorporating pronun-
ciation by using the communicative approach, and in particular, that the stress on
creativity can hinder fluency as it does not make it easy for learners to recall language
and pronunciation features because they are constantly trying to produce new utter-
ances (Bygate, 2009; Kramsch, 2009). Task-based approaches attempt to resolve this
conflict between meaningful, creative activities that might not necessarily facilitate
fluency, and activities that involve repetition that can help learners recall pronuncia-
tion features, but may prove monotonous, by modifying or changing some element of
the task to keep the activity engaging (Thornbury, 2010). Thornbury (2010) gives four
examples of how this can be achieved. Firstly, the instructor can vary the support that
is provided to learners. One example of this is Thornbury’s disappearing dialogues,
where a dialogue that learners have been practising has words removed. Learners
repeat the dialogues and fill in the gaps when the task is repeated (the number of gaps
can be increased as the task is repeated). How the interaction takes place can also
be modified: Thornbury gives the example of paper conversations, where instead
of speaking, learners engage in communication by writing messages as they would
online. A reduction in the time limit for the same conversation task can also be
a way of practising the same target forms and improving fluency (see also Wong,
1987). Finally, the same speaking task can take place with a variety of interlocutors:
Learners can perform the same task (e.g., a role play) with other learners.
Listening exposure can play a vital part in terms of the consciousness-raising
of the role that pronunciation can play. It can expose learners to a wide range of
speakers so that they can become more aware of the linguistic production of, for
example, speakers both young and old, and from different educational backgrounds
and particular regions. It can also show how different pronunciation features are
used according to the formality of context, and also illustrate some of the differences
in terms of intonation that occur between different speakers (Cook, 2008, p. 82).
Furthermore, authentic English extracts which give examples of both native and non-
native varieties (Kachru, 1982) can be included rather than scripted ones (Ockey &
Wagner, 2018; Shockey, 2003, 2011).
Because affective factors might have a significant role to play in the classroom,
and particularly in pronunciation oriented activities, it could be beneficial to engage
learners in relaxation techniques. Bolstad (1997) suggests the use of breathing tech-
niques, imagery, and soft music to place learners in a relaxed state so that they will be
less fearful of making errors and less stressed. Although this is not a pronunciation
technique per se, it could be beneficial in making pronunciation techniques more
effective.
2.11 Pronunciation Techniques and Relevant Research 35
Instructors can provide effective models, not only in terms of pronunciation, but
also in showing how interlocutors can be effective listeners and assist others in main-
taining intelligibility in a conversation. Wong (1987, pp. 15–20) outlines a technique
where the instructor models good interactive listening skills during a conversation
with one learner whilst the learner’s peers take notes on the strategies the instructor
uses. The class then discusses these strategies in groups before engaging in group
work where learners are ascribed different roles: the speaker, the listener, and the
observer. The observer’s task is to note the effect the techniques used by the listener
has on the speaker. The task is repeated as learners need to fulfil all three roles. Tasks
such as these can be beneficial as they show how intelligibility can be aided by the
listener.
Multisensory reinforcement utilizes a variety of the senses to reinforce a partic-
ular pronunciation feature. This can involve auditory reinforcement, using associa-
tion with onomatopoeic sounds (Firth, 1992), and visual reinforcement by utilizing
pictures showing the shape of the mouth when producing particular phonemes (Celce-
Murcia et al., 2010, p. 337). The disambiguation of meaning by identifying language
chunks can also be facilitated by picture discrimination tasks where learners arrange
sets of pictures in accordance with what they hear in a particular story (Beer, as cited
in Gilner, 2008, p. 101). The colour coding of vowel sounds has also been shown to
be beneficial in assisting learners to discern different sounds by categorizing lexical
items containing the same sound and identifying larger lexical chunks and intonation
patterns (Nicolaidis & Mattheoudakis, 2012). Systems involving the notation of the
number of syllables and the placement of prominent stress can also be beneficial, not
only as an instruction aid in class, but also as a tool for learners to recall the word
stress of new vocabulary items (Baker & Burri, 2016). Another tool involving visual
reinforcement is the use of rubber bands by the instructor to demonstrate stressed
syllables (Chan, 2018; Gilbert, 1994). Tactile reinforcement techniques can also be
employed, for example, by asking learners to identify voiced sounds by placing their
fingers on their throats (Celce-Murcia et al., 2010, p. 338). There are also kinaes-
thetic reinforcement techniques which involve movement, such as learners taking a
step forward when they hear a stressed syllable (Grant, 2001) or learners counting
syllables on their fingers (Chan, 2018). There are therefore a range of multisensory
techniques which can aid scaffolding, enhance learners’ receptive and productive
skills, and possibly improve recall.
Techniques involving drama may be effective with some learners as they can
reduce the negative affective factors that can be a barrier for some (Celce-Murcia
et al., 2010, pp. 339–340). They can be used to practise using intonation to denote
attitudinal responses by getting learners to repeat the same utterance or dialogues with
different emotions conveyed (Archibald, 1987). Audio-visual tools can also be used
to provide models of interactions and enable learners to record their own production.
In EAP contexts learners can view presentations given by both native and non-native
speakers. Furthermore, learners can video record their own presentations for self-
study. The rehearsal and performance of a scene from a video/You Tube clip can also
be beneficial; Goodwin (2001, 2005) describes how this might be achieved by firstly
36 2 Pronunciation Instruction: Background, Techniques …
using a video clip with the volume muted so that learners become aware of non-
verbal cues. They then watch the same clip with the volume restored and chunk the
script into thought groups. This is then followed by a rehearsal and performance of the
same scene before they go on to perform and video record their own similar scene. Of
course, the division of a text into chunks can also be done with audio materials (Lewis,
1993, 2002), but the added motivation of taking part in a real performance could make
the activity more meaningful and engaging. Although reading aloud activities have
been criticised for not being engaging and not being effective in aiding pronunciation
instruction, these criticisms have generally been confined to read aloud activities
that do not have a communicative purpose, typically consisting of learners reading
instructions or reading extracts from a coursebook (Wilson, 2010). Production is
unrehearsed in these circumstances, and it is likely that such activities will lead
to some learners feeling more insecure about their pronunciation. However, read
aloud activities that involve rehearsal before performance, such as the performance
of a scene from a film, could be beneficial (Gilner, 2008, p. 100; Harmer, 2009).
These activities, as long as they are implemented in a sensitive manner, can assist in
helping learners identify and produce sentence stress patterns and improve rhythm
and chunking (Gibson, 2008).
Songs, games, and activities can be used in a variety of ways to improve learner
receptive skills. The popularity of songs in English globally means that learners,
particularly teenage learners, depending on the culture, might be well motivated and
enthusiastic to take part in song-based activities. A cloze of the song lyrics can be
used, or alternatively, the lyrics can be provided to learners with deliberate errors for
them to correct. Learners can also use the lyrics in a number of ways, for example,
by reciting them as if they were part of a conversation or chunking the song text
(Celce-Murcia et al., 2010, p. 354). There are also several games and activities that
might be particularly useful for younger learners. Hancock (1995) describes a variety
of these for different levels, including sound mazes (where learners have to match
words with the same target sounds), pronunciation bingo, and other board games.
Jokes, poetry, jazz chants, and limericks can also be used to improve knowledge of
stress and rhythm in an engaging manner (Celce-Murcia et al., 2010, pp. 349–353).
Self-reflection can be encouraged if learners keep an oral journal of their expe-
riences (Baker & Burri, 2016). This can be useful for instructors because they will
be able to provide feedback on the key pronunciation features that they hear on the
recording. Instructors will become more aware of problems that individual learners
might have, not only in terms of linguistic factors but also in terms of social and
psychological factors. Also, learners could feel less stressed about sharing their
concerns or difficulties with the instructor in an indirect way.
Discourse-based techniques can be more suitable for adult learners. Aside from
aspects such as nuclear stress, chunking, and pitch movements, listening exercises can
be used to show the disambiguation of meaning, the organization of discourse, and
particular genre-related features (Tench, as cited in Gilner, 2008). The genre specific
functions of discourse can be particularly useful in EAP, especially in subject specific
courses of instruction and ESP. Intonation training that is based on authentic listening
discourse has been shown to be beneficial (Aufderhaar, 2004), and this training can
2.11 Pronunciation Techniques and Relevant Research 37
be aided by providing visual scaffolding using pitch graphs and intonation contours
(Chun, 1988, 2002).
The growth in technology has also afforded teachers access to a number of tools.
Of course, how these tools are exploited by the teacher is dependent upon the teacher
and teaching context: whether they have access to the technology and whether they
feel confident using it. As mentioned earlier, learners can record their own language
production when completing tasks, but they can also make audio recordings of their
own progress as an oral diary. Feedback on the diary can then be provided by the
teacher through the use of free software, such as vocaroo (https://vocaroo.com/; see
Baker & Burri, 2016, p. 15). There are also a number of speech recognition programs,
such as Google’s speech to text app, which are freely available. These can be valuable
tools for learners to use for self study because they are able to verify the intelligibility
of their production in their own time in a relaxed atmosphere. Another technique
which learners can utilize independently is that of shadowing, which involves learners
listening to an audio model (for example, part of a presentation) and replicating
the model immediately afterwards. Although shadowing has been criticized for the
seeming lack of learner creativity involved, encouraging results have been achieved
in terms of facilitating lively speech with greater pitch variations that listeners find
more comprehensible. Furthermore, these positive results also seem apparent when
the same learners give their own presentations (Foote & McDonough, 2017).
The research that I engaged in involves an investigation into the target of instruction
in terms of whether a native speaker variety is pursued or whether the overriding goal
is one of intelligibility. It is therefore pertinent to examine briefly the research and
contributions made by linguists concerning the status of the native speaker variety,
non-native varieties, and English as a lingua franca. This chapter contains general
background to the issues surrounding English as an international language and the
status of the native speaker variety. A closer examination of Jenkins’ (2000) lingua
franca work and its impact on segmental and suprasegmental instruction can be found
in Chap. 3.
The concept of English as an international language is not a new one. The term
itself can be traced back to the 1980s, as can the concept of World Englishes (Kachru,
1980; Smith, 1983, 1987). Several authors have stressed that English should be
viewed as a language that belongs to a global community of speakers who use it for a
diverse range of purposes, and as such, should encompass a whole range of Englishes
reflecting these purposes (Jenkins, 2000; Kachru, 1982; Matsuda, 2019; Seidlhofer,
2011; Smith, 1976, 1987; Widdowson, 2003). The goal of instruction should therefore
be to enable learners to be intelligible to other interlocutors throughout the world
and not merely intelligible to native speakers (Smith, 1992). Kachru (1982) classified
38 2 Pronunciation Instruction: Background, Techniques …
countries where English is spoken into three categories, describing them in terms
of a concentric circle, encompassing the inner circle, the developing circle, and the
outer circle. Inner circle countries include countries where English is used as a first
language. Outer circle countries are those which are former colonies of the inner circle
counties, and include countries such as Nigeria and India, where English is used as
a second language or lingua franca to facilitate communication, often across ethnic
and cultural boundaries. By contrast, the developing circle countries on the periphery
include countries where English is not the first or second language, but is nonetheless
used in international communication, particularly in business. Kachru (1982) argues
that learners in these countries adhere to native speaker norms set by the inner circle
countries and that there is no real justification for this adherence. Although Kachru’s
description of the concentric circles of Englishes pushed forward the concept of
world Englishes, the classification (as Kachru concedes) is not absolute. There are
countries that do not neatly occupy one of the three circles (Crystal, 2003).
As well as concepts of English as an international language, others have argued
that world standard spoken English (WSSE) will develop as a lingua franca due to
a need for greater cross cultural communication (facilitated by internet communi-
cation) and the growth of global entities, such as the European Union and United
Nations affiliated organisations (Crystal, 2003). Crystal (2003) envisages the growth
of WSSE alongside other non-standard varieties, not necessarily in opposition to
them. Indeed, the use of English as a lingua franca in European Union institutions
already shows some signs of being a different collective non-native entity, being
of a more syllable-timed nature and having less idiomatic and colloquial language
(Crystal, 2003). Crystal (2003) emphasizes the conflicting inertias of identity and
intelligibility and argues that in cross-cultural communication, the overriding need
for intelligibility among diverse interlocutors may lead to the establishment of WSSE.
However, outside of these contexts, non-native varieties can be maintained in outer
circle countries (e.g., Nigeria). One interesting consideration is whether non-native
variety norms will be adopted by any future WSSE. For example, for many speakers
from several expanding and outer circle countries (using Kachru’s terminology), the
dental fricative is often substituted. Crystal (2003, p. 208) views the adoption of such
substitutions by a WSSE as likely because it is a widespread and common deviation
from the native variety.
The idea that native speaker varieties of English should be chosen as a model
without reference to the particular teaching context is clearly problematic (Jenkins,
2000; Seidlhofer, 2001, 2011; Walker, 2010). The choice of a native speaker goal as
the target of instruction is often based upon the flawed view of standard varieties of
English as single, constant, unchanging entities (Seidlhofer, 2011). The assumption
that native speakers of English will be easily intelligible to other native speakers
is not necessarily correct, and native speakers might not always be easily intelli-
gible to non-native speakers. As Trudgill (1999) points out, most native speakers use
non-standard varieties of English. Indeed, native speakers may well be less intel-
ligible to other native speakers and non-native speakers than non-native speakers
of English (Bayyurt, 2018; Smith & Nelson, 1985). Moreover, the adherence to a
native standard as a goal of instruction is counterproductive because it ignores the
2.12 English as an International Language and the Status … 39
very different experience of learning an L2 (Cook, 1999). Cook (1991) uses the
term multi-competence to describe the knowledge of language possessed by a person
who has L1 competence and interlanguage in the L2. It is argued that there is no
logical rationale for the L2 competence to be the same as that of a monolingual
L1 user because the state of being multi-competent entails greater complexity than
the acquisition of a monolingual L1 (Cook, 1999). Also, although second language
acquisition research has concentrated on how the L1 can affect the L2 (Lado, 1957),
there is evidence to suggest that the opposite is true in terms of phonology (Flege,
1987), as well as lexis, grammar, and processing (Balcom, as cited in Cook, 1999,
2003; Beauvillain & Grainger, 1987; Magiste, 1986). Multi-competence users of the
L2 are therefore different from monolingual L1 users in terms of both language and
language processing.
There is often an unequal power dynamic at play between native and non-native
teachers, with native speakers acting as the arbiters of what is intelligible through a
native lens (Kachru, 1985). Although both Widdowson (1994) and Matsuda (2019)
have claimed that the dominance of the native speaker is on the ebb, it is clear that the
status of the native speaker variety remains strong (Anchimbe, 2006; Cook, 1999;
Strauss, 2017). For years, monolingual native speakers of English have been favoured
as experts, often to the detriment of multilingual non-native teachers (Matsuda, 2019).
This has meant that local non-native teachers have often been disadvantaged (Kumar-
avadivelu, 2016). Phillipson (1992, 2016) uses the term linguistic imperialism to
describe the way in which a monolingual approach has been employed to the detri-
ment of local teachers. In particular, attention has been drawn to what Phillipson
terms the monolingual fallacy and the native speaker fallacy: the former empha-
sizing the lack of reference to the learners’ L1 in language teaching and the latter
referring to the idea that native speakers are needed and are deserving of preferential
treatment. These two fallacies are somewhat interdependent and have often led to
the employment of poorly qualified native speakers (Phillipson, 2016). Phillipson
(2016) points out that native English speaker teachers play only a very limited role in
several European countries (particularly Scandinavia) and claims that “most NESTs
[Native English speaker teachers] are not equipped linguistically, culturally, or peda-
gogically for their task” (p. 81). Some of the wider explanations for the salience
of native speaker models and native speakers as experts may lie in the military and
economic power exerted by native speaker countries (Crystal, 2003). Other factors
include the power exerted in the ELT industry by universities, publishers, the British
Council, and a significant number of countries where English is taught for mili-
tary purposes (Phillipson, 2016). It is this promotion of native speaker models that
contributes to the neglect of the diverse range of non-native varieties or lingua franca
models (Jenkins, 2007). Indeed, although Jenkins’ (2007) semi-structured interview
research with non-native speaker teachers of English from a variety of countries
found conflicting attitudes towards native speaker varieties, the consensus view was
that native varieties were more prestigious than lingua franca ones, so it seems that,
using Kachru’s (1985) terminology, the native speaker lens is also applied to non-
native speaker teachers (see also Sifakis & Sougari, 2005). Jenkins (2007) concludes
that this is largely due to teacher beliefs in terms of how a lingua franca model would
40 2 Pronunciation Instruction: Background, Techniques …
be perceived by others, but also believes that there is a possibility that this perception
could change in the future in certain countries. For example, the growing economic
power of China could provide impetus for a shift away from native to non-native
varieties.
Because the number of world Englishes in existence is so vast, teachers do need
to make choices in terms of the target of instruction, both in terms of productive and
receptive skills. Although it is clear that one variety of English should be chosen for
instruction, and it is perfectly possible that a native variety could be chosen if the
context and purpose of instruction meant that this were an appropriate choice, the
choice of a native variety should not be a default position. The decision should be
based on the learner needs, goals, and attitudes in the particular context of instruction
(Matsuda, 2019, p. 148), and teachers should consider carefully the motivations for
their instructional decision-making (Jenkins, 2007, p. 224). Indeed, in outer circle
countries, some have argued that the choice of a native variety is virtually an automatic
choice and that this should not be the case (Bayyurt, 2018). Although the testing of
English may still be a problematic issue because the ELT industry still largely adheres
to a native speaker standard, there are signs that some testing bodies are beginning
to take into account some aspects of regional varieties in their testing materials
(Matsuda, 2019).
The diversity of Englishes throughout the world should not be viewed as some-
thing negative, denoting decay, and non-native varieties should not be delegitimized
and viewed as inferior, temporary substitutes that are used until real native speaker
varieties are attained (Kachru, 1982). One example of this view is the correction of
the dental fricative when learners substitute it with /s/ or /z/. It is doubtful whether
such a substitution interferes with intelligibility in any substantial way (Derwing &
Munro, 2015; Jenkins, 2000). This is rather an issue of accent rather than intelligi-
bility, so should be viewed as a legitimate aspect of a different variety of English.
Various substitutions of the dental fricative already exist in speech communities in
inner circle native speaker countries. If it is patently futile and insensitive to correct
an Irish speaker of English or an Estuary English speaker who is substituting these
sounds, then it should be equally futile and insensitive to correct a non-native speaker
making the same error. Perhaps the reason that this is sometimes not the case is
based purely on the grounds of the perception of English ownership, not intelligi-
bility. However, this perception should be considered erroneous: English ownership
is shared by its diverse range of users in a vast array of contexts, with a variety of
interlocutors, in both formal and informal communication (Seidlhofer, 2011).
It is clear that if learners are not exposed to other varieties of English, this
could lead to them developing negative attitudes to these other non-native vari-
eties (Matsuura, Chiba, & Fujieda, 1999). Furthermore, this lack of exposure can
lead to learners misinterpreting or not comprehending these other varieties (Smith &
Nelson, 2006). There is evidence that exposure to different varieties of English can
make learners more familiar with those varieties, which can improve learners’ inter-
pretation of meaning when engaging in cross cultural communication (Thomson,
2018). Smith’s (1992) pilot study found that both native and non-native speakers
of English (including Chinese learners) benefited from such exposure and became
2.12 English as an International Language and the Status … 41
more effective at interpreting different varieties. It is crucial that at least some expo-
sure to the array of Englishes in existence is provided, particularly if learners are
likely to encounter those Englishes. The dominance of the native speaker in ELT
published materials, in terms of the authors themselves, the situations presented in
them, and the dialogues presented (native speaker to native speaker) is well docu-
mented (Cook, 1999; Strauss, 2017). L2 learners could therefore benefit from L2 posi-
tive role models, both in the listening activities presented and the typical situations
depicted in coursebooks (Cook, 1999). Exposure to listening extracts which present
non-native speakers employing English as a lingua franca, combined with exercises
focused on the features of language that make communication effective could assist
in providing instruction that more accurately represents English throughout the world
(Seidlhofer, 2011).
References
Aitchison, J. (2003). Linguistics (6th ed.) (Teach Yourself Books). London: English Universities
Press.
Anchimbe, E. A. (2006). The native-speaker fever in English language teaching (ELT): Pitting
pedagogical competence against historical origin. Linguistik Online, 26(1), 3–14. https://doi.org/
10.13092/lo.26.618
Archibald, J. (1987). Developing natural and confident speech: Drama techniques in the pronunci-
ation classroom. In Arnold, J. (Ed.), The teaching of pronunciation: An introduction for teachers
of English as a second language [Theme issue], TESL Talk, 17(1), 153–159; (1999). Affect in
language learning. Cambridge, UK: Cambridge University Press.
Arnold, J. (2011). Attention to affect in language learning. Anglistik. International Journal of
English Studies, 22(1), 11–22. https://doi.org/10.14746/n.2019.52.1.2
Arnold, J., & Brown, H. D. (1999). A map of the terrain. In J. Arnold (Ed.), Affect in language
learning (pp. 1–24). Cambridge, UK: Cambridge University Press.
Atherton, M. (2010). ‘The globe of language’: Thomas Prendergast and applied linguistics in the
1870s. Language & History, 53(1), 15–26. https://doi.org/10.1179/175975310X12640878626101
Aufderhaar, C. (2004). Learner views on using authentic audio to aid pronunciation: “You can just
grab some feeling.” TESOL Quarterly, 38(4), 735–746. https://doi.org/10.2307/3588293
Bailey, K. M. (1983). Competitiveness and anxiety in adult second language learning: Looking at
and through the diary studies. In H. W. Seliger & M. H. Long (Eds.), Classroom oriented research
in second language acquisition (pp. 67–102). Rowley, MA: Newbury House.
Baker, A. (2013). Integrating fluent pronunciation use into content-based ESL instruction: Two case
studies. In J. Levis & K. Levelle (Eds.), Proceedings of the 4th pronunciation in second language
learning and teaching conference (pp. 245–254). Ames, United States: Iowa State University.
Baker, A., & Burri, M. (2016). Feedback on second language pronunciation: A case study of EAP
teachers’ beliefs and practices. Australian Journal of Teacher Education, 41(6), 1–19. https://doi.
org/10.14221/ajte.2016v41n6.1
Baker, A., & Murphy, J. (2011). Knowledge base of pronunciation teaching: Staking out the territory.
TESL Canada Journal, 28(2), 29–50. https://doi.org/10.18806
Bayyurt, Y. (2018). Issues of intelligibility in world Englishes and EIL contexts. World Englishes,
37(3), 407–415. https://doi.org/10.1111/weng.12327
Bax, S. (2003). The end of CLT: A context approach to language teaching. ELT Journal, 57(3),
278–287. https://doi.org/10.1093/elt/57.3.278
42 2 Pronunciation Instruction: Background, Techniques …
Beauvillain, C., & Grainger, J. (1987). Accessing interlexical homographs: Some limitations of a
language-selective access. Journal of Memory and Language, 26(6), 658–672. https://doi.org/10.
1016/0749-596X(87)90108-2
Bell, D. M. (2003). Method and post method: Are they really so incompatible? TESOL Quarterly,
37(2), 325–336. https://doi.org/10.2307/3588507
Best, C. T. (1995). A direct realist perspective on cross-language speech perception. In S. Winifred
(Ed.), Speech perception and linguistic experience: Theoretical and methodological issues in
cross-language research (pp. 167–200). Timonium, MD: York Press.
Best, C. T., & Tyler, M. D. (2007). Nonnative and second-language speech perception: Commonal-
ities and complementarities. In O.-S. Bohn & M. J. Munro (Eds.), Language experience in second
language speech learning (pp. 13–34). Amsterdam: John Benjamins.
Bialystok, E. (1997). The structure of age: In search of barriers to second language acquisition.
Second Language Research, 13(2), 116–137. https://doi.org/10.1191/026765897677670241
Bialystok, E., & Hakuta, K. (1999). Confounded age: Linguistic and cognitive factors in age differ-
ences for second language acquisition. In D. Birdsong (Ed.), Second language acquisition and
the critical period hypothesis (pp. 1–22). Mahwah, NJ: L. Erlbaum.
Bigelow, M., & Enser-Kananen, J. (2015). The Routledge handbook of educational linguistics. New
York & Abingdon: Routledge.
Bless, H., & Fiedler, K. (2006). Mood and the regulation of information processing. In J. P. Forgas
(Ed.), Affect in social cognition and behavior (pp. 65–84). New York: Psychology Press.
Bloomfield, G. (1933). Language. London: Allen & Unwin.
Bolstad, R. (1997). Using the language of the brain. The Language Teacher Online, 21(2). Retrieved
27 November, 2019, from https://jalt-publications.org/tlt/articles/2100-using-language-brain.
Brazil, D. (1997). The communicative value of intonation in English. Cambridge: Cambridge
University Press.
Brown, A. (1991). Functional load and the teaching of pronunciation. In A. Brown (Ed.), Teaching
English pronunciation: A book of readings (pp. 211–224). London: Routledge.
Brown, H. D. (2002). English language teaching in the “post-method” era: Toward better diagnosis,
treatment, and assessment. In J. C. Richards & W. A. Renandya (Eds.), Methodology in language
teaching: An anthology of current practice (pp. 9–18). Cambridge: Cambridge University Press.
Brown, H. D. (2006). Principles of language learning and teaching (5th ed.). Upper Saddle River,
U.S.: Pearson Education.
Bygate, M. (2009). Effects of task repetition on the structure and control of oral language. In
K. Van den Branden, M. Bygate, & J. Norris (Eds.), Task-based language teaching: A reader
(pp. 249–274). Amsterdam: John Benjamins. https://doi.org/10.1075/tblt.1.15eff
Carroll, J. B. (1968). Contrastive analysis and interference theory. In J. E. Alatis (Ed.), Mono-
graph series on languages and linguistics 21. Report of the nineteenth annual round table
meeting on linguistics and language studies: Contrastive linguistics and its pedagogical impli-
cations (pp. 113–122). Washington, D.C.: Georgetown University Press. Retrieved June 3,
2018, from https://repository.library.georgetown.edu/bitstream/handle/10822/555461/GURT_1
968.pdf#page=125.
Carpenter, H., Jeon, K. S., MacGregor, D., & Mackey, A. (2006). Learners’ interpretations of
recasts. Studies in Second Language Acquisition, 28, 209–236. https://doi.org/10.1017/S02722
63106060104#
Celce-Murcia, M., Brinton, D., & Goodwin, J. (2010). Teaching pronunciation: A course book and
reference guide (2nd ed.). New York & Cambridge: Cambridge University Press.
Chan, M. J. (2018). Embodied pronunciation learning: Research and practice. The CATESOL
Journal 30(1), 47–68. Retrieved from https://files.eric.ed.gov/fulltext/EJ1174234.pdf.
Chomsky, N. (1959). Review of B. F. Skinner’s Verbal Behavior. Language 35(1), 26–58. Retrieved
from https://cogprints.org/1148/1/chomsky.htm.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: M.I.T. Press.
Chun, D. M. (1988). Teaching intonation as part of communicative competence: Suggestions for
the classroom. Die Unterrichtpraxis, 21(1), 81–88. https://www.jstor.org/stable/3530748.
References 43
Chun, D. M. (2002). Discourse intonation in L2: From theory to practice. Amsterdam and
Philadelphia: John Benjamins Publishing Company.
Cohen, Y., & Norst, M. J. (1989). Fear, dependence and loss of self-esteem: Affective barriers in
second language learning among adults. RELC Journal, 20(2), 61–77. https://doi.org/10.1177/
003368828902000206
Cook, V. J. (1991). The poverty-of-the-stimulus argument and multi-competence. Second Language
Research, 7(2), 103–117. https://doi.org/10.1177/026765839100700203
Cook, V. J. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33(2),
185–209. https://doi.org/10.2307/3587717
Cook, V. J. (Ed.). (2003). The effects of the second language on the first. Clevedon, UK: Multilingual
Matters.
Cook, V. J. (2008). Second language learning and language teaching (4th ed.). London: Hodder
Education.
Corder, S. P. (1967). The significance of learners’ errors. International Review of Applied Linguistics,
5(4), 161–170. https://doi.org/10.1515/iral.1967.5.1-4.161
Crystal, D. (2003). English as a global language. Cambridge: Cambridge University Press.
Derwing, T. M., Diepenbrook, L. G., & Foote, J. A. (2012). How well do general skills ESL textbooks
address pronunciation? TESL Canada Journal, 30(1), 22–44. https://doi.org/10.18806/tesl.v30i1.
1124
Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidence-based perspectives
for L2 teaching and research. Amsterdam: John Benjamins.
Derwing, T. M., Munro, M. J., & Wiebe, G. (1997). Pronunciation instruction for ‘fossilized
learners’: Can it help? Applied Language Learning, 8(2), 217–235. https://eric.ed.gov/?id=EJ5
67518.
Derwing, T. M., Munro, M. J., & Wiebe, G. (1998). Evidence in favour of a broad framework for
pronunciation instruction. Language Learning, 48(3), 393–410. https://doi.org/10.1111/0023-
8333.00047
Derwing, T. M., & Rossiter, M. J. (2003). The effects of pronunciation instruction on the accu-
racy, fluency and complexity of L2 accented speech. Applied Language Learning, 13(1), 1–
17. https://www.researchgate.net/publication/234570703_The_Effects_of_Pronunciation_Instru
ction_on_the_Accuracy_Fluency_and_Complexity_of_L2_Accented_Speech.
Derwing, T. M., Thomson, R. I., & Munro, M. J. (2006). English pronunciation and fluency in
Mandarin and Slavic speakers. System, 34(2), 183–193. https://doi.org/10.1016/j.system.2006.
01.005
Dlaska, A., & Krekeler, C. (2008). Self-assessment of pronunciation. System, 36(4), 506–516.
https://doi.org/10.1016/j.system.2008.03.003
Eckman, F. R. (1977). Markedness and the contrastive analysis hypothesis. Language Learning,
27(2), 315–330. https://doi.org/10.1111/j.1467-1770.1977.tb00124.x
Ellis, G. (1996). How culturally appropriate is the communicative approach? ELT Journal, 50(3),
213–218. https://doi.org/10.1093/elt/50.3.213
Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press.
Engin, M. (2017). Contributions and silence in academic talk: Exploring learner experiences of
dialogic interaction. Learning, Culture and Social Interaction, 12, 78–86. https://doi.org/10.1016/
j.lcsi.2016.11.001
Estaire, S., & Zanon, J. (1994). Planning classwork: A task-based approach. Oxford: Macmillan
Heinemann.
Firth, S. (1992). Developing self-correction and self-monitoring strategies. In P. Avery & S. Ehrlich
(Eds.), Teaching American English pronunciation (pp. 215–219). Oxford: Oxford University
Press.
Flege, J. E. (1987). The production of “new” and “similar” phones in a foreign language: Evidence
for the effect of equivalence classification. Journal of Phonetics, 15(1), 47–65. https://doi.org/
10.1016/S0095-4470(19)30537-6
44 2 Pronunciation Instruction: Background, Techniques …
Flege, J. E. (1995). Second language speech learning theory, findings, and problems. In S. Winifred
(Ed.), Speech perception and linguistic experience: Theoretical and methodological issues in
cross-language research (pp. 233–277). Timonium, MD: York Press.
Foote, J. A., & McDonough, K. (2017). Using shadowing with mobile technology to improve L2
pronunciation. Journal of Second Language Pronunciation, 3(1), 34–56. https://doi.org/10.1075/
jslp.3.1.02foo
Fraser, H. (2006). Helping teachers help students with pronunciation: A cognitive approach.
Prospect, 21(1), 80–96. Retrieved from https://hdl.handle.net/1959.14/329794.
Fraser, H. (2009). Pronunciation as categorization: The role of contrasts in teaching English /r/ and
/l/. In A. Mahboob & C. Lipovsky (Eds.), Studies in applied linguistics and language learning
(pp. 289–306). Newcastle upon Tyne: Cambridge Scholars Publishing.
Fries, C. C. (1945). Teaching and learning English as a foreign language. Ann Arbor, MI: University
of Michigan Press.
Gadd, N. (1996). Towards less humanistic English teaching. ELT Journal, 52(3), 223–234. https://
doi.org/10.1093/elt/52.3.223
Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic Books.
Gardner, R. C. (1985). Social psychology and second language learning: The role of attitudes and
motivation. London: Edward Arnold.
Gattegno, C. (1972). Teaching foreign languages in schools: The silent way (2nd ed.). New York:
Educational Solutions.
Gibson, S. (2008). Reading aloud: A useful learning tool? ELT Journal, 62(1), 29–36. https://doi.
org/10.1093/elt/ccm075
Gilbert, J. B. (1994). Intonation: A navigation guide for the listener. In J. Morley (Ed.), Pronunciation
pedagogy and theory: New views, new dimensions (pp. 36–48). Alexandria, VA: Teachers of
English to Speakers of Other Languages.
Gilner, L. (2008). Pronunciation: A review of methods and techniques. Journal of the School of
Foreign Languages, Nagoya University of Foreign Studies, 35, 93–108. https://id.nii.ac.jp/1095/
00000275/.
Gooch, R., Saito, K., & Lyster, R. (2016). Effects of recasts and prompts on second language
pronunciation development: Teaching English/r/ to Korean adult EFL learners. System, 60, 117–
127. https://doi.org/10.1016/j.system.2016.06.007
Goodwin, J. (2001). Teaching pronunciation. In M. Celce-Murcia (Ed.), Teaching English as a
second or foreign language (3rd ed., pp. 117–137). Boston, MA: Heinle & Heinle.
Goodwin, J. (2005). The power of context in teaching pronunciation. In J. Frodeson & C. Holten
(Eds.), The power of context in language teaching and learning (pp. 225–236). Boston, MA:
Thomson Heinle.
Gordon, J., & Darcy, I. (2016). The development of comprehensible speech in L2 learners: A class-
room study on the effects of short-term pronunciation instruction. Journal of Second Language
Pronunciation, 2(1), 56–92. https://doi.org/10.1075/jslp.2.1.03gor
Grant, L. (2001). Well said: Pronunciation for clear communication (2nd ed.). Boston: Heinle &
Heinle.
Hahn, L. D. (2004). Primary stress and intelligibility: Research to motivate the teaching of
suprasegmentals. TESOL Quarterly, 38(2), 201. https://doi.org/10.2307/3588378
Halliday, M. A. K. (1973). The functional basis of language. In B. Bernstein (Ed.), Class, codes
and control (Vol. II, pp. 343–366). London: Routledge and Kegan Paul.
Halliday, M. A. K. (1975). Learning how to mean: Explorations in the development of language.
London: Edward Arnold.
Hancock, M. (1995). Pronunciation games. Cambridge: Cambridge University Press.
Handschin, C. H. (1912). A historical sketch of the Gouin series-system of teaching modern
languages and of its use in the United States. The School Review, 20(3), 170–175.
Retrieved from https://www.jstor.org/stable/pdf/1077247.pdf?refreqid=excelsior%3A665bc8b
f1c9da4502e831624a5ad282f.
References 45
Harmer, J. (2003). Popular culture, methods and context. ELT Journal, 57(3), 288–294. https://doi.
org/10.1093/elt/57.3.288
Harmer, J. (2009). Is reading aloud allowed? English Teaching Professional, 65(4). Retrieved from
https://www.etprofessional.com/is-reading-aloud-allowed.
Ho, W. K., & Wong, R. Y. L. (Eds.). (2003). English language teaching in East Asia today: Changing
policies and practices. Singapore: Eastern Universities Press.
Horwitz, E. K. (1986). Preliminary evidence for the reliability and validity of a foreign language
anxiety scale. TESOL Quarterly, 20(3), 559–562. https://doi.org/10.2307/3586302
Horwitz, E. K. (1996). Even teachers get the blues: Recognizing and alleviating language teachers’
feelings of foreign language anxiety. Foreign Language Annals, 29(3), 365–372. https://doi.org/
10.1111/j.1944-9720.1996.tb01248.x
Horwitz, E. K., & Young, D. J. (Eds.). (1991). Language anxiety: From theory and research to
classroom implications. Upper Saddle River, NJ: Prentice Hall.
Howatt, A. P. R. (1984). A History of English language teaching. Oxford: Oxford University Press.
Hymes, D. H. (1966). Two types of linguistic relativity. In W. Bright (Ed.), Sociolinguistics (pp. 114–
158). The Hague: Mouton.
Hymes, D. H. (1972). On communicative competence. In J. B. Pride & J. Holmes (Eds.),
Sociolinguistics: Selected readings (pp. 269–293). Harmondsworth: Penguin.
Jenkins, J. (2000). The phonology of English as an international language. Oxford: Oxford
University Press.
Jenkins, J. (2007). English as a lingua franca: Attitude and identity. Oxford: Oxford University
Press.
Johnson, K. E. (2006). The sociocultural turn and its challenges for second language teacher
education. TESOL Quarterly, 40(1), 235–257. https://doi.org/10.2307/40264518
Kachru, B. B. (1980). Non-native literatures in English as a resource for language teaching. RELC
Journal, 11(1), 1–9. https://doi.org/10.1177/003368828001100201
Kachru, B. B. (Ed.). (1982). The other tongue: English across cultures. Urbana, IL: Illinois
University Press.
Kachru, B. B. (1985). Standards, codification and sociolinguistic realism: The English language in
the outer circle. In R. Quirk & H. Widdowson (Eds.), English in the world: Teaching and learning
the language and literatures (pp. 11–30). Cambridge: Cambridge University Press.
Kirch, M. S. (1967). Direct method and the audio-lingual approach. The French Review, 41(3),
383–385. Retrieved from https://www.jstor.org.hallam.idm.oclc.org/stable/385169.
Kjellin, O. (1999). Accent addition: prosody and perception facilitates second language learning.
In O. Fujimura, B. D. Joseph, & B. Palek (Eds.), Proceedings of LP’98 (linguistics and phonetics
conference) at Ohio State University, Columbus, Ohio, September 1998 (Vol. 2, pp. 373–398).
Prague: The Karolinum Press.
Kramsch, C. (2009). The multilingual subject. Oxford: Oxford University Press.
Krashen, S. D., & Terrell, T. D. (1983). The natural approach: Language acquisition in the
classroom. California: Pergamon Press/Alchemy Press.
Kumaravadivelu, B. (2001). Toward a postmethod pedagogy. TESOL Quarterly, 35(4), 537–560.
https://doi.org/10.2307/3588427
Kumaravadivelu, B. (2003). Beyond methods: Macrostrategies for language teaching. New Haven,
CT: Yale University Press.
Kumaravadivelu, B. (2006). TESOL methods: Changing tracks, challenging trends. TESOL
Quarterly, 40(1), 59–81. https://doi.org/10.2307/40264511
Kumaravadivelu, B. (2016). The decolonial option in English teaching: Can the subaltern act?
TESOL Quarterly, 50(1), 66–85. https://doi.org/10.1002/tesq.202
Lado, R. (1957). Linguistics across cultures: Applied linguistics for language teachers. Ann Arbor:
University of Michigan Press.
Larsen-Freeman, D. (2000). Techniques and principles in language teaching (2nd ed.). Oxford:
Oxford University Press.
46 2 Pronunciation Instruction: Background, Techniques …
LeDoux, J. (1996). The emotional brain: The mysterious underpinnings of emotional life. New
York: Simon & Schuster.
Lenneberg, E. (1967). Biological foundations of language. New York: Wiley & Sons.
Levis, J. M., & Cortes, V. (2008). Minimal pairs in spoken corpora: Implications for pronunciation
assessment and teaching. In C. A. Chapelle, Y. R. Chung, & J. Xu (Eds.), Towards adaptive
CALL: Natural language processing for diagnostics language assessment (pp. 197–208). Ames,
IA: Iowa State University.
Lewis, M. (1993). The lexical approach. Hove: Language teaching publications.
Lewis, M. (2002). Implementing the lexical approach: Putting theory into practice (2nd ed.). Boston:
Heinle.
Lightbown, P. L., & Spada, N. (2013). How languages are learned (4th ed.). Oxford: Oxford
University Press.
Littlewood, W. (1981). Communicative language teaching: An introduction. Cambridge: Cambridge
University Press.
Littlewood, W. (2004). The task-based approach: Some questions and suggestions. ELT Journal,
58(4), 319–326. https://doi.org/10.1093/elt/58.4.317
Littlewood, W. (2011). Communicative language teaching: An expanding concept for a changing
world. In E. Hinkel (Ed.), Handbook of research in second language teaching and learning (Vol.
II, pp. 541–547). Abingdon, UK: Routledge.
Liu, M., & Jackson, J. (2008). An exploration of Chinese EFL learners’ unwillingness to commu-
nicate and foreign language anxiety. The Modern Language Journal, 92(1), 71–86. https://doi.
org/10.1111/j.1540-4781.2008.00687.x
Long, M. H. (1990). Maturational constraints on language development. Studies in Second Language
Acquisition, 12(3), 251–285. https://doi.org/10.1017/S0272263100009165
Lorch, M. P. (2016). A late 19th-century British perspective on modern foreign language
learning, teaching, and reform: The legacy of Prendergast’s “Mastery System.” Historiographia
Linguistica, 43(1–2), 175–208. https://doi.org/10.1075/hl.43.1-2.06per
Lyster, R. (2007). Learning and teaching languages through content: A counterbalanced approach.
Amsterdam/Philadelphia: John Benjamins.
Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake. Studies in Second Language
Acquisition, 19(1), 37–66. https://doi.org/10.1017/S0272263197001034
Macdonald, S. (2002). Pronunciation-views and practices of reluctant teachers. Prospect, 17(3),
3–16. https://hdl.handle.net/1959.14/323083.
MacIntyre, P. D., & Gardner, R. C. (1994). The subtle effects of language anxiety on cognitive
processing in the second language. Language Learning, 44(2), 283–305. https://doi.org/10.1111/
j.1467-1770.1994.tb01103.x
Magiste, E. (1986). Selected issues in second and third language learning. In J. Vaid (Ed.), Language
processing in bilinguals: Psycholinguistic and neurolinguistic perspectives (pp. 97–122).
Hillsdale, NJ: Lawrence Erlbaum.
Marinova-Todd, S., Marshall, D., & Snow, C. (2000). Three misconceptions about age and L2
learning. TESOL Quarterly, 34(1), 9–34. https://doi.org/10.2307/3588095
Matsuda, A. (2019). World Englishes in English language teaching: Kachru’s six fallacies and the
TEIL paradigm. World Englishes, 38(1–2), 144–154. https://doi.org/10.1111/weng.12368
Matsuura, H., Chiba, R., & Fujieda, M. (1999). Intelligibility and comprehensibility of American
and Irish Englishes in Japan. World Englishes, 18(1), 49–62. https://doi.org/10.1111/1467-971X.
00121
McNeil, D. (1966). Developmental psycholinguistics. In F. Smith & G. A. Miller (Eds.), The genesis
of language: A psycholinguistic approach (pp. 15–84). Cambridge, MA: MIT Press.
Meddings, L., & Thornbury, S. (2009). Teaching unplugged: Dogme in English language teaching.
Peaslake: Delta Publishing.
Mitchell, R., & Myles, F. (2004). Second language learning theories (2nd ed.). London: Hodder
Arnold.
References 47
Morley, J. (1991). The pronunciation component in teaching English to speakers of other languages.
TESOL Quarterly, 25(3), 481–520. https://doi.org/10.2307/3586981
Morley, J. (1992). Rapid review of vowel and prosodic contexts. Ann Arbor: University of Michigan.
Neufeld, G. G. (1987). On the acquisition of prosodic and articulatory features in adult language
learning. In G. Ioup & S. Weinberger (Eds.), Interlanguage phonology: The acquisition of a
second language sound system (pp. 321–332). Cambridge, MA: Newbury House Publishers.
Neufield, G. G. (2001). Non-foreign-accented speech in adult second language learners: Does it
exist and what does it signify? ITL Review of Applied Linguistics, 133(1), 185–206. https://doi.
org/10.1075/itl.133-134.01neu
Nikolaidis, K., & Mattheoudakis, M. (2012). The pro-voc method: Combining pronunciation and
vocabulary teaching. International review of Applied Linguistics in Language Teaching, 50(4),
303–321. https://doi.org/10.1515/iral-2012-0012
Norris, J. M., & Ortega, L. (2008). Effectiveness of L2 instruction: A research synthesis and quan-
titative meta-analysis. Language Learning, 50(3), 417–528. https://doi.org/10.1111/0023-8333.
00136
Nunan, D. (1987). Communicative language teaching: Making it work. ELT Journal, 41(2), 136–
145. https://doi.org/10.1093/elt/41.2.136
Ockey, G. J., & Wagner, E. (2018). Assessing L2 listening: Moving towards authenticity.
Amsterdam/Philadelphia: John Benjamins.
Oyama, S. (1976). A sensitive period for the acquisition of a nonnative phonological system. Journal
of Psycholinguistics Research, 5(3), 261–283. https://doi.org/10.1007/BF01067377
Pennington, M. C. (1998). The teachability of phonology in adulthood: A re-examination. Interna-
tional Review of Applied Linguistics in Language Teaching, 36(4), 323–341. https://doi.org/10.
1515/iral.1998.36.4.323
Pennycook, A. (1989). The concept of method, interested knowledge, and the politics of language
teaching. TESOL Quarterly, 23(4), 589–618. https://doi.org/10.2307/3587534
Phillipson, R. (1992). Linguistic imperialism. Oxford: Oxford University Press.
Phillipson, R. (2016). Native speakers in linguistic imperialism. Journal for Critical Education
Policy Studies, 14(3), 80–96.
Piccardo, E. (2016). Council of Europe: Common European framework of reference for languages:
Learning, teaching, assessment phonological scale revision process report. Retrieved from
Council of Europe: https://rm.coe.int/phonological-scale-revision-process-report-cefr/168073
fff9.
Prabhu, N. (1990). There is no best method—why? TESOL Quarterly, 24(2), 161–176. https://doi.
org/10.2307/3586897
Purcell, E. T., & Suter, R. W. (1980). Predictors of pronunciation accuracy: A re-examination.
Language Learning, 30(2), 271–288. https://doi.org/10.1111/j.1467-1770.1980.tb00319.x
Reid, J. (1995). Learning styles in the ESL/EFL classroom. Boston: Heinle & Heinle.
Ricard, E. (1986). Beyond fossilization: A course on strategies and techniques in pronunciation for
adult learners. TESL Canada Journal, Special Issue 1, 243–249. https://doi.org/10.18806/tesl.
v3i0.1009
Richards, J. C. (1971). A non-contrastive approach to error analysis. English Language Teaching,
25(3), 204–219. Retrieved from https://doi.org/10.1093/elt/XXV.3.204
Richards, J. C., & Rodgers, T. S. (2001). Approaches and methods in language teaching (2nd ed.).
Cambridge: Cambridge University Press.
Rivers, W. (1964). The psychologist and the foreign language teacher. Chicago: University of
Chicago.
Saito, K. (2015). Variables affecting the effects of recasts on L2 pronunciation development.
Language Teaching Research 19(3), 276–300. https://doi.org/10.1177/1362168814541753
Saville-Troike, M. (2006). Introducing second language acquisition. Cambridge: Cambridge
University Press.
Scarcella, R. C., & Oxford, R. L. (1994). Second language pronunciation: State of the art instruction.
System, 22(2), 221–230. https://doi.org/10.1016/0346-251X(94)90058-2
48 2 Pronunciation Instruction: Background, Techniques …
Schumann, J. H. (1978). The acculturation model for second language acquisition. In R. C. Gingras
(Ed.), Second language acquisition and foreign language teaching (pp. 27–50). Arlinton, VA:
Center for Applied Linguistics.
Scovel, T. (1978). The effect of affect on foreign language learning: A review of the anxiety research.
Language Learning, 28(1), 129–142. https://doi.org/10.1111/j.1467-1770.1978.tb00309.x
Seidlhofer, B. (2001). Closing a conceptual gap: The case for a description of English as a lingua
franca. International Journal of Applied Linguistics, 11(2), 133–158. https://doi.org/10.1111/
1473-4192.00011
Seidlhofer, B. (2011). Understanding English as a lingua franca. Oxford: Oxford University Press.
https://doi.org/10.1002/9781405198431.wbeal0718.pub2
Selinker, L. (1972). Interlanguage. International Review of . Applied Linguistics, 10(3), 209–231.
https://doi.org/10.1515/iral.1972.10.1-4.209
Shachter, J. (1974). An error in error analysis. Language Learning, 24(2), 205–214. https://doi.org/
10.1111/j.1467-1770.1974.tb00502.x
Sheen, Y. (2006). Exploring the relationship between characteristics of recasts and learner uptake.
Language Teaching Research, 10(4), 361–392. https://doi.org/10.1191/1362168806lr203oa
Shockey, L. (2003). Sound patterns of spoken English. Oxford: Blackwell.
Shockey, L. (2011). Understanding L2 and the perspicacious Pole. In J. Arabski & A. Wojtaszek
(Eds.), The acquisition of L2 phonology (pp. 27–36). Bristol: Multilingual Matters.
Sifakis, N. C., & Sougari, A.-M. (2005). Pronunciation issues and EIL pedagogy in the periphery:
A survey of Greek state school teachers’ beliefs. TESOL Quarterly, 39(3), 467–488. https://doi.
org/10.2307/3588490
Singleton, D. (1989). Language acquisition: The age factor. Clevedon: Multilingual Matters.
Skinner, B. F. (1957). Verbal communication. New York: Appleton-Century-Crofts.
Smith, L. E. (1976). English as an international auxiliary language. RELC Journal, 7(2), 38–42.
https://doi.org/10.1177/003368827600700205
Smith, L. E. (Ed.). (1983). Readings in English as an international language. Oxford: Pergamon.
Smith, L. E. (Ed.). (1987). Discourse across cultures: Strategies in world Englishes. London:
Prentice Hall.
Smith, L. E. (1992). Spread of English and issues of intelligibility. In B. B. Kachru (Ed.), The other
tongue: English across cultures (2nd ed., pp. 75–90). Urbana, IL: University of Illinois Press.
Smith, L. E., & Nelson, C. (1985). International intelligibility of English: Directions and resources.
World Englishes, 4(3), 333–342. https://doi.org/10.1111/j.1467-971X.1985.tb00423.x
Smith, L. E., & Nelson, C. L. (2006). World Englishes and issues of intelligibility. In B. B. Kachru, Y.
Kachru, & C. Nelson (Eds.), The handbook of world Englishes (pp. 428–445). Oxford: Blackwell.
Sparks, R. L., & Ganschow, L. (1995). A strong inference approach to causal factors in foreign
language learning: A response to MacIntyre. The Modern Language Journal, 79(2), 235–244.
https://doi.org/10.2307/329622
Steinberg, F. S., & Horwitz, E. K. (1986). The effect of induced anxiety on the denotative and
interpretative content of second language speech. TESOL Quarterly, 20(1), 131–136. https://doi.
org/10.2307/3586395
Stevick, E. W. (1980). Teaching languages: A way and ways. Rowley, MA: Newbury House.
Stockwell, R., & Bowen, D. (1965). The sounds of English and Spanish. Chicago: University of
Chicago Press.
Strauss, P. (2017). “It’s not the way we use English”-Can we resist the native speaker stranglehold on
academic publications? Publications, 5(27), 1–7. https://doi.org/10.3390/publications5040027
Thomson, R. I. (2018). High variability [pronunciation] training (HVPT): A proven technique
about which every language teacher and learner ought to know. Journal of Second Language
Pronunciation, 4(2), 208–231. https://doi.org/10.1075/jslp.17038.tho
Thornbury, S. (1996). Teachers research teacher talk. ELT Journal, 50(4), 279–289. https://doi.org/
10.1093/elt/50.4.279
References 49
Thornbury, S. (2000). A dogma for EFL. IATEFL Issues, 153(2). Retrieved January 28,
2019, from https://esol.britishcouncil.org/sites/default/files/attachments/informational-page/Ado
gmaforEFL.pdf.
Thornbury, S. (2010, December 5). R is for repetition [Blog post]. Retrieved from https://scottthor
nbury.wordpress.com/2010/12/05/r-is-for-repetition/.
Tomlin, R. S. (1990). Functionalism and second language acquisition. Studies in Second Language
Acquisition, 12(2), 155–177. https://doi.org/10.1017/S0272263100009062
Trudgill, P. (1999). Standard English: What it isn’t. In T. Bex & R. J. Watts (Eds.), Standard English:
The widening debate (pp. 117–127). London: Routledge.
Venkatagiri, H., & Levis, J. (2007). Phonological awareness and speech comprehensibility: An
exploratory study. Language Awareness, 16(4), 263–277. https://doi.org/10.2167/la417.0
Vygotsky, L. S. (1962). Thought and language. Cambridge MA: MITT Press.
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes.
Cambridge: Cambridge University Press.
Walker, R. (2010). Teaching the pronunciation of English as a lingua franca. Oxford: Oxford
University Press.
Wang, Q. (2007). The national curriculum changes and their effects on English language teaching
in the People’s Republic of China. In J. Cummins & C. Davison (Eds.), International handbook
of English language teaching (pp. 87–105). Boston: Springer Science & Business Media.
Widdowson, H. G. (1972). The teaching of English as communication. ELT Journal, 27(1), 15–19.
https://doi.org/10.1093/elt/XXVII.1.15
Widdowson, H. G. (1978). Teaching language as communication. Oxford: Oxford University Press.
Widdowson, H. G. (1994). The ownership of English. TESOL Quarterly, 28(2), 377–389. https://
doi.org/10.2307/3587438
Widdowson, H. G. (2003). Defining issues in English language teaching. Oxford: Oxford University
Press.
Williams, M. W., & Burden, R. L. (1997). Psychology for language teachers: A social constructivist
approach. Cambridge: Cambridge University Press.
Willis, J. (1996). A framework for task-based learning. London: Longman.
Wilson, K. (2010, October 14). Reading aloud in class is a complete waste of time—discuss [Blog
post]. Retrieved from https://kenwilsonelt.wordpress.com/2010/10/14/reading-aloud-in-class-is-
a-complete-waste-of-time-discuss/.
Wong, R. (1987). Learner variables and prepronunciation considerations in teaching pronunciation.
In J. Morley (Ed.), Current perspectives on pronunciation: Practices anchored in theory (pp. 13–
28). Alexandria, VA: Teachers of English to Speakers of Other Languages.
Young, D. J. (1990). An investigation of students’ perspectives on anxiety and speaking. Foreign
Language Annals, 23(6), 539–553. https://doi.org/10.1111/j.1944-9720.1990.tb00424.x
Young, D. J. (1991). Creating a low-anxiety classroom environment: What does the anxiety research
suggest? The Modern Language Journal, 75(4), 426–439. https://doi.org/10.2307/329492
Young, D. J. (Ed.). (1999). Affect in foreign language and second language learning: A practical
guide to creating a low-anxiety classroom atmosphere. Boston, MA: McGraw-Hill College.
Yu, L. (2001). Communicative language teaching in China: Progress and resistance. TESOL
Quarterly, 35(1), 194–198. https://doi.org/10.2307/3587868
Chapter 3
Suprasegmental Instruction
and Intelligibility
3.1 Introduction
the syllable do vary, Parker and Graham (2009, p. 18) identify the common feature
as the presence of a vowel sound (with very few exceptions) at the centre of the
syllable.
The issue of stress is obviously important when dealing with disyllabic or polysyl-
labic words. Stress is considered a suprasegmental aspect of pronunciation because
it involves the stressing of a syllable, not merely sounds (Daniel, 2011). A stressed
syllable will be louder or longer than other syllables in the word; in other words,
it will be more prominent. Roach (1983, p. 85) emphasizes the greater articulatory
force required when stressing a syllable. The syllable with primary stress has the most
prominence, and other syllables will either have secondary stress or be unstressed.
Stress in English is relatively complex and, as Roach points out (1983, p. 88), the
rules of where the primary stressed syllable will fall involve several exceptions. This
syllabic complexity is also augmented by several word class pairs that have different
stress patterns (and also the pronunciation of single sounds within those syllables),
for example, record (/’rEkO d/ noun) and record (/rI kO d/ verb).
to/O / in connected speech, for example, two or three (/tu wO θri /); when/j/ is added to
facilitate the move from/i / to/O /, for example, three or four (/θri jO fO /); and when/r/ is
added to facilitate a move from/O / to/O / and make the transition smoother and easier to
articulate, for example, four or five (/fO rO faIv/). These features of connected speech
are important features of prosody which native speakers acquire and use, and may
therefore be of importance when considering what might be included in pronuncia-
tion instruction for foreign learners. However, whether or not intelligibility is affected
by these suprasegmental elements is an important consideration, particularly in light
of the work of Jenkins (2000). This is discussed later in the chapter.
Intonation is a vital aspect of prosody that native speakers acquire and use, and may
also be important for foreign learners. Parker and Graham (2009) define intonation as
follows: “The element of phonology concerned with the use of pitch-in other words
the system which determines the various musical notes at which different parts of
utterances are said” (p. 210).
However, this narrow definition of intonation is usually expanded to include
rhythm, loudness and voice quality (Nolan, 2006). Halliday (1967) makes a distinc-
tion between tonality, tone, and tonicity. Tonality is used to describe tone units or
intonational phrases that are constrained both physically (speakers need to breathe at
certain points whilst speaking) and cognitively (utterances are used to express ideas).
These intonation boundaries are marked by a fall in pitch and a reduction in speed,
for example, after the final accented syllable (Halliday, 1967). Brazil (1997) argues
that boundaries are also marked by a pause. Tone is used to describe the fall or rise in
pitch during utterances, and tonicity is used to describe the placement of the nucleus
in an intonational phrase.
Intonational phrases consist of a “single unbroken tune” (Parker & Graham, 2009,
p. 23) and a prominent nucleus. The nucleus is the most prominent syllable in the
phrase, and the change of pitch that begins with the nucleus and ends with the phrase
is known as the nuclear tone. The nucleus will necessarily be louder and longer in
duration than other syllables, and as with contrastive stress, the default positioning
of the nucleus in the phrase can be changed to give prominence to a different word,
54 3 Suprasegmental Instruction and Intelligibility
thus changing the meaning. The nucleus will have the most prominent pitch accent
(Cruttenden, 1997), but the placement of prominence is not governed by language,
but the interlocutor (Underhill, 1994). Before the nucleus, there may also be a head
that stretches from the first stressed syllable to the nucleus; a pre-head, consisting
of the unstressed syllables before the head; and a tail, which includes the syllables
that follow the nucleus (Underhill, 1994, p. 63). For native speakers, the nuclear tone
is of vital importance in interaction, with different tones used for varying types of
utterances (Parker & Graham, 2009).
Much work on intonational phrases is shaped by Halliday (1985), who viewed
them as informational units giving appropriate intonation for grammatical functions,
e.g., the rising tone of polar questions. However, this functional approach has little
support in terms of evidence (Levis, 2018; McCarthy, 1991, p. 106). For example,
although Halliday’s approach seems applicable in lists, where a falling nuclear tone
is used to complete the list, the use of a rising tone in polar questions is far from
consistent (Fries, 1964; Lee, 1980; Levis, 1999a), and attitudinal approaches, which
view the role of nuclear tones as that of expressing emotion, seem to have more
support (McCarthy, 1991, p. 107). Examples of such nuclear tones are a low, falling
or flat tone to signify boredom, and a fall-rise pattern to indicate sarcasm (Parker &
Graham, 2009, pp. 72 and 74). Ladd (1980) also outlines how L1 users of English use
intonation to express what they want in a subtle manner by implicature, and Nolan
(2006) gives several other examples of how quite complex changes in meaning can
be conveyed by subtle changes in pitch and tone contours. However, the difficulty
lies in how to make learners aware of and use the vast array of tones (Underhill, 1994,
p. 83). For example, although there is a relationship between attitudes or emotions
and intonation, this relationship is not always consistent, can vary from speaker to
speaker, and can be somewhat indirect (Levis, 2018, p. 261). Contrastive prominence,
used to denote new information (as opposed to information that has already been
provided), is something that is much easier to identify and is not subject to such
variability (Levis, 2018, p. 261).
In terms of discourse, it is clear that intonation can play a vital role in organizing
discourse in interactions, aiding comprehensibility in communication (Chun, 1988).
A simple rise can be employed in discourse to signify that the speaker wishes to
continue their turn. Conversely, a fall-rise can signal that the speaker welcomes input
(Parker & Graham, 2009, p. 74). Researchers such as Brown and Yule (1983), who are
more interested in analysing pitch variations over longer stretches of discourse, refer
to these stretches as paratones. Paratone analysis is driven by topic rather than by the
structure of information, and the identification of the nuclear tone is abandoned, with
all prominent syllables marked (McCarthy, 1991, p. 101). Paratones can be defined
as “the phenomenon whereby speakers expand their pitch range at the beginning of
a new topic and compress it at the end” (Wennerstrom, 1998, p. 5). They can be vital
tools in organizing longer stretches of speech into distinct sections and can signal
topic change. This means they can assist the listener in understanding discourse
meaning and aid comprehensibility.
It is clear that, for native speakers of English, prosody plays a vital role in
conveying meaning, indicating turn-taking intentions in discourse, and expressing
3.2 What Are Suprasegmentals? 55
emotions (Roach, 1983). Indeed, Nolan (2006) argues that the influence of prosody
on English is particularly significant, largely due to the relatively clear distinction
between elements of an utterance which are prominent and those which are not. The
distinction in English is therefore more salient than, for example, in French.
along with drilling and repetition, almost always with a native speaker model as the
target.
Chomsky’s critique of behaviourist theory (1959) had a profound effect on pronun-
ciation teaching as it played a pivotal role in questioning the assumptions that under-
pinned the direct method and the audiolingual method. Chomsky argued that the
reduction of language learning to a stimulus–response model was erroneous because
it could not explain the absorption of vast amounts of language sounds and grammar
rules. Instead, it was argued that children had some kind of innate language learning
ability, which Chomsky termed the language acquisition device, that facilitated the
acquisition of the L1 (Chomsky, 1965). This new view of language acquisition gave
rise to a number of communicative approaches which concentrated upon meaning
rather than study at the phonological level (e.g., Widdowson, 1978). Techniques such
as minimal pair drilling were therefore viewed as being of little theoretical or prac-
tical value as they did not deal with meaning (Celce-Murcia et al., 2010, p. 11). Other
classroom activities, such as discourse analysis (Brazil, 1997) and chunking activi-
ties (Lewis, 1993, 2002), aim to improve pronunciation in terms of suprasegmental
features, such as intonation and rhythm (see also Szczepek-Reed, 2012). Most impor-
tantly, these activities are placed within a communicative context. For example, it
is recommended that chunking activities should take place after a general listening
comprehension so that meaning is provided (Brazil, 1994, p. 3).
The approach to pronunciation teaching with communicative approaches is there-
fore one which emphasizes communication, not the imitation of single sounds. A
native speaker goal is viewed as unattainable and not conducive to fulfilling the
communicative goals of learners (Chun, 2002; Levis, 2005, 2018; Tarone, 2005). As
both Kachru (1992) and Jenkins (1998, 2000, 2007) have pointed out, learners may
be learning English in the context of international communication as a lingua franca,
and these learners will not benefit from instruction adhering to a native speaker stan-
dard. However, if the new target is one of intelligibility rather than a native speaker
standard, there are a number of issues that need to be addressed.
most commonly used in the speech community (Derwing & Munro, 2015; Munro &
Derwing, 1995).
One difference between these three concepts is that it could be said that intelligi-
bility can be measured more reliably: both comprehensibility and accentedness rely
on the perception of listeners, and are therefore more subjective in nature, whereas
intelligibility can be assessed by listeners by getting them to write down what they
have heard. However, to a certain extent, subjectivity still exists because intelligibility
scores will depend on individual raters. More crucially, accentedness does not neces-
sarily affect intelligibility: It is possible for a speaker to have a strong accent and still
be intelligible and comprehensible (Derwing & Munro, 2015). Put simply, intelligi-
bility is the understanding of the utterance, comprehensibility is the perceived level
of difficulty or strain the listener experiences in trying to interpret the utterance, and
accentedness is the listener’s perception of the difference between the native variety
and the speaker.
Accent can certainly play a vital role in the value judgements that listeners form
about the intelligence, education, friendliness, and social attractiveness of the speaker
(Lindemann, 2017). There does seem to be a demonstrable bias against non-native
accents in the above-mentioned categories (Eisenchlas & Tsurutani, 2011; Linde-
mann, 2003), and this bias also exists in the judgement of non-native listeners too
(McKenzie, 2010). Furthermore, it seems to be the case that this bias is greater
when the speaker is non-white, and Buckingham (2014) found that listeners gave
higher ratings to a speaker when they were told that the speaker was from the UK.
However, although it is clear that accent does play a role in the listener’s perception
of the intelligence, education, and social attractiveness of the speaker, Derwing and
Munro (2011) argue that far too much emphasis is placed on accentedness and that
this emphasis is unproductive because possessing a strong accent does not neces-
sarily have an effect on intelligibility. One example of this could be learners being
unaware of elision or assimilation. Although such a lack of awareness may lead to
them possessing a noticeable foreign accent, it is highly unlikely that it will impede
intelligibility (McCarthy, 1991, p. 89).
Jenkins (2000, 2002) attempted to establish the aspects of pronunciation that were
necessary for intelligibility between non-native speakers by analysing communi-
cation breakdowns. As a result, a Lingua Franca Core was proposed, consisting
of nuclear stress, contrastive stress, vowel length distinctions, and most consonant
sounds. One of the most significant aspects of the core was that certain elements,
such as weak forms, were eliminated, either because they were considered unneces-
sary for intelligibility or because they were not considered teachable. Similarly, the
dental fricative sounds/θ/ and/ð/, often substituted with/t/ and/d/ or/f/ and/v/ by some
learners, were not included as they were not necessary for intelligibility. Jenkins’
58 3 Suprasegmental Instruction and Intelligibility
work on intelligibility and the Lingua Franca Core was valuable because it forced a re-
evaluation of common teaching practices. Walker (2010) sees the focus on high func-
tional load segmentals as a key advantage due to the bottom-up processing employed
by L2 English learners (as opposed to native speakers). However, it did draw criticism
for omitting certain suprasegmental features from the core. The lack of attention to
weak forms and vowel reduction for receptive study is viewed as a weakness because
learners need to be aware of such features so that they can understand native speaker
varieties (Levis, 2018, p. 259). Kennedy’s study (as cited in Levis, 2018, p. 46) in a
lingua franca context found that word stress was the only suprasegmental feature to
impede intelligibility, and it is perhaps the omission of this suprasegmental feature
that has attracted the most criticism (Dauer, 2005; Field, 2005; Levis, 2018). Others
have pointed to the lack of empirical rigour in Jenkins’ research to justify the choices
made in terms of the items included (Szpyra-Kozłowska, 2015, pp. 77–83). There is
also evidence that learners actually prefer a native model (Timmis, 2002; Waniek-
Klimczak, Rojczyk, & Porzuczek, 2015). Derwing (2003) found that the vast majority
of ESL learners in her study in Canada wanted to attain native-like pronunciation.
Also, there is evidence that non-native speaker teachers seem to favour a native-
speaker model of English (Jenkins, 2007; Sifakis & Sougari, 2005). Although it may
be tempting to adhere to the advice of Harmer (2001) and provide these students
with a native-like goal, the evidence suggests that, in some contexts at least, depar-
ture from a native standard goal can enhance intelligibility. A lingua franca can be
defined as “any use of English among speakers of different first languages for whom
English is the communication medium of choice” (Seidlhofer, 2011, p. 7). Seidlhofer
(2011) argues that critics of lingua franca models not only fail to acknowledge the
fallacy of native English as a supreme, unitary entity, but also fail to appreciate that
native speakers can make use of a lingua franca as well as non-native speakers. There
are certainly problematic areas within Jenkins’ core, particularly with regard to the
omission of word stress, and any proposed lingua franca proposal will face difficulty
because of a lack of agreement on which elements should be included. However, the
alternative is to acquiesce to the imposition of a variety of English that may not be
beneficial for many of its users. Some have argued that Jenkins’ (2000) criticism of
native varieties being imposed from the top down could just as easily be levelled
at any lingua franca core proposal (Jarosz, 2019, p. 18). Although this criticism is
certainly valid, the key point is surely who or what is attempting to impose a particular
language variety and for what purpose? A lingua franca proposal is one that attempts
to include features necessary for the mutual intelligibility of all speakers, a native
speaker variety does not. The fact that reaching a consensus in terms of what consti-
tutes international English has proved problematic is largely due to the attempt to
accommodate a multiplicity of learner contexts and learner needs in those contexts.
This democratization of English language learning, although far from straightfor-
ward, is surely a valuable process, and even when it is not possible to establish an
English as a lingua franca that is broadly accepted globally, teachers should make
pragmatic classroom decisions based on the research (e.g., by not concentrating
on low functional load segmentals) with due consideration to the learner context
and learner goals. Research into L2 learners from several Asian and South East
3.5 The Lingua Franca Core 59
Asian countries, and also in African countries, seems to show that a form of interna-
tional English is being adopted and used because in contexts where interaction with
native speakers is not required, adherence to a native-like target is counter-productive
(Deterding & Kirkpatrick, 2006; Gramley & Patzold, 2004). There also appears to be
a broad consensus that a native speaker goal can be unrealistic in many contexts and
can lead to learner disillusionment (Celce-Murcia et al., 2010; Derwing & Munro,
2005; Levis, 2018; Wagner & Toth, 2017, p. 87).
There are a number of issues that are important to consider when examining the
concept of intelligibility. Firstly, the audience is an incredibly important considera-
tion: If learners are to be intelligible to other non-native speakers, they will require
exposure to non-native speaker materials. According to the Levis model of intel-
ligibility (2005), non-native speakers find it more difficult to communicate with
other non-native speakers than with native speakers. This is an important consid-
eration as teachers may not only be native speakers, but skilled practitioners who
are better able to understand non-native speakers (Kennedy & Trofimovich, 2008,
p. 460; Parker & Graham, 2009). Teachers should therefore be aware of this and be
willing to intervene to assist students should communication breakdown between
themselves and their peers (Kennedy & Trofimovich, 2008, pp. 481–482). This is
particularly important in the light of self-assessment research, which suggests that
learners who have a low level of comprehensibility tend to overestimate their own
comprehensibility levels; the opposite is true of those with higher comprehensibility
levels (Trofimovich, Isaacs, Kennedy, Saito, & Crowther, 2016). In short, students
may have a false perception of their own comprehensibility to others, and the teacher
should play an important role in sensitively bridging the gap between the students’
perception and the reality.
It is also crucial that listening materials reflect a range of English accents, both
native and non-native. Deterding’s (2005) research from Singapore discovered that
even though students regularly substituted/f/ and/v/ for/θ/ and/ð/, they had much
difficulty in comprehending these substitutions in Estuary English. This shows that
both native and non-native accents are important for raising student awareness and
that we should not assume that just because students use those substitutions, they
will be able to understand them when used in a different variety. It also seems
appropriate to draw a distinction between listening and speaking in other ways.
Japanese speakers can learn to better identify the differences between/l/ and/r/, and
this receptive training has been shown to assist the production of these often problem-
atic phonemes (Bradlow, Pisoni, Akahane-Yamada, & Tokhura, 1997). Furthermore,
Jenkins (2002) points out that receptive instruction can be valuable, even when it is
not successful in causing a positive change in learner production because it can be
helpful in aiding acquisition outside the classroom at a later stage. This is particu-
larly true of suprasegmentals, which appear to be acquired subconsciously. Therefore,
60 3 Suprasegmental Instruction and Intelligibility
nonetheless cause intelligibility problems in crucial words that learners might need to
use. Deterding (as cited in Levis, 2018, p. 244) gives the example of the word usually,
which can cause Chinese learners problems because of the segmental/Z /. Instead of
drilling a phoneme that occurs relatively rarely in English, Deterding suggests getting
learners to say You Shirley. This should produce an intelligible realization of the word
that is understandable for listeners. Levis (2018, p. 244) also proposes using other
communication strategies: the -ty and -teen endings of numbers can be made clearer
by enunciating the single digits after the number, for example, by saying fourteen:
that’s one, four.
Segmental instruction has traditionally not discriminated between high and low
functional load phonemes, and relatively recent research has found that many of the
words used to generate minimal pair examples in coursebooks were words that were
rarely used by English speakers in practice (Levis & Cortez, 2008). It is clear that a
universal approach to segmental instruction that does not recognize the diverse range
in the gravity of errors, but instead adheres to a mythical native standard variety as a
yardstick to measure learners, is counter-productive in many teaching contexts, and
particularly in the context of short EAP pre-sessional courses. The aim should be to
ensure that learners can communicate with native and non-native speakers on their
destination courses and that they can also communicate in the local area where the
university is based. The idea that learners with an IELTS score of 5 will be able to
produce native-like English after ten weeks of instruction, most of which is based
on reading and writing, is clearly unworkable in any case.
There does seem to be substantial evidence that suprasegmentals play a vital role
in affecting intelligibility and that suprasegmental instruction can enable learners to
become more intelligible (Derwing et al., 1997; Derwing & Rossiter, 2003; Hahn,
2004). However, it is clear that, as with segmentals, suprasegmental features can also
be prioritized in terms of those features that have a more serious impact on intelligi-
bility and comprehensibility. Features that merely impact accentedness should not be
a priority. The experimental study of Hahn (2004) showed the importance of primary
stress and its effect, not only on intelligibility, but also sociolinguistically, in terms
of role relationships in discourse. The research by Derwing et al. (1997) is particu-
larly encouraging, showing an improvement in eight of thirteen fossilized learners in
either intelligibility, comprehensibility or accentedness. The researchers attributed
this success to the lack of focus on segmentals in favour of a broader suprasegmental
approach. Furthermore, these improvements in scores (based on listeners before and
after estimates and an intelligibility transcription task) occurred over a relatively
short period of instruction: three months.
The work of Pickering (2009) is significant because it shows that certain features
of intonation can be effective in aiding interaction, even when communication occurs
between non-native speakers only. Both pitch movement and key (low, middle, or
3.7 Suprasegmental Instruction 63
high) were found to be significant tools that students used to manage discourse,
particularly when communication broke down. Similarly, pitch variation has been
shown to be a useful tool in aiding the comprehensibility of academic preparations
and longer stretches of speech, with greater pitch variation being easier for listeners
to decode, and monotone deliveries interpreted by listeners as a sign that speakers
lacked involvement in the topic (Pickering, 2001; Wennerstrom, 1994). Assisting
learners in managing longer speaking stretches and making their output easier to
process may therefore make learner output more comprehensible for a range of
listeners.
However, in spite of this substantial evidence, there still seems to be a firm
emphasis on segmentals in terms of the research areas addressed (Piasecka, 2011,
p. 135; Thomson & Derwing, 2014). Thomson and Derwing’s (2014) review of
pronunciation research showed that over half of the studies they reviewed not only
assumed a native speaker goal, but also concentrated on segmentals. Levis (2018)
stresses the interdependence of segmentals and suprasegmentals, argues that the
ongoing debate is a false one, and states that a balanced approach is required in most
teaching contexts. Indeed, Derwing and Munro (2015) specifically address this issue
because the earlier publications (Derwing et al., 1998; Derwing & Rossiter, 2003)
were erroneously interpreted as being ones suggesting that segmental instruction
had little or no value. This interdependence is supported by research which shows
that intelligibility is affected much more when an incorrect placement of lexical
stress is accompanied by changes in vowel quality (Field, 2005). Although it seems
wise to have a nuanced approach that includes both segmentals and suprasegmentals
(Celce-Murcia et al., 2010; Derwing & Munro, 2015; Derwing et al., 1998, p. 407),
in adult teaching contexts, particularly on relatively short courses of instruction, it
would seem more appropriate to focus on suprasegmental instruction (McNerney
& Mendelsohn, 1992, p. 186; Piasecka, 2011, p. 136). Adults may well be at a
disadvantage when compared to children in terms of their ability to benefit from
pronunciation instruction (Cook, 2008, p. 148; Singleton, 1989), and it seems highly
unlikely for learners to attain native-like pronunciation after early childhood (Scovel,
2000). However, perhaps the focus should be on what adults can achieve (perfectly
intelligible production) rather than dwelling on what might not be possible (Derwing
& Munro, 2015, p. 53). Also, as Johnstone (2002, p. 13) points out, adult learners
do have the ability to cope with discourse, and this ability makes suprasegmental
instruction particularly appropriate for them. Alternatives to the listen and repeat
type activities associated with segmental instruction could include the utilization of
reading aloud and self-directed learning (Ricard, 1986; Underhill, 1994, p. 202),
activities generated largely by the learners themselves and guided by the teacher
(Meddings & Thornbury, 2009), discourse-based activities (McCarthy, 1991), role-
playing (Celce-Murcia, 2010), intonation exercises (Brazil, 1997), down chunking
(Lewis, 1993, 2002; Piccardo, 2016, p. 11), and listening discrimination exercises
(Bowen & Marks, 1992; Halliday & Greaves, 2008).
This focus on how suprasegmentals can play a vital role in promoting intelligibility
and comprehensibility, as well as being key to providing the type of instruction
that is much more suited to adult learners, seems to show that they should be a
64 3 Suprasegmental Instruction and Intelligibility
vital element of any type of English language course, and particularly ones where
speaking skills form a prominent element (Piasecka, 2011, p. 136). There is also
evidence that suprasegmental instruction may be beneficial in the acquisition of
reading skills because phonological decoding and other phonological processes (e.g.,
learning vocabulary through rhyming) are inextricably linked with the acquisition
of reading skills (Kusiak, 2011; Piasecka, 2011). EAP courses should aim to equip
learners with the skills to interact verbally with a variety of both L1 and L2 English
interlocutors as it is likely that their fellow learners will be comprised of these two
groups. Courses of instruction should therefore focus on suprasegmental instruction
with intelligibility and comprehensibility as the goals of instruction. However, there
is some evidence that, in spite of research suggesting such a focus is desirable, with
intelligibility as the goal, it is sometimes not apparent in the classroom (Derwing
& Rossiter, 2002; Foote et al., 2011; Piasecka, 2011; Szpyra-Kozłowska, 2008; see
also Sifakis and Sougari (2005), for a description of conflicting teacher beliefs and
practices regarding the goal of instruction in the state school sector in Greece). This
lack of an intelligibility-based focus in the classroom could be explained by the
fact that those errors which do not necessarily affect intelligibility may nonetheless
still be more noticeable for teachers (Levis, 2018, p. 29). Also, the implementation
of instruction based on intelligibility requires careful thought in terms of priorities,
whereas instruction based on nativeness, although less effective, is easier for teachers
to implement as this prioritization is not required (Levis, 2018, p. 31).
Piccardo (2016, p. 14) outlines other possible explanations for the lack of mean-
ingful instruction. These include the negative stereotypes that teachers might have
concerning the instruction or assessment of pronunciation, and a general lack of
research in the area of pronunciation (and particularly research with a pedagog-
ical focus), coupled with the technical nature of much research into phonological
phenomena that may be difficult for practitioners to discern and transfer in a prac-
tical setting. Limited training, materials, guidance, and support are also cited. The
idea that there is a lack of guidance for teachers was certainly apparent in the teacher
interviews conducted by Macdonald (2002), and was also noted by Foote et al. (2011)
in their survey of ESL programmes in Canada. Others have pinpointed the specific
areas of intonation (Couper, 2017; Sonsaat, 2017) and stress (Couper, 2017) as those
where instructors seem less confident, and the lack of provision of effective feedback
and scaffolding has also been shown to be apparent (Baker, 2014; Baker & Burri,
2016; Couper, 2017). Also, the materials that are available for pronunciation instruc-
tion in coursebooks could be problematic due to a concentration on words rarely
used (Cook, 2008, p. 82; Levis & Cortez, 2008), a lack of guidance for the teacher
and a lack of variety in the types of task (Derwing et al., 2012), and inappropriate
tasks: Levis (1999b) found that some coursebooks presented intonation patterns that
even native speakers struggle to identify.
Although Pawlak’s (2011) study, based on learner diaries, took place in an EFL
setting (undergraduates in Poland), the findings are of note because a number of
issues involving learner and institutional goals, as well as pronunciation materials
and instruction techniques arose. In particular, it was found that learner goals were
3.7 Suprasegmental Instruction 65
types of both non-verbal communication (e.g., eye contact and smiling) and verbal
communication by the teacher (referring to the student by name, using personal exam-
ples, and using humour) can be instrumental in fostering immediacy with learners,
i.e., the feeling of communicative closeness and security (Gorham, 1988). Also,
the way that teachers use their voice is of utmost importance; utilizing a range of
suprasegmentals not only models these features for learners, but it also shows that
they are approachable (Gregersen, 2011). In particular, teachers should avoid a mono-
tone delivery and use vocal cues to emphasize salient points for learners (Richmond
& McCroskey, 2004). By enhancing teacher immediacy and using suprasegmentals
clearly in classroom talk, teachers can promote positive affective factors in their
learners (Gregersen, 2011).
Although the impact of Jenkins’ (2000) work was instrumental in causing a re-
examination of the target of instruction in situations where non-native speakers need
to be intelligible to each other, the effect of adopting Jenkins’ core wholesale would
be that many suprasegmental features would be omitted. Although nuclear stress,
contrastive stress, and initial consonant clusters are within the core, word stress,
intonational tones, rhythm, and final consonant clusters are excluded. Such a whole-
sale adoption of the core is unlikely, and there is certainly evidence to suggest that
instruction in word stress can benefit learners and should also be included in pronun-
ciation instruction (Field, 2005; Waniek-Klimczak, 2015). It is perhaps the exclusion
of word stress from the core that has been opposed the most. Levis (2018, p. 124)
and Isaacs and Trofimovich (2012) see word stress as an essential component of
intelligibility-based instruction, and the work of Cutler, Dahan and van Donselaar
(1997), Field (2005), and Zielinski (2008) lends weight to the inclusion of word
stress in an intelligibility-based syllabus. The review of the research by Cutler et al.
(1997) suggests that word stress patterns can play a pivotal role in word recognition
and therefore greatly enhance both intelligibility and comprehensibility. Listeners
may struggle with stress patterns that are not familiar and may either struggle to
interpret the words (affecting comprehensibility) or misidentify words completely
(affecting intelligibility). Both Field (2005) and Richards (2016) found that a shift
in the placement of word stress, as well as changes in vowel quality could gravely
impact intelligibility for both native and non-native listeners. However, it is important
to note that not all word stress errors are equal in terms of the gravity of their impact
on intelligibility. There is evidence that an erroneous placement of stress to the right
of disyllabic words creates greater unintelligibility than one to the left. For example,
folLOW, instead of FOllow, would be more unintelligible than BElow, instead of
beLOW (Field, 2005). Also, Cutler (1986) found that stress shifts in noun–verb pairs
(e.g., PERmit/perMIT ) had little noticeable impact on intelligibility. This appears to
hold true even when the incorrect stress placement includes vowel reduction (Small,
Simon, & Goldberg, 1988). However, Field’s (2005) research showed that certain
types of word stress errors could have a detrimental effect on intelligibility. This
is particularly significant in view of the crucial role that stress placement has in
terms of word recognition, particularly for native speaker listeners (Cutler, 2012;
Zielinski, 2008), and also in view of the evidence that listeners utilize phonological
detail before attending to semantic information (Levis, 2018, p. 102). Indeed, the
3.7 Suprasegmental Instruction 67
Council of Europe (2018, p. 136) includes both sentence stress and word stress in its
assessment descriptors for pronunciation. This suggests that it views word stress as
both learnable and teachable, particularly in light of the phonological scale revision
that led to the new framework of 2018, which stresses that phonological competence
is teachable (Piccardo, 2016, p. 18).
Levis (2018) suggests certain guidelines for the promotion of intelligibility in
suprasegmental instruction. These are guided by the basic principles that instruction
and the techniques chosen to facilitate learning should only include elements that
assist intelligibility, and that both collective and individual learner needs should be
catered for (Levis, 2018, p. 230). Of course, intelligibility, and therefore what is
included in instruction, is inextricably linked with learner context and setting. Levis
(2005, p. 374) argues that inner circle countries may impose a higher threshold of
what is considered intelligible, providing the example of teaching assistants in the
United States who are regularly tested in terms of spoken proficiency. This higher
threshold could also apply to medical professionals and translators from outside
the inner circle. In addition, there are contexts in which the potential for penalty is
greater. The seriousness of errors can also be differentiated in other ways according
to Levis (2018): Errors which give offence by the mistaken production of taboo
words (e.g., shit/sheet) should be seen as errors worthy of greater attention, as should
words which carry greater lexical importance as those words are instrumental in
affecting understanding. Intelligibility can also be affected by processing constraints,
determined by whether the message is familiar or unfamiliar. Unfamiliar messages
will be more difficult to process (Levis, 2018). These processing difficulties may have
a marked effect on comprehensibility, causing strain for the listener: If this strain
becomes too great, it could render the message unintelligible. In sum, intelligibility
can be determined by who is interacting, the situation in which they are interacting,
the required register, and the general communicative context. Instruction should
account for these factors and also be guided by the learnability of the features taught
(Levis, 2005, 2018).
Levis (2018) also points out the importance of instruction which views pronuncia-
tion in context, as part of general spoken language. As such, it should be multi-modal
in nature, using body language and those elements of orthography and morphology
that are learnable to assist instruction. The opaque relationship between English
orthography and phonology makes this challenging, but not always impossible (Dick-
erson, 1990). Crucially, the features focused on should only be those that impact
intelligibility or comprehensibility, and should be ones necessary for a communica-
tive goal to be reached (e.g., allomorphic-ed past tense endings). In terms of receptive
skills, it is particularly important for learners to be exposed to several different vari-
eties of English, and especially to those that they are likely to encounter (Cauldwell,
2013; Levis, 2018, p. 241).
Although they are guidelines, and therefore may vary according to context, Levis
(2018) makes a number of suggestions in terms of the prioritization of certain
suprasegmentals more likely to impact intelligibility. Drawing on the work of Field
(2005), Hahn (2004), and Zielinski (2008), Levis concludes that word stress is an
68 3 Suprasegmental Instruction and Intelligibility
important area of instruction in all learner contexts and that an incorrect stress place-
ment shift to the right is one that has a greater impact on intelligibility, so should
receive greater attention. In terms of rhythm, perception is the main focus, with expo-
sure to naturalistic, as well as controlled samples, focused on the identification of
key words, weak forms, the identification of the relationship between word class and
stress, and features of connected speech (e.g., linking) viewed as important areas.
Production is generally confined to the vowel length of stressed and unstressed sylla-
bles. Certain elements of intonation should also be prioritized. In particular, Levis
(2018) emphasizes the importance of the most common nuclear stress pattern: the
final content word in thought groups, as well as contrastive stress and stress denoting
new information. Other priority areas include the perception training of non-final
contours and attention to the falling intonation of pauses that extends when multiple
phrases are used (see also Pickering, 2001). Levis also recommends instruction to
encourage the production of lively speech, utilizing chunking and pitch variation to
aid comprehensibility (see also Hincks, 2003; Hincks & Edlund, 2009; Lewis, 1993,
2002) and also key pitch changes in discourse signposting used to organize longer
stretches of learner output (e.g., The FIRST point I’d like to make is…/The SECOND
point…). Some of the key areas omitted are the intonation of polar questions and
intonation indicating attitude as it is difficult to provide general intonation rules
because, contrary to the guidance provided in many coursebooks, intonation can be
quite variable in these areas (Levis, 2018, pp. 262–263). Question tag intonation is
also viewed as being unnecessary, largely because learners hardly ever use these,
and Levis suggests that alternatives can be employed. In terms of general speaking
skills, Levis also advocates rehearsing sections of presentations with particular atten-
tion to thought groups, reading aloud activities, and spontaneous speaking activities
to develop fluency. Speech rate has been identified as a factor affecting compre-
hensibility, so fluency-based activities could be particularly effective (Kang, 2010).
However, perhaps the most significant point made is that the materials used should
be authentic real world examples of language use, utilizing a variety of speakers
and several varieties of native and non-native English (Levis, 2018, p. 263; see also
Ockey & Wagner, 2018).
of Harding (2017), and among others, the criticisms of Derwing and Munro (2015)
and Trofimovich and Isaacs (2012) with regard to nativeness. There is an acceptance
that reference to native accents should not be included and that a re-evaluation of
the native speaker model is needed. Indeed, the new scales contain a descriptor of
intelligibility at every level (Piccardo, 2016, p. 16). The new scale (Council of Europe,
2018, p. 136) divides the assessment of pronunciation into three categories: overall
phonological control, sound articulation, and prosodic features. If we briefly examine
Table 3.1, which is taken from the 2018 descriptors (Council of Europe, 2018, p. 136),
it can be seen that all three categories are guided by the principle of intelligibility.
This means that the sound articulation category (dealing with segmentals) enables
raters to discriminate between low functional load errors that are unlikely to interfere
with intelligibility and high functional load errors that could have more of an impact.
One final point to be made is that when we consider testing pronunciation, there
are major difficulties in terms of how the words used in the descriptors may be
interpreted by different markers. Teacher interview data have shown that teacher
perceptions of what nativeness constitutes can vary considerably, making the provi-
sion of clearly defined pronunciation testing constructs even more important (Bohn
& Hansen, 2017). Thornbury (2017) points out that descriptors used in many tests and
exams may lead to a variability in marking, also emphasizing a point raised by many
others too: that marking is highly subjective and is a product of human interaction.
This can mean that intelligibility might simply be interpreted to mean that the speaker
does not have an intrusive accent or is fairly fluent (see also Pennington, 1998). The
use of terms of quantity, such as limited or some, can be particularly problematic in
descriptors (Piccardo, 2016), and significant amendments to the CEFR descriptors
were made in 2018 to reflect this (Council of Europe, 2018).
3.9 Studies Involving Chinese Learners 71
In terms of Hong Kong English, which has been the subject of a number of studies
(Setter, 2006, 2008; Setter, Wong, & Chan, 2010; Sewell, 2015), there are a number
of salient points that may be pertinent when considering possible pronunciation
difficulties. Firstly, there may be variability in intelligibility, not only between Hong
Kong English and other Englishes, but also within Hong Kong English itself (Setter
et al., 2010). Caution is therefore needed when drawing general conclusions about
problematic areas. Secondly, intelligibility studies have tended to concentrate on
the problematic areas in terms of intelligibility and therefore have tended to neglect
the analysis of phonological features that are easily transferred (Sewell, 2015). This
could mean that important phonological aspects might not be exploited fully. Sewell
(2015) does argue that the substitution and deletion of phonemes is vitally important
in affecting intelligibility, and that the mispronunciation of consonants is of partic-
ular significance although some have stressed the importance of vowels in stressed
syllables and their importance in affecting intelligibility (Zielinski, 2008).
Setter (2006, 2008) has identified a number of differences between Hong Kong
English and UK English pronunciation that could be instrumental in affecting intelli-
gibility. Firstly, the differences in syllable duration of stressed, unstressed, and weak
syllables in Hong Kong English are not so great. Another factor that could also affect
intelligibility is the frequency of consonant clusters in Hong Kong English: It seems
that there are far fewer consonant clusters in Hong Kong English when compared
to British English (Setter, 2006, 2008). This could mean that in the EFL or EAP
classroom learners from Hong Kong may have difficulty in identifying clusters in
listening activities/classroom discussions and also may prove to be less intelligible
for native speakers if they do not produce consonant clusters that would normally
occur in British English (if these clusters are ones that affect intelligibility).
In terms of analyses of Chinese students in different educational settings, Ma’s
(1994) contrastive, computer aided examination of 16 Mandarin speaking undergrad-
uate students’ vowel production showed that students could produce near-native like
frequencies for many phonemes and one diphthong. Of course, this research was set
in an ESL context, so the importance of exposure to language outside the classroom
could be critical. Nonetheless, there does seem to be some evidence that near-native
competence could be possible in some aspects of pronunciation after a prolonged
period of language exposure in a target language country (in this case, the USA).
However, the successful production of diphthongs may still be a problematic area
for Chinese L1 learners (Beebe, 1984).
Although exposure to language outside the classroom is undoubtedly of impor-
tance, other longitudinal studies over relatively short courses of instruction have
also emphasized the importance of focused instruction alongside this exposure.
Derwing et al. (2006) and Derwing et al., (2008) conducted a longitudinal study
of Mandarin, Ukrainian, and Russian speaking beginner students, examining their
language production over the duration of a ten-month ESL programme in Canada.
Student pronunciation was rated at the beginning of the course, after two months of
72 3 Suprasegmental Instruction and Intelligibility
the course, and at the end of the ten-month course. One of the more important findings
from their research was that beginners showed very little improvement indeed without
the impetus of explicit pronunciation instruction. Similarly, the one year follow-up
study (one year after the completion of the ten month course, at 22 months) showed
that although the Slavic speakers had made some progress without any specific
pronunciation instruction, Mandarin speakers made no similar progress. The study
involved 32 beginner level students who were rated in terms of comprehensibility by
native speakers, and it seems to show that beginner students were disadvantaged due
to a lack of pronunciation instruction. For my own research, the study is of impor-
tance because it appears to show that Mandarin speakers struggled more than other
learner groups when deprived of targeted pronunciation instruction.
Contrastive analysis has been used to explain a number of common phonological
errors of Chinese speakers of English. For example, because tone is an important
element of Mandarin, this may lead to the length of a syllable being extended so much
that the syllable becomes a tonic (Juffs, 1990). The pronunciation of the/θ/ sound by
Chinese speakers was the major focus of research which sought to identify which
substituted sound was viewed as the norm by Chinese speakers. Rau, Chang and
Tarone (2009), in their study of 27 Chinese English graduates and undergraduates
living in the USA, identified/s/ as the sound that was viewed as the norm substi-
tution and, extending the work of Labov (1966) on speech communities, point out
that the important factor is not necessarily performance, i.e., whether the/s/ sound is
actually used, but whether it is viewed as the norm. Rau et al. (2009) also examined
the factors which could be used to explain the erroneous pronunciation of/θ/. They
found that frequency of occurrence could be a factor that could aid students; for
example, if/θ/ occurred in the same word repeatedly, learner strategy was identified
as a major factor affecting correct pronunciation. Another important finding was that
those students who used monitoring strategies to improve their pronunciation were
more successful in the pronunciation of/θ/ than those who merely relied on phonetic
salience. So it appears that if students are self-directed and are encouraged to actively
monitor their own production, they could be more successful, at least in terms of the
pronunciation of certain segmentals. However, as was mentioned earlier, whether this
particular substitution actually affects intelligibility or comprehensibility is debat-
able. Deterding (2010, p. 7) argues that these substitutions are perfectly intelligible
in lingua franca contexts, and also points out that a/t/ substitution is recommended
in the Aviation Radiotelephony Manual: an international guide for communication
for those involved in aviation.
In terms of word stress, it is clear that the fact that Mandarin is a tone language
may cause difficulties for Chinese speakers in terms of using word stress correctly
in English (Archibald, 1997). The adult undergraduate Chinese speakers of English
who had been living in the USA for two years in Yu and Andruski’s study (2010)
could identify lexical stress fairly well, but showed a marked preference for iambic
stress patterns due to L1 interference. Furthermore, whereas the English subjects
in the study relied upon segmental information and contextual cues to assist in the
correct identification and placement of stress, the Chinese subjects relied upon pitch
cues. This reliance on pitch cues was due to L1 interference and obviously led to
3.9 Studies Involving Chinese Learners 73
the fourth syllable carries the prominent stress. This could be a valuable and rela-
tively simple way to facilitate instruction in terms of word stress. Other research has
also concentrated on how colour coding can be used to clarify both vocabulary and
particular pronunciation features (Nikolaidis & Mattheoudakis, 2012).
In terms of enhancing monitoring strategies to improve students’ pronunciation
skills, The University of Sheffield (2017) suggests a number of activities that could
be beneficial. Of course, as Dlaska and Krekeler (2008) point out in their investi-
gation of student self-assessment of their own pronunciation, students may require
scaffolding and support from their instructor and the activities themselves so that they
are aware of L1 interference and made aware of possible problematic areas. In terms
of the problematic area of diphthongs, Foster (in The University of Sheffield, 2017)
suggests using visual cues, for example, by using different parts of the teacher’s face
to use words to represent particular diphthongs. It is also argued that the identifi-
cation of nuclear stress is a particularly problematic area for Chinese learners, and
that listening exercises that focus on the identification of nuclear stress can be of
particular importance for learners being able to fully comprehend lectures. Further-
more, learners can be encouraged to record themselves and identify the nuclear
stress in their own presentations (Foster, in The University of Sheffield, 2017).
Learners’ own presentations can be exploited in more productive ways too. The
video recording of learner presentations has been shown to be an important tool
for improving oracy skills, particularly when a variety of feedback is provided that
can also lead to more intelligible performances (Heron, 2018, 2019; Ritchie, 2016).
In addition, the identification of discourse markers in planned presentations can be
transferred to unplanned speech and assist the comprehensibility of learner output
(see Williams’ [1992] teacher assistant research). Instruction aimed at varying the
pitch in learner academic presentations has also been shown to improve compre-
hensibility (Wennerstrom, 1994). Three further examples of how student monitoring
strategies can be promoted to assist instruction are provided by Jones, Argent and
Sesserman (in The University of Sheffield, 2017). Jones gives the example of using a
reading text in preparation for a seminar, where students are asked not only to high-
light arguments for and against a particular viewpoint in the text, but also to identify
and check words that could be difficult to pronounce. Learners can also be instructed
to note the words/expressions that they did not understand during the seminar task,
and significantly, these problematic words are also noted by the teacher. They can
then be focused on (particularly by using word recognition apps for self-study).
Argent (in The University of Sheffield, 2017) concentrates on the value of
enhancing student monitoring strategies in terms of intonation choices. In partic-
ular, Argent gives the examples of tone choices for confirming, and also concession
contra-expectation, where the author concedes that the expected results did not mate-
rialize. There is also a description of what Woolard (2013) termed messaging, that
is, expanding students knowledge of intonation choices by giving examples of the
same pattern, but with the noun form changed. Argent gives the examples of that’s
the wrong data/that’s the wrong e-mail/that’s the wrong article: all with the same
falling intonation pattern.
76 3 Suprasegmental Instruction and Intelligibility
It has also been suggested that awareness of elision could be improved if students
are exposed to two recordings: one with elements of connected speech included,
and the other with the elided elements omitted (Sesserman, in The University of
Sheffield, 2017). Students could then be asked to identify the elided elements from the
listening. This could be beneficial in improving students’ receptive skills, and could
be particularly useful for improving understanding of lectures and native speaker
students with whom they will be expected to interact with on their future university
courses.
Finally, another obstacle to providing meaningful instruction could be a lack of
guided practice activities. There is some evidence that teachers may not be employing
sufficient guided practice activities to provide scaffolding for learners during pronun-
ciation instruction (Baker, 2014, pp. 153–154). The provision of such guided practice,
rather than controlled practice activities, could be instrumental in terms of assisting
learners and helping them to bridge the gap between their L1 and the target language.
In conclusion, the view that suprasegmental pronunciation instruction should play
a vital part in instruction is now well established. Some have suggested that courses
should initially focus on speaking skills and then focus attention on the pronunciation
areas that can be improved to facilitate intelligibility (Firth, 1992), whereas others
have advocated the employment of global speaking skills alongside suprasegmental
instruction (Gordon & Darcy, 2016). What seems clear is that pronunciation instruc-
tion choices should be based on evidence (Derwing & Munro, 2015; Levis, 2018), and
the evidence points to suprasegmental instruction as an effective tool in promoting
learner intelligibility (Derwing & Munro, 2015; Derwing et al., 1997, 1998; Field,
2005; Hahn, 2004; Levis, 2018; Pickering, 2001, 2009; Wennerstrom, 1994, 1998).
On short EAP university pre-sessional courses, a goal of intelligibility rather than
native-like production is a much more realistic one, but the goal of intelligibility
necessitates prioritization (Levis, 2018). If segmental instruction is undertaken, it
should be done so with due consideration to the concept of functional load in order to
ensure that this goal is adhered to. Likewise, suprasegmental instruction should prior-
itize particular features that are attainable over a short course that have been shown
to aid intelligibility and comprehensibility: word stress (with particular attention
paid to erroneous placements to the right of disyllabic words, e.g., folLOW ), nuclear
stress, contrastive stress, and pitch variation, combined with chunking activities to
enhance comprehensibility (Levis, 2018). Listening exercises are crucial and should
include a range of both native and non-native varieties, with particular attention paid
to those varieties that the learner will be likely to encounter. Furthermore, there are
ways in which listening activities outside the classroom can be utilized to promote
independent learning, for example by utilizing instructional videos (Sonsaat, 2017,
p. 183) or shadowing techniques (Foote & McDonough, 2017). Also, with respect to
how instruction takes place, the scaffolding of activities can be aided by guided prac-
tice activities, accompanied by clear, form-focused feedback and supportive teacher
body language (Baker, 2011, 2014; Baker & Burri, 2016; Couper, 2017). Most impor-
tantly, the communicative value of the pronunciation exercises employed should be
made transparent to learners. Finally, in view of the evidence suggesting a lack of
teacher confidence in providing suprasegmental instruction based on intelligibility
3.10 Instruction in EAP Contexts 77
(Couper, 2017; Foote et al., 2011; Macdonald, 2002; Sonsaat, 2017), the goal of
instruction should be made clear by course leaders, and priorities should be apparent
in the syllabus content. Guidance in how such instruction might be implemented
would also be beneficial, particularly in light of the poor guidance often provided in
coursebook materials and teacher coursebook guides (Derwing et al., 2012; Sonsaat,
2017).
References
Catford, J. C. (1988). Functional load and diachronic phonology. In Y. Tobin (Ed.), The Prague
School and its legacy: In linguistics, literature, semiotics, folklore and the arts. (3–20).
Amsterdam/Philadelphia: John Benjamins. https://doi.org/10.1075/llsee.27.04cat
Cauldwell, R. (2013). Phonology for listening. Birmingham: Speech in Action.
Celce-Murcia, M., Brinton, D., & Goodwin, J. (2010). Teaching pronunciation: A course book and
reference guide (2nd ed.). New York & Cambridge: Cambridge University Press.
Chomsky, N. (1959). Review of B. F. Skinner’s, Verbal behavior. Language 35(1), 26–58. Retrieved
from https://cogprints.org/1148/1/chomsky.htm
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: M.I.T. Press.
Chrabaszcz, A., Winn, M., Linn, C. Y., & Idsardi, J. (2014). Acoustic cues to perception of word
stress by English, Mandarin, and Russian Speakers. Journal of Speech, Language, and Hearing
Research, 57(4), 1468–1479. https://doi.org/10.1044/2014_JSLHR-L-13-0279
Chun, D. M. (1988). The neglected role of intonation in communicative competence and proficiency.
The Modern Language Journal, 72(3), 259–303. https://doi.org/10.1111/j.1540-4781.1988.tb0
4192.x
Chun, D. M. (2002). Discourse intonation in L2: From theory to practice. Amsterdam and
Philadelphia: John Benjamins Publishing Company.
Cook, V. J. (2008). Second language learning and language teaching (4th ed.). London: Hodder
Education.
Council of Europe (2018). Common European Framework of Reference for Languages: Learning,
teaching, assessment: companion volume with new descriptors. Retrieved February 13, 2019,
from https://rm.coe.int/cefr-companion-volume-with-new-descriptors-2018/1680787989
Couper, G. (2017). Teacher cognition of pronunciation teaching: Teachers’ concerns and issues.
TESOL Quarterly, 51(4), 820–843. https://doi.org/10.1002/tesq.354
Cruttenden, A. (1997). Intonation (2nd ed.). Cambridge, UK: Cambridge University Press.
Cutler, A. (1986). Forbear is a homophone: Lexical prosody does not constrain lexical access.
Language and Speech, 29(3), 201–220. https://doi.org/10.1177/002383098602900302
Cutler, A. (2012). Native listening. Cambridge, MA: MIT Press.
Cutler, A, Dahan, D., & van Donselaar, W. (1997). Prosody in the comprehension of spoken
language: A literature review. Language and Speech, 40(2), 141–201. https://doi.org/10.1177/
002383099704000203
Daniel, I. O. (2011). Introductory phonetics and phonology of English. Newcastle Upon Tyne:
Cambridge Scholars Publishing.
Dauer, R. M. (1983). Stress-timing and syllable-timing reanalyzed. Journal of Phonetics, 11(1),
51–62. https://psycnet.apa.org/record/1983-29886-001
Dauer, R. M. (1993). Accurate English. Englewood Cliffs, NJ: Regents/Prentice Hall.
Dauer, R. M. (2005). The lingua franca core: A new model for pronunciation instruction? TESOL
Quaterly, 39(3), 543–550. https://doi.org/10.2307/3588494
Derwing, T. M. (2003). What do ESL students say about their accents? Canadian Modern Language
Review, 59(4), 547–566. https://doi.org/10.3138/cmlr.59.4.547
Derwing, T. M., Diepenbrook, L. G., & Foote, J. A. (2012). How well do general skills ESL textbooks
address pronunciation? TESL Canada Journal, 30(1), 22–44. https://doi.org/10.18806/tesl.v30i1.
1124
Derwing, T. M., & Munro, M. J. (2005). Second language accent and pronunciation teaching: A
research-based approach. TESOL Quarterly, 39(3), 379–398. https://doi.org/10.2307/3588486
Derwing, T. M., & Munro, M. J. (2011). The foundations of accent and intelligibility in pronunciation
research. Language Teaching, 44(3), 316–327. https://doi.org/10.1017/s0261444811000103
Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidence-based perspectives
for L2 teaching and research. Amsterdam: John Benjamins.
Derwing, T. M., Munro, M. J., & Thomson, R. I. (2008). A longitudinal study of ESL learners’
fluency and comprehensibility development. Applied Linguistics, 29(3), 359–380. https://doi.org/
10.1093/applin/amm041
References 79
Derwing, T. M., Munro, M. J., & Wiebe, G. (1997). Pronunciation instruction for ‘fossilized
learners’: Can it help? Applied Language Learning, 8(2), 217–235. https://eric.ed.gov/?id=EJ5
67518
Derwing, T. M., Munro, M. J., & Wiebe, G. (1998). Evidence in favour of a broad framework for
pronunciation instruction. Language Learning, 48(3), 393–410. https://doi.org/10.1111/0023-
8333.00047
Derwing, T. M., & Rossiter, M. J. (2002). ESL learners’ perceptions of their pronunciation needs
and strategies. Science Direct, 30(2), 155–166. https://doi.org/10.1016/S0346-251X(02)00012-X
Derwing, T. M., & Rossiter, M. J. (2003). The effects of pronunciation instruction on the accu-
racy, fluency and complexity of L2 accented speech. Applied Language Learning, 13(1), 1–
17. https://www.researchgate.net/publication/234570703_The_Effects_of_Pronunciation_Instru
ction_on_the_Accuracy_Fluency_and_Complexity_of_L2_Accented_Speech
Derwing, T. M., Rossiter, M. J., & Munro, M. J. (2002). Teaching native speakers to listen to foreign-
accented speech. Journal of Multilingual and Multicultural Development, 23(4), 245–259. https://
doi.org/10.1080/01434630208666468
Derwing, T. M., Thomson, R. I., & Munro, M. J. (2006). English pronunciation and fluency in
Mandarin and Slavic speakers. System, 34(2), 183–193. https://doi.org/10.1016/j.system.2006.
01.005
Deterding, D. (2005). Listening to Estuary English in Singapore. TESOL Quarterly, 39(3), 425–440.
https://doi.org/10.2307/3588488
Deterding, D. (2010). ELF-based pronunciation teaching in China. Chinese Journal of Applied
Linguistics, 33(6), 3–15. https://www.researchgate.net/publication/263010941_ELF-based_pro
nunciation_teaching_in_China/link/02e7e5399310b479ba000000/download
Deterding, D., & Kirkpatrick, A. (2006). Emerging South-East Asian Englishes and intelligibility.
World Englishes, 25(3), 391–409. https://doi.org/10.1111/j.1467-971X.2006.00478.x
Dickerson, W. B. (1990). Morphology via orthography: A visual approach to oral decisions. Applied
Linguistics, 11(3), 238–252. https://doi.org/10.1093/applin/11.3.238
Dlaska, A., & Krekeler, C. (2008). Self-assessment of pronunciation. System, 36(4), 506–516.
https://doi.org/10.1016/j.system.2008.03.003
Eisenchlas, S. A., & Tsurutani, C. (2011). You sound attractive! Perceptions of accented English
in a multilingual environment. Australian Review of Applied Linguistics, 34(2), 216–236. https://
doi.org/10.1075/aral.34.2
Field, J. (2005). Intelligibility and the listener: The role of lexical stress. TESOL Quarterly, 39(3),
399–423. https://doi.org/10.2307/3588487
Firth, S. (1992). Developing self-correction and self-monitoring strategies. In P. Avery & S. Ehrlich
(Eds.), Teaching American English pronunciation (pp. 215–219). Oxford: Oxford University
Press.
Foote, J. A., Holtby, A. K., & Derwing, T. M. (2011). Survey of the teaching of pronunciation in
adult ESL programs in Canada, 2010. TESL Canada Journal, 29(1), 1–22. https://doi.org/10.
18806/tesl.v29i1.1086
Foote, J. A., & McDonough, K. (2017). Using shadowing with mobile technology to improve L2
pronunciation. Journal of Second Language Pronunciation, 3(1), 34–56. https://doi.org/10.1075/
jslp.3.1.02foo
Fries, C. C. (1964). On the intonation of ‘yes-no’ questions in English. In D. Abercrombie, D. B.
Fry, P. A. D. MacCarthy, N. C. Scott, & J. L. Trim (Eds.), In honour of Daniel Jones: Papers
contributed on the occasion of his eightieth birthday (pp. 242–244). London: Longman.
Gass, S., & Varonis, E. M. (1984). The effect of familiarity on the comprehensibility of nonnative
speech. Language Learning, 34(1), 65–87. https://doi.org/10.1111/j.1467-1770.1984.tb00996.x
Gordon, J., & Darcy, I. (2016). The development of comprehensible speech in L2 learners: A class-
room study on the effects of short-term pronunciation instruction. Journal of Second Language
Pronunciation, 2(1), 56–92. https://doi.org/10.1075/jslp.2.1.03gor
Gorham, J. (1988). The relationship between verbal teacher immediacy behaviours and student
learning. Community Education, 37(1), 40–53. https://doi.org/10.1080/03634528809378702
80 3 Suprasegmental Instruction and Intelligibility
Gramley, S., & Patzold, K. M. (2004). A survey of modern English (2nd ed.). London: Routledge.
Gregersen, T. (2011). Suprasegmentals: Tools for increased language teacher effectiveness. In J.
Arabski & A. Wojtaszek (Eds.), The acquisition of L2 phonology (pp. 27–36). Bristol: Multilingual
Matters.
Hahn, L. D. (2004). Primary stress and intelligibility: Research to motivate the teaching of
suprasegmentals. TESOL Quarterly, 38(2), 201. https://doi.org/10.2307/3588378
Halliday, M. A. K. (1967). Intonation and grammar in British English. The Hague: Mouton.
Halliday, M. A. K. (1985). An introduction to functional grammar. London: Edward Arnold.
Halliday, M. A. K., & Greaves, W. S. (2008). Intonation in the grammar of English. London:
Equinox Publishing Ltd.
Harding, L. (2017). What do raters need in a pronunciation scale? The user’s view. In T. Isaacs & P.
Trofimovich (Eds.), Second language pronunciation assessment (pp. 12–34). Bristol: Multilingual
Matters.
Harmer, J. (2001). The practice of English language teaching (3rd ed.). Harlow, England: Pearson
Education.
Hattie, J. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to achievement.
Florence, US: Routledge.
Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1),
81–112. https://doi.org/10.3102/003465430298487
Heron, M. (2018). Communicating in a business-like way: developing oracy skills in higher educa-
tion. [Power Point slides]. Retrieved January 29, 2019, from https://www.baleap.org/event/spe
aking
Heron, M. (2019). Making the case for oracy skills in higher education: Practices and opportunities.
Journal of University Teaching & Learning Practice, 16(2). Retrieved from https://ro.uow.edu.
au/jutlp/vol16/iss2/9
Hincks, R. (2003). Speech technologies for pronunciation feedback and evaluation. ReCALL
15(1):3–20. https://doi.org/10.1017/S0958344003000211
Hincks, R., & Edlund, J. (2009). Promoting increased pitch variation in oral presentations with tran-
sient visual feedback. Language, Learning and Technology, 13(3), 3–20. https://doi.org/10125/
44190
Isaacs, T., & Harding, L. (2017). Pronunciation assessment. Language Teaching, 50(3), 347–366.
https://doi.org/10.1017/S0261444817000118
Isaacs, T., & Trofimovich, P. (2012). Deconstructing” comprehensibility: Identifying the linguistic
influences on listeners’ L2 comprehensibility ratings. Studies in Second Language Acquisition,
34(4), 475–505. https://doi.org/10.1017/S0272263112000150
Isaacs, T., & Trofimovich, P. (2017). Assessment of second language pronunciation. In T. Isaacs
and P. Trofimovich (Eds.), Second language pronunciation assessment (pp. 3–11). Bristol:
Multilingual Matters. https://doi.org/10.21832/ISAACS6848
Jarosz, A. (2019). English pronunciation in L2 instruction: The case of secondary school learners.
Cham, Switzerland: Springer.
Jenkins, J. (1998). Which pronunciation norms and models for English as an international language?
ELT Journal, 52(2), 119–126. https://doi.org/10.1093/elt/52.2.119
Jenkins, J. (2000). The phonology of English as an international language. Oxford: Oxford
University Press.
Jenkins, J. (2002). A sociolinguistically based, empirically researched pronunciation syllabus for
English as an international language. Applied Linguistics, 23(1), 83–103. https://doi.org/10.1093/
applin/23.1.83
Jenkins, J. (2007). English as a lingua franca: Attitude and identity. Oxford: Oxford University
Press.
Johnstone, R. (2002). Addressing ‘the age factor’: Some implications for languages policy. Council
of Europe. Retrieved from: https://rm.coe.int/addressing-the-age-factor-some-implications-for-
languages-policy-/1680886e92
References 81
Juffs, A. (1990). Tone, syllable structure and interlanguage phonology: Chinese learners’ stress
errors. International Review of Applied Linguistics, 28(2), 99–117. https://doi.org/10.1515/iral.
1990.28.2.99
Kachru, B. B. (1992). Teaching world Englishes. In B. B. Kachru (Ed.), The other tongue: English
across cultures (2nd ed., pp. 355–365). Urbana, IL: University of Illinois Press.
Kang, O. (2010). Relative salience of suprasegmental features on judgments of L2 comprehensibility
and accentedness. System, 38(2), 301–315. https://doi.org/10.1016/j.system.2010.01.005
Kennedy, S., & Trofimovich, P. (2008). Intelligibility, comprehensibility, and accentedness of L2
Speech: The role of listener experience and semantic context. The Canadian Modern Language
Review, 64(3), 450–489. https://doi.org/10.3138/cmlr.64.3.459
Kusiak, M. (2011). The role of phonological awareness in beginning reading. In J. Arabski & A.
Wojtaszek (Eds.), The acquisition of L2 phonology (pp. 77–92). Bristol: Multilingual Matters.
Labov, W. (1966). The social stratification of English in New York City. Washington, DC: Centre
for Applied Linguistics.
Ladd, D. R. (1980). The structure of intonational meaning: Evidence from English. Bloomington,
IN: Indiana University Press.
Lee, W. R. (1980). A point about the rise-endings and fall-endings of yes-no questions. In L.
R. Waugh & C. H. van Schooneveld (Eds.), The melody of language intonation and prosody
(pp. 165–168). Baltimore: University Park Press.
Levis, J. M. (1999a). The intonation and meaning of normal yes/no questions. World Englishes,
18(3), 373–380. https://doi.org/10.1111/1467-971X.00150
Levis, J. M. (1999b). Intonation in theory and practice, revisited. TESOL Quarterly, 33(1), 37–63.
https://doi.org/10.2307/3588190
Levis, J. M. (2005). Changing contexts and shifting paradigms in pronunciation teaching. TESOL
Quarterly, 39(3), 369–378. https://doi.org/10.2307/3588485
Levis, J. M. (2018). Intelligibility, oral communication, and the teaching of pronunciation.
Cambridge, UK & New York, NY: Cambridge University Press.
Levis, J. M., & Cortes, V. (2008). Minimal pairs in spoken corpora: Implications for pronunciation
assessment and teaching. In C. A. Chapelle, Y. R. Chung, & J. Xu (Eds.), Towards adaptive
CALL: Natural language processing for diagnostics language assessment (pp. 197–208). Ames,
IA.: Iowa State University.
Lewis, M. (1993). The lexical approach. Hove: Language teaching publications.
Lewis, M. (2002). Implementing the lexical approach: Putting theory into practice (2nd ed.). Boston:
Heinle.
Lindemann, S. (2003). Koreans, Chinese, or Indians? Attitudes and ideologies about nonnative
English speakers in the United States. Journal of Sociolinguistics, 7(3), 348–364. https://doi.org/
10.1111/1467-9481.00228
Lindemann, S. (2017). Variation or ‘Error’? Perception of pronunciation variation and implications
for assessment. In T. Isaacs & P. Trofimovich (Eds.), Second Language Pronunciation Assessment
(pp. 193–209). Bristol: Multilingual Matters.
Littlewood, W. (1984). Foreign and second language learning: Language acquisition and ts
implications for the classroom. Cambridge: Cambridge University Press.
Loewen, S., & Philp, U. (2006). Recasts in the adult English L2 classroom: Characteristics, explicit-
ness, and effectiveness. The Modern Language Journal, 90(4), 536–556. https://doi.org/10.1111/
j.1540-4781.2006.00465.x
Ma, L. (1994). English Learning: An analysis of Chinese students’ problems in pronunciation.
Retrieved February 13, 2019, from https://eric.ed.gov/?id=ED411668
Macdonald, S. (2002). Pronunciation-views and practices of reluctant teachers. Prospect, 17(3),
3–16. https://hdl.handle.net/1959.14/323083
Marks, J. (2002). Pronunciation in the lexical approach. In M. Lewis (Ed.), Implementing the lexical
approach (pp. 156–162). Boston: Heinle.
McCarthy, M. (1991). Discourse analysis for language teachers. Cambridge: Cambridge University
Press.
82 3 Suprasegmental Instruction and Intelligibility
Ricard, E. (1986). Beyond fossilization: A course on strategies and techniques in pronunciation for
adult learners. TESL Canada Journal, Special Issue 1, 243–249. https://doi.org/10.18806/tesl.
v3i0.1009
Richards, M. G. (2016). Not all word stress errors are created equal: Validating an English word
stress error gravity hierarchy (Doctoral dissertation, Iowa State University, Ames, US). Retrieved
from https://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=7008&context=etd
Richmond, V. P., & McCroskey, J. C. (2004). Nonverbal behaviour in interpersonal communications
(5th ed.). Boston, MA: Allyn & Bacon.
Ritchie, S. M. (2016). Self-assessment of video-recorded presentations: Does it improve skills?
Active Learning in Higher Education, 17(3), 207–221. https://doi.org/10.1177/146978741665
4807
Roach, P. (1983). English phonetics and phonology. Cambridge: Cambridge University Press.
Rossiter, M. (2003). ‘It’s like chicken but bigger’: Effects of communication strategy in the ESL
classroom. The Canadian Modern Language Review, 60(2), 105–121. https://doi.org/10.3138/
cmlr.60.2.105
Saito, K., & Lyster, R. (2012). Effects of form-focused instruction and corrective feedback on L2
pronunciation development of/ô/ by Japanese learners of English. Language Learning, 62(2),
595–633. https://doi.org/10.1111/j.1467-9922.2011.00639.x
Saito, K., Trofimovich, P., Isaacs, T., & Webb, S. (2017). Re-examining phonological and lexical
correlates of second language comprehensibility: The role of rater experience. In T. Isaacs
& P. Trofimovich (Eds.), Second Language Pronunciation Assessment (pp. 141–156). Bristol:
Multilingual Matters.
Scovel, T. (2000). A critical review of the critical period research. Annual Review of Applied
Linguistics, 20, 213–223. https://doi.org/10.1017/S0267190500200135
Seidlhofer, B. (2011). Understanding English as a lingua franca. Oxford: Oxford University Press.
Setter, J. (2006). Speech rhythm in world Englishes: The case of Hong Kong. TESOL Quarterly,
40(4), 763–782. https://doi.org/10.2307/40264307
Setter, J. (2008). Consonant clusters in Hong Kong English. World Englishes, 27(3–4), 502–515.
https://doi.org/10.1111/j.1467-971X.2008.00581.x
Setter, J., Wong, C. S. P., & Chan, B. H. S. (2010). Hong Kong English. Edinburgh: Edinburgh
University Press.
Sewell, A. (2015). The intranational intelligibility of Hong Kong English accents. System, 49, 86–97.
https://doi.org/10.1016/j.system.2015.01.003
Shockey, L. (2003). Sound patterns of spoken English. Oxford: Blackwell.
Shockey, L. (2011). Understanding L2 and the perspicacious Pole. In J. Arabski & A. Wojtaszek
(Eds.), The acquisition of L2 phonology (pp. 27–36). Bristol: Multilingual Matters.
Sifakis, N. C., & Sougari, A. M. (2005). Pronunciation issues and EIL pedagogy in the periphery:
A survey of Greek state school teachers’ beliefs. TESOL Quarterly, 39(3), 467–488. https://doi.
org/10.2307/3588490
Singleton, D. (1989). Language acquisition: The age factor. Clevedon: Multilingual Matters.
Skinner, B. F. (1957). Verbal Communication. New York: Appleton-Century-Crofts.
Small, L. H., Simon, S. D., & Goldberg, J. S. (1988). Lexical stress and lexical access: Homographs
versus nonhomographs. Perception & Psychophysics, 44(3), 272–280. https://doi.org/10.3758/
BF03206295
Smith, L. E., & Nelson, C. (1985). International intelligibility of English: Directions and resources.
World Englishes, 4(3), 333–342. https://doi.org/10.1111/j.1467-971X.1985.tb00423.x
Sonsaat, S. (2017). The influence of an online pronunciation teacher’s manual on teachers’ cogni-
tions. (Doctoral dissertation, Iowa State University, US). https://doi.org/10.31274/etd-180810-
5849
Szczepek-Reed, B. (2012). A conversation analytic perspective on teaching English pronunciation:
The case of speech rhythm. International Journal of Applied Linguistics, 22(1), 67–87. https://
doi.org/10.1111/j.1473-4192.2011.00293.x
84 3 Suprasegmental Instruction and Intelligibility
Yates, L., Zielinski, B., & Pryor, E. (2011). The assessment of pronunciation and the new IELTS
pronunciation scale. In J. Osborne (Ed.), IELTS research reports 12 (pp. 1–46). Melbourne and
Manchester: IDP IELTS Australia and British Council.
Yu, V. W., & Andruski, J. E. (2010). Cross-language study of perception of lexical stress in English.
Journal of Psycholinguist Research, 39(4), 323–344. https://doi.org/10.1007/s10936-009-9142-2
Zielinski, B. (2008). The listener: No longer the silent partner in reduced intelligibility. System,
36(1), 69–84. https://doi.org/10.1016/j.system.2007.11.004
Chapter 4
Methodology: The Mixed Methods
Approach
4.1 Introduction
The following chapter includes a description of the research context, the Research
Questions I sought to answer, the research timetable, an explanation of the pilot study
that led to the current research, a rationale for employing a mixed methods approach
and the particular methods used, and an explanation of how particular methods and
tools were employed. This is then followed by a description of the participants
involved in the study, along with ethical considerations and a stakeholder analysis
of the participants.
to note that although the university refers to IELTS scores, assessment on the course
may be quite different from that in IELTS exams. Assessments are set and controlled
internally and are entirely independent of the official IELTS examination system.
Another important point is that although most learners have an average of 5, the
individual IELTS scores that they possess prior to registering on the course may vary.
This means that learners may have an IELTS score of 4.5 in one or two language skill
areas, but still have an overall average of 5. This is particularly important if learners
happen to have the lower score of 4.5 in either listening or speaking because it may
mean that they will have greater difficulties in pronunciation than other students.
There are a number of aspects of the university course that should also be clarified,
including other pre-sessional courses that are also run by the university in the summer
months, the nationalities of learner groups, the nature of the course (general or
specific), the summative assessments, the number of hours of instruction, and the
organization of the classes.
Aside from the ten-week pre-sessional course, there are similar courses of five and
fifteen-week duration operating at the university during the summer months, with the
five-week course typically aiming to improve learners’ IELTS scores from 5.5 to 6,
and the fifteen-week course aiming to improve scores from 4.5 to 6. The vast majority
of learners are Chinese speakers although other learner groups, such as Russian,
Thai, and Emirati are present. All three courses teach English for General Academic
Purposes, as opposed to English for Specific Academic Purposes although students
are encouraged to explore their target subject area by writing a project and giving
a presentation on their chosen project. These two assignments are only assessed
formatively though, and the focus of the course is on general essay writing skills
using sources, speaking in seminars, reading, and listening. Summative assessments
consist of a timed essay using sources (40%); a group speaking exam consisting of
three or four students, where students discuss a topic and a series of questions that
they know of beforehand for between 15 and 20 min (40%); a listening exam (10%);
and a reading exam (10%). In recent years, the listening exam has moved away from
the typical IELTS type of exam to include longer listening extracts from lectures,
presumably in an attempt to improve a key skill learners will need on their future
courses: the ability to listen to and understand lectures. In the year in which the
research took place (2018), there were 15 h per week of face to face instruction, with
reading, writing, and grammar accounting for 10.5 h per week, speaking sessions 3 h
per week, and listening sessions 1.5 h per week. A total of 4 h per week were also
devoted to individual tutorials with learners (approximately 20 min tutorial time per
student), and some weeks involved a 1.5 h session where students had the opportunity
to meet the lecturers from their target courses.
Teachers work with a teaching partner and are assigned a home class and a second
class. The teacher is responsible for teaching reading, writing and grammar to their
4.2 Research Context 89
home class (10.5 h per week), and listening and speaking to their second class (4.5 h
per week). There are therefore some days when teachers swap groups with their
partner teacher and teach their second class. Teachers follow a syllabus over the
ten-week course (referred to as the scheme of work or SOW in teacher interviews)
although they are able to use alternative materials. Students are supported on the
course by the course’s website, which includes support resources and an EAP toolkit
for independent study.
As well as being a researcher during the course, I also taught on the ten-week
course. Being acutely aware of the possible difficulty of this dual insider/outsider
role, I was particularly careful to ensure that the research adhered to the university’s
own ethics policy and its key principles of beneficence (doing positive good), non-
malfeasance (doing no harm), integrity, informed consent, confidentiality/anonymity,
and impartiality.
The research I conducted is based upon the four Research Questions I sought to
answer. Each Research Question is followed by subsidiary questions to clarify the
area of inquiry.
(1) What are the goals in terms of suprasegmental pronunciation instruction and
intelligibility-based instruction?
• what are the institutional goals?
• are they officially stated (documents) and how are they formulated?
• how do the documents and course leader shape the goals?
(3) To what extent are these goals reflected by teacher instruction, attitudes and
beliefs?
• what teacher factors influence the instruction provided (e.g., experience,
education, motivation, beliefs, attitudes, confidence, and guidance received)?
• what are the teachers’ goals, attitudes, and beliefs on the pre-sessional?
90 4 Methodology: The Mixed Methods Approach
Table 4.1 illustrates the research timeline and provides information on which partic-
ular research methods were used to answer particular Research Questions. A variety
of quantitative and qualitative methods were used.
First, course documents were examined in order to ascertain what the formal goals
of the pre-sessional were, how these were assessed, and how clearly these goals were
transmitted to both teachers and learners. These documents included the syllabus
content in terms of pronunciation instruction, the associate lecturer supplement
and student handbook that were distributed during the induction, and the seminar
marksheet used for the seminar exam (see Appendix A), together with the seminar
guidelines (Appendix B) and standardization document that were used during the
seminar exam standardization session. Second, action research, encompassing self-
observation and participant observation (including notes taken on syllabus content,
guidance provided during induction and staff meetings, and my own efforts to provide
meaningful suprasegmental instruction) were employed in order to explore the guid-
ance provided to teachers in terms of the course goals, how they could be imple-
mented, and to investigate any difficulties experienced in attempting to implement
these goals. Third, teacher observation, in the form of one video recorded lesson
and the completion of teacher self-observation checklists, was used in order to view
how suprasegmental instruction and error correction were provided. Fourth, teacher
grading of a recorded speaking practice exam session, together with a brief follow-
up interview with the teachers to provide explanations for the pronunciation grades
awarded, was employed in order to investigate assessment goals and the interpreta-
tions of the descriptors by the teacher raters. Fifth, a semi-structured interview with
the course leader (CL) gave more data in terms of clarifying the goals of the pre-
sessional. Sixth, student questionnaires were then distributed and collected towards
the end of the course to uncover learner attitudes towards pronunciation instruc-
tion (e.g., preferences) and collate their opinions in terms of the effectiveness of
pronunciation instruction on the course. This was then followed by three in-depth,
semi-structured interviews with teacher participants to investigate teacher attitudes,
4.3 Research Process 91
In the summer of 2017, after receiving informed consent from the teacher participants
and CL, I conducted exploratory interviews with four teachers in order to gain greater
insight into teacher attitudes and beliefs, and how well they thought instruction was
incorporated into the course. I also obtained informed consent from learners in my
own class for the video recording of a practice seminar exam that was utilized in the
current research: the teacher assessment of student seminar phase. Biographical and
background data of the teacher participants are outlined in Table 4.2.
All teachers were experienced, qualified teachers of English, who had been
working on the same pre-sessional course for over five years. Three teachers were
native speakers and one was a non-native speaker. Estimations of how well teacher
training prepared them varied, with two teachers stating that it did not assist them at
all in terms of being able to provide pronunciation instruction. The other two stated
that it did help to a certain extent.
Table 4.3 gives a summary of teacher confidence and the stated target of instruc-
tion (intelligibility/comprehensibility versus native speaker-like production). Two
4.3 Research Process 93
Table 4.2 Teacher estimations of how well their teacher training prepared them
Teacher (NS or Approximate years Teacher Estimate of how well
NNS/male/female) experience qualification teacher training
prepared them to teach
pronunciation
1 NS/M Over 30 years CELTA, MA Not at all
2 NS/F 37 years ESOL Not at all
3 NS/M 11 years CELTA, working Fairly basic, but some
towards Diploma useful aspects
TESOL
4 NNS/F Over 30 years BEd, Diploma Teacher training helped
TESOL
Source Author’s own
Table 4.3 Levels of teacher confidence and the stated target of instruction
Teacher (NS or NNS; Levels of confidence Stated target of instruction
male/female)
1 NS/M Low confidence Intelligibility
2 NS/F Generally avoids it due to lack of Intelligibility (but would correct
time an incorrect pronunciation of /θ/)
3 NS/M Confident and enjoys teaching it Intelligibility and
comprehensibility. An erroneous
pronunciation of the /θ/ sound
does not affect intelligibility
4 NNS/F Low confidence Intelligibility
Source Author’s own
teachers reported low confidence in providing instruction and one reported a high
level of confidence. Teacher 2 simply stated that pronunciation instruction was gener-
ally avoided due to a lack of time. All teachers stated that intelligibility was the target
of instruction although teacher 2 stated that she would correct a mispronounced /θ/
and /ð/ sound: something that may seem at odds with intelligibility as a target of
instruction. Teacher 4 intimated that her experience of being raised in a country
where English is commonly spoken as a second language informed her belief that
different accents (rather than a native speaker target) and an alternative pronunciation
of the /θ/ and /ð/ sounds (e.g., with a substituted /t/ and /d/) are perfectly acceptable.
There was a general consensus that error correction should be provided individ-
ually, often in tutorial so that student confidence is maintained (teachers 2 and 4
stress this in particular). Teacher 2 also pointed to the lack of time as a reason for her
limiting error correction to words which were subject specific, presumably because
the frequency of use would be greater in the learners’ future studies and those partic-
ular errors could therefore affect learner progress on their future courses. Teacher
1 stated that he would provide instant correction if a word was mispronounced.
94 4 Methodology: The Mixed Methods Approach
However, later in the interview he stated that he probably would not do this and also
pointed out that phonemic errors are often intelligible due to contextual clues. All
teachers reported that the suprasegmental features of stress and intonation were of
greater importance and something they focus on. Teacher 4 also stated that students
enjoy intonation and chunking activities (see Table 4.4).
There were a number of suggestions on how pronunciation instruction might be
improved on the course (Table 4.5). The assertions of Smith (in The University of
Sheffield, 2017), that there may be a lack of suitable materials for instruction in
EAP contexts, seem to be supported by the pilot study findings: There was a unan-
imous dissatisfaction with the pronunciation materials used, with teachers 1, 2, and
3 questioning the quality of the materials, and teacher 4 emphasizing the lack of
materials. A lack of structure to the pronunciation instruction provided was also a
common theme, as was criticism of the listening activities, which teachers found
to be either neglected or not integrated. Also, as in Macdonald’s teacher interviews
Table 4.4 Opinions on error correction and areas of instruction teachers focus on/believe are
important
Teacher (NS or NNS; Statements on error correction Areas of instruction that they
male/female) focus on/believe are important
1 NS/M Would correct instantly if there Believes intonation should be the
was a mispronounced word. main focus
However, errors with phonemes
are not so important as the
meaning is usually clear due to
context
2 NS/F Corrections are usually done on a Individual sounds and words are
one to one level in tutorial. problematic but intonation and
Students often repeat the same sentence stress are the most
errors in spite of correction. Does important considerations.
not correct every word: those Students usually have quite good
words connected with the target intonation by the end of the
subject are important course
3 NS/M Feedback is usually provided Improvement in chunking is the
individually most achievable goal. Students
identifying nuclear stress is also
important. Word stress and weak
forms are important but it is
difficult to teach them over a
short time
4 NNS/F Does not give a lot of feedback. It Although mispronunciation of
is difficult to correct in front of individual sounds can be a
peers. Any correction is best done problem, the focus should be on
individually macro factors, such as stress and
intonation. Students like
exercises involving intonation
and chunking
Source Author’s own
4.3 Research Process 95
Table 4.5 Suggestions on how the pre-sessional can be improved in providing pronunciation
instruction
Teacher (NS or NNS; male/female) Suggestions on how the pre-sessional course can be
improved
1 NS/M Materials could be better. There should be specific
pronunciation lessons and instruction should be more
structured
2 NS/F Materials could be better. Instruction should be more
structured and more thought out. Listening activities
should be better thought out. There is no guidance on how
instruction should be provided
3 NS/M Materials are poor. Listening is not integrated. We should
use listening more intensively to focus on pronunciation.
We should record visiting lecturers to produce NNS
listening materials. There is a lack of scaffolding
embedded on the course. Very little guidance is provided
4 NNS/F A lack of materials. Listening is not systematic and
neglected. There is a lack of time devoted to
pronunciation. We need clear targets and goals. More
guidance is needed
Source Author’s own
(2002), a common theme was that teachers stated that there was very little or no guid-
ance in terms of providing instruction. The general consensus was that the syllabus
should contain more pronunciation elements and that teachers would welcome more
guidance.
During the pre-sessional course of 2017, there did seem to be widely differing opin-
ions on two issues. Firstly, a new listening exam which incorporated dictation was
piloted halfway through the course. This was criticized by many teachers for a number
of reasons, but mainly because it was viewed as unreasonable to test segmental recep-
tive skills on a course where the understanding of the meaning and the message were
paramount (for example, in understanding lectures). The new format was not used
for the final exam, but this pilot did seem to reveal a lack of clarity of the course’s
pronunciation goals. The second issue concerned whether students should be asked
to read aloud instructions to a task in order to improve pronunciation. Both course
managers and teaching staff seemed to disagree on this issue, with some arguing that
it was good practice for students and others arguing that it was rather time-consuming,
that it could destroy the confidence of weaker students, or that it was not authentic.
Again, there seemed to be a lack of clarity for teachers, with those who employed the
strategy receiving different feedback when they were observed by different members
96 4 Methodology: The Mixed Methods Approach
of senior staff. Such issues of institutional goals, to what extent teachers implement
those goals, and how they feel about those goals are precisely the kind of issues that
I wished to explore further in the study proper. As in the main study, I also took notes
and examined course documents in order to detail any particular guidance provided
in terms of pronunciation instruction. Aside from the issues already mentioned (the
proposed dictation element of the exam and encouragement to get students to read
out instructions), the only other input regarding instruction was a one hour session
on how dialogic interaction can be used to enable learners to participate and maxi-
mize their progress, particularly in terms of their ability to engage in purposeful
academic talk. Although not directly related to pronunciation, it could be said that
this session, in an indirect way, was related to pronunciation because it concentrated
on maximising learner spoken output and scaffolding. In terms of assessment, stan-
dardization sessions did take place during the course although the actual meaning and
interpretation of the descriptors and the pronunciation categories were not discussed.
Rationale
Document analysis is a common supplementary method employed in mixed methods
research (Bowen, 2009). In particular, as Yin (1994) points out, the analysis of docu-
ments is particularly useful when combined with qualitative data. It may be employed
in a variety of types of research, ranging from historical to educational, and for a
variety of purposes. Bowen (2009, p. 30) cites examples of how document analysis
has been utilized to study university policy documents in order to gain information
that may then be exploited and investigated in semi-structured interviews. Although
care must be taken in terms of choosing which particular documents to analyse, the
significance of the documents chosen, and how particular words may be interpreted,
document analysis can be a valuable adjunct to other research methods, providing an
unobtrusive and efficient means of identifying official policies and providing material
that can be used to feed into questions at the interview stage (Bowen, 2009).
Implementation
For the research that I undertook, it was necessary to examine certain documents.
Firstly, the syllabus (scheme of work) was an important document, not least because
teachers were advised by the CL during staff induction that although they could
depart from the scheme of work and use different materials, it was perfectly feasible
to adhere to the scheme of work for the entirety of the course. It was also neces-
sary to examine it in order to formulate additional questions regarding its content
during the semi-structured interviews and inspect the specific suggested materials. I
was particularly interested in examining whether the listening extracts were exam-
ples of native speaker or non-native speaker varieties. Another major focus was
the type of pronunciation activities involved, whether they were of a segmental or
suprasegmental nature, and how successful they were according to the three teacher
participants (and myself as part of my own action research).
Another important document I examined was the seminar marksheet that teachers
use when assessing speaking. This was important because it was necessary to examine
how teachers and the CL interpret the particular descriptors in the pronunciation cate-
gory, for example, whether they award broadly similar marks for the same perfor-
mance, and whether they share the same definitions of the errors that interfere with
understanding. I was also interested in examining the number of descriptors used in
order to assess the reliability of marking in the pronunciation category. Furthermore,
the descriptors were particularly relevant because they show the relative weight given
to segmental and suprasegmental features of pronunciation.
It was also necessary to examine a number of documents that were distributed
during the staff induction. Firstly, a guidance document for the seminar marking
was crucial because it was intended to provide an explanation of how the pronun-
ciation category descriptors should be interpreted. Secondly, the associate lecturers
guide was important because it could contain information regarding organizational
goals or guidance in providing pronunciation instruction. The student handbook was
98 4 Methodology: The Mixed Methods Approach
also important because the goals of the course in terms of pronunciation might be
explicitly stated there.
arise in the classroom, and identify possible ways of overcoming these difficulties.
Put simply, the action research component and the video recorded observations are
important because they address the practical element of instruction, something that
Widdowson (1994) argued should be a major concern for applied linguists. From my
own personal point of view, action research in the current study was also valuable in
terms of teacher development. This is also seen as one of the advantages of action
research (Crookes, 1993).
Adelman (1993) outlines the origin of action research, beginning with Lewin’s
study of a factory, which showed that workers were more productive when they
became a meaningful, active part of the decision-making process. Although there
are a number of different types of action research, the current study can be classified
along the lines of the type described by Ellis (1997): “research undertaken by teachers
in their own classrooms with a view to improving local practices” (p. 23).
Carr and Kemmis (1986) outline the action research procedure, which consists of
a plan of action to deal with a problematic area: in my own case, an attempt to improve
the suprasegmental pronunciation skills of language learners. This is then followed
by an action: learner activities and teacher input that will hopefully achieve the goal.
The third stage involves an observation of the results, which is then followed by a
reflection on the success (or failure) of the action initiated. This reflection then feeds
back into the next planning stage. Action research is therefore cyclical in nature. I
did consider the possibility of undertaking critical action research (outlined in Ellis,
1997, p. 24), with other teachers participating, but concluded that teachers would be
reluctant to participate in such research, particularly bearing in mind that they had
already consented to being video recorded and interviewed. I felt that teachers would
be unwilling to sacrifice the time and effort to undertake such action research, and that
being able to review the video recorded lessons and ask questions in a semi-structured
interview towards the end of the course, together with my own action research, would
be sufficient. My own action research is therefore not collaborative, unlike the Adult
Migrant English Program operating in Australia, described by Burns (2009, p. 293),
but individual in nature, being similar to the action research cases described by
Edge (2001), and in particular, similar to the individual action research conducted by
Gunn (2005) and Li (2006). The aim of the action research is to address the challenge
identified by Roberts (1998), that is, to view teacher development within the teaching
context: in my own context, my own attempts to provide suprasegmental instruction
that improves intelligibility and comprehensibility that will assist students to not only
pass their course, but also communicate effectively inside and outside the classroom
during their degree courses.
Implementation
My action research consisted of four main elements. First, maintaining a record
of the guidance provided on instruction during induction sessions and professional
development sessions. Second, maintaining a diary detailing activities I chose to use
and reflections on their success. Third, completing a checklist of any error correction
of pronunciation undertaken during my class in weeks three, four and eight. Finally, I
recorded one of my lessons in order to better reflect on how pronunciation instruction
100 4 Methodology: The Mixed Methods Approach
was incorporated into my speaking classes and to perhaps provide ideas for future
improvement. The video recording would also be useful in checking how accurately
the checklist reflected the actual content of the lesson (I filled in a checklist for my
video recorded lesson in week four) and would also provide observation data to
compare with other teacher observations.
I took notes during the induction and staff meetings that occurred throughout the
course, typing up the notes later in the evening (see Appendix C). Likewise, after
teaching a speaking lesson, I made notes detailing, for example, how I attempted
to incorporate pronunciation instruction into the lesson, how effective the listening
materials were, any difficulties I faced, and how I estimated the effectiveness of the
instruction that I provided. I did also record certain pronunciation activities in my
writing class although there were obviously far fewer opportunities to incorporate
pronunciation instruction into those lessons. Other particular events that were relevant
for pronunciation instruction were also recorded. I made notes of these reflections
throughout the course, usually at the end of the teaching day (see Appendix D).
The three checklists on pronunciation error correction were completed after
sessions with my speaking class in weeks three, four, and eight (the three teacher
participants also filled in a checklist for the same sessions in weeks three and four).
I filled in the checklist that I had devised directly after the lessons in the classroom.
These checklists were piloted by myself in week one. I found that it was possible
to fill in the particular categories in two or three minutes after the lesson. I believe
that the checklist produced provided valuable data without being too complicated to
complete.
The speaking lesson that I chose to video record was relatively easy to organize.
Firstly, I obtained consent for the filming from the students. They were fully informed
of their anonymity, their right to withdraw from the research, and how the video
recording would be used. I set up the camera in the left corner of the classroom
facing the front of the class. This was so that the video would capture any particular
use of body language used by myself to aid instruction.
Participant Observation: Rationale
Participant observation is particularly useful in terms of examining groups, orga-
nizations, and institutions (Kurz, 1983). Although participant observation has been
employed in educational settings (Ahola & Lucas, 1981), my own participant obser-
vation differs from the traditional type because I am not an outsider. Also, my own
research is limited to the specific focus of pronunciation instruction and particu-
larly to the guidance provided in official course documents (syllabus, student/staff
handbook, and seminar exam standardization materials) and by the CL during staff
induction, teachers’ meetings, and teacher development sessions. I documented how
pronunciation instruction was dealt with and also made notes on my own experiences
to further illustrate and catalogue the challenges of providing instruction. Participant
observation is often associated with investigating the extent to which institutional
goals are implemented (one of the most notable examples being Goffman, 1961) and
has been shown to be a useful method when applied to educational contexts (Yazan,
2015; Yin, 1994). My own overt participant observation was valuable in describing
4.4 Methods and Tools 101
Rationale
Teacher observation formed an important part of my research. The value of using
teacher observation to view actual teacher practices directly, as opposed to relying
solely on what teachers say they do in the classroom, can play a valuable role in
mixed methods research (Dörnyei, 2007, pp. 176–177). Indeed, direct classroom
observation has been a major element of research into pronunciation instruction in
recent times, particularly when combined with semi-structured or stimulated recall
interviews (Baker, 2011; Baker & Burri, 2016; Tergujeff, 2012). Baker (2011) argues
that a lack of direct observation in investigations into pronunciation instruction has
often led to research that, although valuable, lacked the important practical dimen-
sion in its analysis (see also Borg, 2006). The valuable opportunity to observe teacher
practices in a natural, authentic setting, and therefore to hopefully yield extremely
102 4 Methodology: The Mixed Methods Approach
participants become more relaxed and accustomed to being filmed. Other difficulties
posed by video recording mentioned by Dörnyei (2007) include video blind spots
and positioning the camera so that the full scope of classroom interaction is captured.
Video recordings have been used to examine pronunciation activity types and
techniques employed (Baker, 2011) and pronunciation feedback (Baker & Burri,
2016). These two pieces of research seem to show that direct video recorded obser-
vation, when combined with semi-structured interviews, can be valuable in exploring
the complex relationship between teacher attitudes, beliefs, and practice. The former
study involved video recording alongside an observation protocol. However, because
classroom interaction was recorded, Baker (2011) was able to collect the data and
classify the results at a later stage. The initial classifications that were made in terms
of pronunciation activities involved several categories that were later grouped into
controlled, guided or free activities. Free activities were only tallied if they were
directly linked to the development of pronunciation. The examination of teacher
cognition was addressed by both semi-structured and stimulated recall interviews.
The former were coded by utilizing Shulman’s theory of knowledge (1986, 1987)
in order to account for content knowledge, pedagogical knowledge, knowledge of
student learners, and knowledge of the curriculum. Furthermore, in order to enhance
the reliability of the data, coders were employed in order to cross-check the data
coding. My own research observation is more descriptive in nature, and observation
is utilized to better ascertain whether the syllabus content and goals are adhered to in
terms of suprasegmental instruction, how instruction is implemented, and how error
correction is conducted.
Checklists
Checklists have been used in several fields of research, for example, in testing
computer tools and systems (Wilson, 2013). Roberts (1998, p. 171) points out the
advantage of providing a standard instrument for research, and a certain flexibility
is also provided as checklists can be modified (Wilson, 2013). Both of these advan-
tages persuaded me further that the use of checklists, combined with one video
recording, would be the best way to observe what was happening in the classroom.
Self-completion checklists can also be much less intrusive than video recorded obser-
vations and may therefore be easier to employ if participants are more willing to fill
in checklist data rather than be observed.
Implementation of Recordings and Checklists in Teacher Observations
Mindful of the difficulty in obtaining informed consent, mentioned by Dörnyei
(2007), I felt that it was unrealistic to observe teachers on several occasions and
that I would be more successful in gaining consent if teachers were video recorded
on one occasion and filled in a self-observation checklist on other occasions. Between
weeks two and nine, as well as for the video recorded lesson, teachers were asked to
fill in a self-observation checklist after the lesson (see Appendix E: Self-completion
checklist). I believed that it would be easier to persuade teachers and students to
participate in research involving only one video recorded lesson rather than three.
There was another reason for the choice of a self-observation checklist as a method to
104 4 Methodology: The Mixed Methods Approach
supplement video recording: Although, as Roberts (1998, p. 171) points out, check-
lists have the potential to be affected by teacher bias, I hoped to compare discrepancies
between checklist data and the recorded lesson in weeks three and four to examine
whether teachers’ perception of what they do in the classroom matched the reality
of what they actually do. It was also a useful way of examining whether teachers
followed the proposed scheme of work (the syllabus), and if they did, how the lessons
of the three teachers differed in terms of how pronunciation instruction was provided.
In terms of possible pitfalls, I was mindful of the need to devise a checklist that was
not too complicated, whilst still being complex enough to yield valuable data. The
checklist was piloted by myself with my own learner group in the first two weeks of
the course. In addition to the checklists completed by the three observed teachers, two
other teachers agreed to fill in one checklist each. Although this was not originally
planned, I decided that this additional data could be useful in providing a broader
picture of error correction. The two teachers who filled in one checklist each are
given the pseudonyms Alice and June in the research.
I video recorded four lessons (three teachers’ lessons and my own) during weeks
three and four of the course. These weeks were chosen so that observation would not
clash with any mid-term practice assessments and would also afford teachers (and
myself) some time to become acquainted with their learner groups and their individual
students. Speaking/listening lessons usually only took place on three sessions per
week, so two weeks seemed a reasonable amount of time to allow for this. I use the
pseudonyms, Bruce, Olivia, and Mark to refer to the three teacher participants. Bruce
and Olivia were recorded for the same session in week three on 25/7/2018; Mark
and myself were recorded for the same session in week four on 1/8/2018. Every
effort was made to ensure that the video recorders were placed to ensure that both
teacher and student interactions were captured. The mentor on the course, responsible
for aspects of course administration, assisted in the setting up and collection of
the video recordings. On the first recording day, the mentor set up the equipment
in the room for Bruce, and I set up the recording equipment for Olivia. On the
second recording day, the mentor set up the recording for Mark, and I set up the
recording in my own classroom. I believed that observations of simultaneous sessions
on two occasions would serve to reveal any particularly different approaches taken
by teachers to the same materials, the extent to which teachers build and reinforce
previous pronunciation work, and the type of instruction and error correction that is
favoured. A preferable strategy would have been to combine video recording with
live observation. This would have been beneficial in terms of eliminating camera
blind spots, mentioned by Dörnyei (2007, p. 185), and would have made observation
a little easier as the recordings could have been used to check certain phenomena
noted during the live observation. However, this was not possible because I was
conducting my own lessons and my own action research at the same time that other
teachers’ lessons were taking place.
Unfortunately, being unable to undertake live observations meant that I was unable
to conduct stimulated recall interviews similar to those conducted by Baker (2011),
Baker and Burri (2016), and myself in my previous research (Hodgetts, 2014).
However, I felt that being able to review the observations and address issues in a
4.4 Methods and Tools 105
semi-structured interview towards the end of the course would be valuable in terms
of exploring how teacher beliefs and attitudes combine with actual teaching prac-
tice. It would also enable me to investigate reasons for and explanations of possible
discrepancies between beliefs, attitudes and declared instruction practices, and actual
instruction practices. Observation is important for a number of reasons, and in light
of the work of Scollon (2001), may be particularly relevant in shedding light on the
cultural norms that exist among learner groups, and in particular, Chinese cultural
norms that view the instructor as a much respected provider of knowledge. It was
interesting to observe how these cultural norms manifest themselves in the classroom,
whether they impact the effectiveness of pronunciation instruction, and if they do,
what strategies teachers employ to make instruction more effective. Semi-structured
interviews then enabled me to explore some of the choices that teachers made and
some of the difficulties that they faced.
Rationale
Semi-structured teacher interviews have been used as an important method for uncov-
ering the reasons why some teachers have become reluctant to teach pronunciation,
and have been particularly effective when used in conjunction with other methods as
part of a mixed methods approach (Baker, 2011, 2014; Baker & Murphy, 2011, Cohen
& Fass, 2001). Macdonald (2002), in an Australia ESL study, showed that teachers
who were reluctant to provide instruction cited poor guidance, training policies, and
a lack of adequate materials as explanations for this reluctance. These teachers also
stated that they employed no systematic instruction and that they only drew attention
to pronunciation or made any intervention when intelligibility was threatened. In my
own research, I combined general questions relating to teacher beliefs with ques-
tions which arose from the observations in order to explore how beliefs and attitudes
informed instruction. The interviews enabled a comparison of stated practice and
actual practice, and facilitated an in-depth exploration of the specific challenges of
providing instruction on a ten-week EAP pre-sessional course. They also yielded
some possible solutions to stated problems. Phipps and Borg (2007) used classroom
observation and semi-structured interviews in order to compare teacher beliefs with
actual instruction. Although there was no specific focus on pronunciation instruc-
tion, the study showed that teacher trainees (on a DELTA course) did experience
difficulty in translating their stated beliefs into classroom instruction. The combina-
tion of observation and semi-structured interviews in my own research (Hodgetts,
2014) was also useful in examining the connections between stated beliefs and prac-
tice. Like Cohen and Fass (2001), in my own research, certain discrepancies between
teacher perceptions of their own practice and actual practice were uncovered.
106 4 Methodology: The Mixed Methods Approach
Implementation
The CL interview was semi-structured in nature in order to afford flexibility and
the freedom to explore particular areas in depth (the CL interview guide can be
found in Appendix F). The major issues addressed are as follows: institutional goals
and the target of instruction (e.g., segmental vs. suprasegmental instruction/native
speaker production as the target vs. intelligibility), whether teachers are aware of these
goals and the target of instruction, error correction (to enable a comparison with the
observation/checklist data), listening materials and other pronunciation materials,
the CL’s viewpoint on the difficulties teachers face, how well staff training/induction
prepares the staff in providing instruction, how instruction has evolved, and the
frequency of instruction. The set of 12 principal questions was augmented by seven
further questions that I added as a result of observations that I had made during the
course. For example, the CPD session stressed the importance of learners giving
academic presentations because this is something that the vast majority of learners
have to do on their future courses. I was therefore interested in why the assessment
of the presentation strand of the pre-sessional course was only formative rather than
summative in nature. I was also able to address the advice given by the CL during
induction regarding the erroneous production of the dental fricative (i.e., that it was
an error that was easily remedied).
The teacher interview guide was adapted from the pilot interview guide and
can be found in Appendix G. The teacher interview guide, like the pilot guide,
includes a setting at ease phase involving questions that participants are easily able
to answer, such as those regarding early teaching experiences, and progresses to rather
more specific questions concerning pronunciation instruction practice and attitudes
towards instruction. This progression from the general to the specific is advocated
by Dörnyei (2007, p. 140). I also aimed to remain, as far as is possible, neutral,
allowing a free expression of viewpoints, yet also focused on eliciting responses to
the specific questions in the schedule. Also, in line with Dörnyei’s recommenda-
tion (2007, p. 143), I summarized the participants viewpoints and answers at the
end of the interview, giving them the opportunity to add further information. The
teacher interviews were semi-structured in nature in order to allow an exploration
of several areas of pronunciation instruction and to allow a full investigation into
the teacher’s viewpoints. I was aware of the fact that although I was broadly satis-
fied with the pilot interviews, perhaps there were occasions when I had become
a little too involved when I should have perhaps remained silent to encourage the
participants to expand on their own views. The teacher interview guide consists
of seven main categories: background, attitudes towards instruction on the course
(syllabus), frequency of instruction, the type of instruction teachers actually provide,
teacher goals (e.g., segmental vs. suprasegmental instruction/native speaker produc-
tion as target vs. intelligibility), the origin of teacher goals, and suggestions for the
improvement of pronunciation instruction on the course.
An exploration of the rationale for actual practices and an investigation of choices
made by teachers were facilitated as the interviews took place after the observations.
The interview guide for the teachers is divided into seven distinct categories, but also,
4.4 Methods and Tools 107
like the CL guide, includes a set of questions that was added as a result of particular
teacher actions I had noted during the video recorded observations. For example,
I was interested in whether teachers believed that the modelling of the chairing
and discussion language that they provided was successful. Other issues included
the materials that they had used during the observed lessons and their rationale
for modelling (or not modelling) new vocabulary. The interview guide begins with
a background section dealing with early pronunciation teaching experiences and
teacher training before progressing to their experience on the pre-sessional. Sections
connected with teacher beliefs and practices are then addressed.
For both the teacher interviews and the CL interview, the guide was not strictly
adhered to in terms of question order. If, as sometimes happened during the inter-
views, the interviewees were speaking freely and addressed another key area that
was pertinent, I followed the natural flow of the conversation. There were also some
occasions when pertinent questions occurred to me during the interview as a result of
what the interviewee had said. In this way, I believe I was able to obtain more valid
data that assisted me in answering the Research Questions regarding the institutional
goals that existed on the course and the teacher goals, beliefs, and attitudes. As with
other research elements, participants had the right to withdraw from the research at
any stage and they were fully informed of this.
Rationale
Questionnaires are an important research tool because they are a relatively cost-
effective method of collecting large amounts of data (Dörnyei, 2003; Gillham, 2000).
Also, because they are quantitative in nature, it is easier to employ computer tech-
nology in order to collate and analyse the data (Borg, 2006; Dörnyei, 2003). Gillham
(2000) also points out that subjects may give more objective answers because they
are not under pressure and are anonymous. However, although self-completion ques-
tionnaires are free of interviewer bias, they can be problematic if questions are vague
or if piloting is inadequate (Baker, 2011; Dörnyei, 2007). Dörnyei (2003) also points
out the potential problems caused by poorly worded questionnaires, the possibility
of dishonest responses, and lack of motivation in learner questionnaires. However, I
felt that after piloting, my questionnaire data would prove useful in supplying infor-
mation relating to student preferences in terms of pronunciation instruction and a
description and evaluation of the pronunciation instruction they had received.
There have been several studies that have used questionnaires to investigate a
variety of aspects of pronunciation instruction. Foote, Holtby and Derwing (2011)
surveyed ESL teachers across Canada and found that teachers declared employing
more segmental activities rather than suprasegmental ones when compared to the
earlier study of Breitkreutz, Derwing and Rossiter (2001). Student questionnaires
have revealed a learner preference for a native-like model of instruction, with learners
108 4 Methodology: The Mixed Methods Approach
generally regarding a native accent as both more important and more desirable (Baker,
2011; Derwing, 2003; Sifakis & Sougari, 2005; Timmis, 2002). The Timmis (2002)
study surveyed learner attitudes in 14 different countries and found that only learners
from South Africa, India, and Pakistan favoured intelligibility as the main target of
instruction, with the remaining countries favouring a native-like model. The same
study illustrated that the preferred target of instruction for teachers was that of intel-
ligibility. There are certain valid criticisms of the way the research was gathered,
and Timmis acknowledges the limitations of the research (Shoro, 2018, pp. 4–9).
However, a learner preference for native speaker models was also found in the ESL
contexts of Canada (Derwing, 2003), the United States (Baker, 2011), and also in the
EFL context of primary and secondary school students in Greece (Sifakis & Sougari,
2005).
Other questionnaires have concentrated on learner preferences in terms of pronun-
ciation instruction techniques (Baker, 2011; Hismanoglu & Hismanoglu, 2010) and
preferences in terms of error correction (Baker, 2011; Cathcart & Olsen, 1976).
Couper’s (2003) study of attitudes to pronunciation instruction revealed that although
learners expressed a low level of confidence in their own pronunciation, they viewed
pronunciation instruction as an important element of instruction. A low level of
confidence and a desire for students to improve their own pronunciation was also
apparent in Kang’s (2010) ESL research in New Zealand. A similar desire was also
expressed by Baker’s learners (2011) although the learners in this sample displayed
quite high confidence levels. Questionnaire data on student attitudes towards pronun-
ciation instruction and the pronunciation instruction they receive on the course were
instrumental in allowing a more thorough insight into what learners on the course
think of instruction.
The Construction and Implementation of the Learner Questionnaire
The learner questionnaire (see Appendix H) is an adaptation of the one utilized by
Baker (2011). Likert scales are particularly appropriate for eliciting opinions (Brown
& Rodgers, 2002, p. 120), and I therefore decided to use Baker’s questionnaire as
a starting point. Question 1 consists of 30 statements. The first thirteen statements,
aside from small changes made to statements 7 and 11, are identical to Baker’s. These
broadly cover: confidence and motivation (1–4), learner preferences in terms of error
correction (5–8), and learner preferences in terms of language models and the type
of accented English they want to speak (9–13). The questionnaire was augmented by
three categories not included in the Baker (2011) study: the first encompasses learner
evaluation of their own progress, materials used, and listening activities (14–20); the
second is directed towards the type of error correction that was present during classes
(rather than what they prefer: 21–27); and the third category aims to assess broadly
three ways in which independent learning may have been encouraged on the course
(28–30).
Although the biographical data, open-ended questions, and quasi-open ended
questions at the end of the questionnaire (questions 2–7) are broadly similar to
Baker’s study, one or two slight alterations were made. For example, I replaced a
question concerning the specific programme that Baker was investigating with a
4.4 Methods and Tools 109
question asking students to identify the activity that helped them improve (question
4). As Dörnyei recommends (2007), I endeavoured to ensure that the closed ques-
tions on the Likert scale were unambiguous, brief, and written in relatively simple
language (particularly in view of the fact that respondents would most likely have
an IELTS level at or below 6). The questionnaire consisted of two pages (again,
in line with Dörnyei’s recommendations) so as to be less daunting. It was clearly
legible, contained four sides of information (the two A4 size sheets being stapled
together), and questions progressed from direct, relatively simple, Likert scale ones
at the beginning, to open questions later in the questionnaire (also recommended by
Dörnyei, 2007, pp. 111–112).
Although some researchers suggest the biographical section may be more appro-
priate at the beginning of a questionnaire (Brown & Rodgers, 2002, pp. 120–121),
I was persuaded of the value of locating the section at the end of the questionnaire
to encourage completion. Dörnyei (2007, p. 111) points out that people have to fill
in a lot of biographical data in their everyday lives when they encounter bureau-
cratic forms; it could therefore be de-motivating to encounter such a section at the
beginning of the process.
Apart from the questions themselves, there are other ways in which the ques-
tionnaire was modified. Firstly, Baker’s study consisted of a five point Likert scale:
strongly agree, agree, maybe, disagree, and strongly disagree. The first choice I made
was to remove the maybe category. I felt that the term was needlessly ambiguous. I
decided that instead of a no opinion or don’t know category at the centre of the scale,
learners would be encouraged to express an opinion by having an entirely separate
don’t know category to the right of the scale. I also chose to expand the Likert scale to
include the categories rather agree and rather disagree to further encourage respon-
dents to express an opinion rather than choosing the don’t know category. There does
seem to be some evidence that the use of a seven point Likert scale may yield more
reliable results (Krosnick & Presser, 2009, p. 20). I chose to begin with strongly
disagree on the left rather than the right, progressing to the strongly agree category
on the right.
Piloting of the Learner Questionnaire
I initially distributed the questionnaires to my own writing class in week eight in
order to pilot the questionnaire and identify any problematic questions. My own
writing class was the weakest in terms of the IELTS scores that students had before
embarking on the pre-sessional. I therefore surmised that if my own students were
able to understand and complete the questionnaire in around 20 min, then students
in other classes should also be able to complete it. The only change I made to the
questionnaire at this stage was to simplify the language a little on the first page
(information about the questionnaire). This was because I noticed that students in
my own class were having difficulty with some words, and it was time consuming
to explain some of the vocabulary items involved. I attempted to simplify the dense
language whilst retaining the important information regarding the nature and scope of
the research, the guarantee of anonymity, and the right to withdraw from the research.
I noticed that the open questions were answered by few students, but decided to
110 4 Methodology: The Mixed Methods Approach
retain these questions because students in other classes generally had a higher level
of English (according to their IELTS scores). I therefore concluded that those higher
level learners might be more likely to contribute to the open questions section. I
distributed the questionnaires for teachers to hand out to their classes on the ten-week
course at the end of their lessons in week eight. I stressed that teachers could explain
any problematic language in order to assist students in completing the questionnaire,
and in particular, I emphasized that teachers should stress that the questionnaire was
not a test and was anonymous (i.e., that students did not have to write their names on
the questionnaire). The teachers then brought the questionnaires from their classes
back to the staffroom for me to collect. All teachers successfully distributed and
returned the questionnaires.
The Construction and Implementation of the Teacher Questionnaire
The teacher questionnaire can be found in Appendix I. It is an adaptation of the ques-
tionnaire used by Foote et al. (2011), and question 1 includes statements replicated
from the study, including statements regarding the target of instruction (9–10) and
ones involving segmental and suprasegmental instruction (11–14). The statements
that I devised were centred around the areas of teacher experiences/training (1–2),
the perceived value/difficulty of providing instruction (3–5), listening materials (6–
8), feedback (15–17), assessment (18–19), and the desire for more training (20).
Aside from the Likert scale statements in question 1, the questionnaire also includes
closed questions of a different type. Question 2 (i) is an open multiple choice ques-
tion which addresses the origin of knowledge regarding pronunciation instruction,
and question 2 (ii) requires teachers to nominate the most important source of their
knowledge in providing pronunciation instruction. Question 3 is a ranking question
which addresses the perceived importance of segmentals and suprasegmentals.
As in the student questionnaire, open questions are included towards the end of the
questionnaire. These questions refer to suggestions for improvements on the course
(question 4) and the major difficulties faced (question 5). There are also biographical
data questions which, as in the student questionnaire, were included at the end (in
line with Dörnyei’s recommendations, 2007, p. 111).
The questionnaire was also adapted in exactly the same way as the student ques-
tionnaire: the Likert scale was extended to a seven point scale in order to encourage
responses, and the don’t know category is expressed as a distinct category, separate
from the rest of the scale.
Piloting of the Teacher Questionnaire
The questionnaire was distributed to teachers from the five, ten and fifteen-week
pre-sessional courses. Each course had nine, eleven, and five teachers respectively
(excluding myself from the ten-week course). I had originally intended to distribute
the questionnaires earlier in the course in order to rectify any potentially problem-
atic questions. However, due to the demands of the course and other elements of
the research, such as the video recording of lessons and keeping a record of my
teaching, teacher questionnaires were not distributed until the beginning of week
nine. I distributed the questionnaires at times when teachers were less busy and
4.4 Methods and Tools 111
explained the purpose of the questionnaire, that anonymity was guaranteed, and that
completion of the questionnaire was entirely voluntary. Because teachers returned
their questionnaires intermittently, I was able to monitor the first five or six that
were returned as a very basic way of piloting the questionnaire. At that time, I was
satisfied that all questions had been understood in the way they were intended, so I
was satisfied that no further adjustments or amendments were needed. When more
questionnaires were returned, it seemed clear that one of the questions had been
misunderstood. Question 3, the ranking question, was graded with a mark, rather
than a rank by some participants. This meant that different items in that question
were given the same score. Aside from this question, I was satisfied that the ques-
tions seemed to have been understood and interpreted as I had intended. From a total
of 25 teachers, 23 questionnaires were returned (see Table 4.6).
Rationale
A qualitative investigation of raters assessment of pronunciation was used in the
Kennedy et al. (2017) research into French language raters. In their study, four raters
were asked to rate learners according to four constructs: accentedness, comprehen-
sibility, fluidity, and communicative effectiveness. The researchers were not only
interested in the marks awarded to the candidates, but also the rationale that was
provided for their marking. The raters were therefore asked to make comments on
the pronunciation of the particular candidates. The rationale for decision making
is important because it may well be the case that the same marks are awarded for
completely different reasons (Isaacs & Thomson, 2013). If researchers wish to gain
knowledge of teacher cognition, it is therefore useful to ask raters to try to explain their
decision-making. Although my own research was specifically focused on how raters
scored a summative assessment, audio and video recordings have also been used in
other research in order to uncover rater attitudes towards the social attractiveness of
accents (Buckingham, 2014; Lindemann, 2017, pp. 201–203).
112 4 Methodology: The Mixed Methods Approach
Implementation
I chose to investigate how teachers mark the pronunciation category by asking Mark,
Bruce, and Olivia to mark a student seminar. The three teacher participants evaluated
one student seminar (consisting of three students) in weeks five and six. The seminar
was video recorded from my speaking class from the 2017 pre-sessional. It was a
practice exam, and therefore followed exactly the same procedure and format as the
final exam. I chose a seminar that involved students of varying ability and a recording
with the clearest audio. As mentioned earlier, all participants, including those filmed
during the 2017 course, were informed of the nature of the research, their guaranteed
anonymity, and the right to withdraw from the research at any time.
In order to ensure authenticity, subjects were asked to mark all categories (as in the
exam). They were then asked to provide a rationale for their pronunciation mark. This
would prove useful in explaining which pronunciation features teachers concentrate
upon and identifying any discrepancy in the mark ranges. In order to ensure that
teachers were not influenced by other markers, the participants completed the task
individually in a quiet area of the university building. Student seminars consist of
three or four students (depending on the number of students per class). I chose a
video recorded seminar with three students from the previous year’s practice test. A
seminar with three students was chosen for two reasons. Firstly, an examination group
of three students is much more typical. Secondly, it is a little easier to mark a seminar
group with three students. I wanted the results to be an accurate representation of
the marking that we usually do, rather than an examination of more difficult marking
scenarios. Furthermore, seminar groups of four students are usually double marked,
so any variability in marking should be less of an issue with those groups. The teacher
participants were asked to watch the recording of the seminar practice and grade the
performances, underlining descriptors on the seminar assessment worksheet, very
much in the same way as in the actual seminar exam. I did not ask teachers to
simply complete the pronunciation category as that would not have been an authentic
representation of the marking process. I then asked teachers to explain what they
thought about the students’ pronunciation. These responses were recorded in order
to help me recall and analyse them. In particular, I was interested in how teachers
interpreted segmental and suprasegmental errors and how the words in the descriptors
were interpreted. Of course, the marks given by teachers in the pronunciation category
for each student would also serve to show whether the three participants interpreted
the descriptors in a broadly similar way.
involved the video recording of one of my own lessons. I was therefore a participant
in the research myself in addition to being a researcher. The participants in the video
recording and checklist data consisted of three teachers (two males and one female)
and myself. The choice of the three teacher participants was dependent on a number
of factors. Firstly, I intended to avoid involving teachers who were new employees.
This was because I was aiming to choose participants who were more representative
of the overall sample: of the twelve teachers on the ten-week course (including
myself), only four were new teachers. Another reason for not choosing new teachers
as participants was that I did not wish to add further burdens or introduce stressful
situations (e.g., the recording of lessons) to teachers who were in a new working
environment and probably getting to grips with new organizational procedures and
working practices. I therefore attempted to persuade experienced teachers on the pre-
sessional to participate. Another consideration was willingness to participate. The
three participants would have to consent to the video recording of one lesson, filling
in checklists after lessons, grading and explaining their marks of a student seminar,
and an interview at the end of the course. Although using checklists meant that
participation was less onerous, with only one recorded lesson required, nonetheless,
a certain amount of the teachers’ time would be required (for the interviews and
seminar grading), and not all teachers would be comfortable being filmed. Indeed, I
had originally planned to involve the only non-native speaker teacher on the ten-week
course in order to gain an insight into the perspective of non-native teachers (as I
had done in the pilot interviews), but was unable to obtain consent. However, I was
able to obtain consent from two male colleagues and one female colleague: Mark,
Bruce and Olivia. They are all experienced teachers, having worked on the course for
between five and eight summer sessions. Mark and Bruce both have Trinity TESOL
Diploma level qualifications; Olivia has both a PGCE level qualification and a Ph.D.
in the study of a foreign language. All three teachers were made fully aware of the
scope and nature of the research and were content to participate. In addition to the
video recording of one lesson of the three teacher participants, I also recorded one
of my own lessons as part of my own action research.
The CL was an important participant in the research, providing valuable data
during the semi-structured interview. The CL was a teacher and assistant course
leader on the pre-sessional for many years before becoming the CL in 2017. Before
becoming an EAP teacher, he worked overseas as an EFL teacher for several years
and is MA TESOL and Trinity Diploma qualified.
Firstly, during the 2017 pre-sessional course, on 21/08/2017, at the university where
the main study was held, informed written consent was given by the CL and student
participants for the filming of a student practice seminar. Student participants were
fully informed that they would remain anonymous, of the voluntary nature of the
research and their right to withdraw from the research should they so wish, and
114 4 Methodology: The Mixed Methods Approach
that the recorded materials would be used for research purposes only. The recorded
materials were then used for the teacher assessment of the student seminars element
of the research in 2018.
Secondly, informed written consent was sought and received for the video
recording of one lesson from four learner classes: the speaking groups led by myself,
Mark, Bruce, and Olivia. Once again, students were fully informed of the nature of
the research, their anonymity, that their participation was entirely voluntary, and that
they could withdraw from the research at any stage. Having been fully informed,
students gave written consent for their participation. Table 4.7 outlines the number
of students in each class, their L1, and the date that written consent was obtained.
The four learner groups consisted of learners intending to study a variety of courses.
Tables 4.8, 4.9, 4.10, and 4.11 detail students’ IELTS scores in each group when they
arrived on the pre-sessional course.
The learner groups of Mark, Olivia, and myself were broadly similar in terms of
the level of the students (according to their IELTS scores), and they all consisted
of Chinese speakers. Bruce’s group differed, not only because it contained learners
with an L1 other than Chinese, but also because the level of students’ listening and
speaking skills (as well as their overall score) was much higher than the other groups.
Indeed, I was satisfied that Bruce agreed to participate in the research because his class
provided an opportunity to view how instruction might differ due to the circumstances
of the group being multilingual and of a higher level.
There were a total number of 177 students on the course; of this number, 161 learners
completed the questionnaire. The groups consisted of 136 Chinese speakers, 10
Arabic speakers, 5 Thai speakers, 4 Russian speakers, and 1 Japanese speaker. There
4.5 The Participants 115
Table 4.8 John’s group: IELTS scores at the start of the course
Listening, speaking and overall IELTS
Student Listening Speaking Overall
1 5.0 5.5 5.0
2 5.0 4.5 5.0
3 5.0 5.0 5.0
4 5.0 5.0 5.0
5 5.0 5.0 5.0
6 5.0 4.5 5.0
7 4.5 4.5 5.0
8 5.0 4.5 5.0
9 5.0 5.5 5.0
10 5.5 4.5 5.0
11 4.5 4.5 5.0
12 4.5 4.5 5.0
13 4.5 4.5 5.0
14 5.0 4.5 5.0
15 5.0 4.5 5.0
Source Author’s own
were also 5 no data questionnaires where the participants did not fill in the question
about their native language. Although there is no precise data on the gender of
the learners as this was not a variable I was interested in examining, there were
approximately equal numbers of male and female learners on the course.
The research fully complies with the ethics guidance of the British Educational
Research Association, adhering to the guiding principles of beneficence, non-
malfeasance, integrity, informed consent, confidentiality and anonymity, and inde-
pendence and impartiality. In particular, I would like to stress that all participants
(CL, teachers, and students who were video-recorded, interviewed or who filled in
self-completion questionnaires) were fully informed of the nature of the research,
their anonymity, and their right to withdraw at any stage. Furthermore, the univer-
sity was offered free access to a final version of the thesis. I would like to stress that
during the research, I strived to be flexible and accommodating in terms of the timing
of the research involving the teacher, CL, and student participants, being willing to
schedule the research elements for their convenience. The consent forms for the
CL, the teacher participants, and learners can be found in Appendices J, K, and L
respectively.
The pre-sessional course is an important source of revenue for the university and a
method by which multi-cultural and multi-lingual undergraduate and postgraduate
classes may be created to encourage a valuable exchange of ideas and knowledge
across cultures. I believe that my research could play a small part in attempting to raise
awareness of discrepancies in marking between teachers and therefore ensure more
standardized testing of pronunciation: surely a desirable outcome for the univer-
sity. Furthermore, it is hoped that the research will raise awareness of pronuncia-
tion instruction among the teacher participants and that this will in turn encourage
self reflection: also a desirable outcome. Any suggestions on how this course, and
EAP courses in general, can improve pronunciation instruction, and particularly
intelligibility-based suprasegmental instruction that better facilitates learning, are
tentatively offered.
118 4 Methodology: The Mixed Methods Approach
For the CL, the research should be illuminating in terms of providing a snapshot of
teacher/student attitudes and beliefs gleaned from the self-completion questionnaire
data. This knowledge may be helpful in identifying teacher and student needs, and
could be particularly useful in identifying how those needs may be met, for example,
by making adjustments to the syllabus, the course induction, or guidance that is
provided. It could also be of use in providing knowledge that can be utilized for
teacher training purposes in CPD sessions. Similarly, the element of the research
involving teacher assessment of seminars could lead to possible improvements in
terms of improving the standardization of marks. Once again, any suggestions on
how the course may be improved are tentatively offered.
4.6.4 Myself
The self-reflection that may be encouraged in the teacher participants also extends to
myself. Indeed, this self-reflection is enhanced because I engaged in action research
and took notes, detailing observations on how I provided instruction and the diffi-
culties I encountered. Again, this self-reflection should be beneficial in assisting my
own development, both as a researcher and practitioner. The research gave me a
valuable insight into the various actors on the pre-sessional course, particularly the
CL and the teachers. This knowledge and understanding should also benefit my own
teaching development. One of the most satisfying and enlightening elements of the
research from my own point of view was interviewing the teachers and learning of
their own journeys as teachers in terms of how their pronunciation instruction beliefs
were formed. Although teachers do interact on a day to day basis, it strikes me that
4.6 Stakeholder Analysis 119
many of these interactions are either general conversational interactions or, if they
do involve teaching, comprise of simple discussions about teaching activities. Such
conversations rarely extend to teacher beliefs, if at all, and I am grateful to those
participants for being so open and honest about their beliefs, attitudes, and practices.
4.6.5 Learners
For the learners, the video recording of one lesson was consented to by student
participants in the four classes (including my own). There was no external observer
in the classroom, and teacher participants stated that learners soon became unaware of
the video camera after the first few minutes of the lesson. I noticed a similar effect in
my own class; after the first few minutes of excitement and curiosity, learners became
oblivious to the camera. Therefore, I do not believe that the research had either a
positive or negative effect on those students. Similarly, for the students who chose to
fill in the self-completion questionnaire, I anticipate no positive or negative effects.
However, for future students, it is hoped that the research could have a positive effect
if information is gained that could assist in improving the provision of pronunciation
instruction and its assessment.
References
Adelman, C. (1993). Kurt Lewin and the origins of action research. Educational Action Research,
1, 7–24. https://doi.org/10.1080/0965079930010102
Ahola, A. A., & Lucas, B. G. (1981). Participant observation in educational research: Dynamics of
role adjustment in a high school setting. Mc Gill Journal of Education 16(1). 76–90. Retrieved
January 21, 2019, from https://mje.mcgill.ca/article/viewFile/7382/5312.
Angrosino, M. V., & dePerez, K. A. (2000). Rethinking observation: From method to context. In
N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp. 673–702).
Thousand Oaks, CA: Sage.
Baker, A. (2011). Pronunciation pedagogy: Second language teacher cognition and practice
(Doctoral dissertation, Georgia State University, Atlanta, US). Retrieved March 26, 2018, from
https://scholarworks.gsu.edu/cgi/viewcontent.cgi?article=1017context=alesl_diss.
Baker, A. (2014). Exploring teachers’ knowledge of second language pronunciation techniques:
Teacher cognitions, observed classroom practices, and student perceptions. TESOL Quarterly,
48(1), 136–163. https://doi.org/10.1002/tesq.99
Baker, A., & Burri, M. (2016). Feedback on second language pronunciation: A case study of EAP
teachers’ beliefs and practices. Australian Journal of Teacher Education, 41(6), 1–19. https://doi.
org/10.14221/ajte.2016v41n6.1
Baker, A., & Murphy, J. (2011). Knowledge base of pronunciation teaching: Staking out the territory.
TESL Canada Journal, 28(2), 29–50. https://doi.org/10.18806
Bartels, N. (2002). Professional preparation and action research: Only for language teachers? TESOL
Quarterly, 36(1), 71–79. https://doi.org/10.2307/3588361
Borg, S. (2006). Teacher cognition and language education: Research and practice. London:
Continuum.
120 4 Methodology: The Mixed Methods Approach
Kang, O. (2010). ESL learners’ attitudes toward pronunciation instruction. In J. M. Levis & K.
LeVelle (Eds.), Proceedings of the 1st pronunciation in second language learning and teaching
conference (pp. 105–118). Ames, IA: Iowa State University.
Kennedy, S., Blanchet, J., & Guénette, D. (2017). Teacher-raters’ assessment of French lingua franca
pronunciation. In T. Isaacs & P. Trofimovich (Eds.), Second language pronunciation assessment
(pp. 210–236). Bristol: Multilingual Matters.
Krosnick, J. A., & Presser, S. (2009). Question and questionnaire design. In J. D. Wright & P. V.
Marsden (Eds.), Handbook of survey research (2nd ed., pp. 1–81). San Diego, CA: Elsevier.
Kurz, D. E. (1983). The use of participant observation in evaluation research. Evaluation and
Program Planning, 6(2), 93–102. https://doi.org/10.1016/0149-7189(83)90022-8
Li, N. (2006). Researching and experiencing motivation: A plea for ‘balanced research’. Language
Teaching Research, 10(4), 237–456. https://doi.org/10.1191/1362168806lr206pr
Lindemann, S. (2017). Variation or ‘error’? Perception of pronunciation variation and implications
for assessment. In T. Isaacs & P. Trofimovich (Eds.), Second language pronunciation assessment
(pp. 193–209). Bristol: Multilingual Matters.
Macdonald, S. (2002). Pronunciation-views and practices of reluctant teachers. Prospect, 17(3),
3–16. Retrieved from https://hdl.handle.net/1959.14/323083.
Nunan, D. (1990). The teacher as researcher. In C. Brumfit & R. Mitchell (Eds.), Research in the
language classroom (pp. 16–32). London: Modern English Publications.
Phipps, S., & Borg, S. (2007). Exploring the relationship between teachers’ beliefs and their
classroom practice. The Teacher Trainer, 21(3), 17–19. https://doi.org/10.1016/j.system.2009.
03.002
Roberts, J. (1998). Language teacher education. London, UK: Arnold.
Scollon, R., & Scollon, S. W. (2001). Intercultural communication (2nd ed.). Oxford: Blackwell.
Shoro, S. (2018). Teaching and learning English in non-English-speaking countries. Newcastle
upon Tyne: Cambridge Scholars Publishing.
Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educational
Researcher, 15(2), 4–14. https://doi.org/10.3102/0013189X015002004
Shulman, L. S. (1987). Knowledge and teaching: Foundations of the new reform. Harvard
Educational Review, 57(1), 1–23. https://doi.org/10.17763/haer.57.1.j463w79r56455411
Sifakis, N. C., & Sougari, A. M. (2005). Pronunciation issues and EIL pedagogy in the periphery:
A survey of Greek state school teachers’ beliefs. TESOL Quarterly, 39(3), 467–488. https://doi.
org/10.2307/3588490
Tergujeff, E. (2012). English pronunciation teaching: Four case studies from Finland. Journal of
Language Teaching and Research, 3(4), 599–607. https://doi.org/10.4304/jltr.3.4.599-607
The University of Sheffield. (2017, July 12). BALEAP Webinar: Ten practical ideas for teaching
pronunciation in EAP [Video webinar]. Retrieved January 28, 2019, from https://uni-sheffield.
adobeconnect.com/pno7lgb00vgh/.
Timmis, I. (2002). Native-speaker norms and international English: A classroom view. ELT Journal,
56(3), 240–249. https://doi.org/10.1093/elt/56.3.240
Wajnryb, R. (1992). Classroom observation tasks: A resource book for language teachers and
trainers. Cambridge: Cambridge University Press.
Widdoson, H. G. (1994). The ownership of English. TESOL Quarterly, 28(2), 377–389. https://doi.
org/10.2307/3587438
Wilson, C. (2013). Credible checklists and quality questionnaires: A user centred design method.
Waltham, MA: Elsevier Inc.
Yazan, B. (2015). Three approaches to case study methods in education: Yin, Merriam, and Stake.
The Qualitative Report, 20(2), 134–152. Retrieved from https://nsuworks.nova.edu/tqr/vol20/iss
2/12.
Yin, R. K. (1994). Case study research: Design and methods (2nd ed.). Thousand Oaks, CA: Sage.
Zielinski, B., & Yates, L. (2004). Myth 2: Pronunciation instruction is not appropriate for beginner-
level students. In L. J. & Grant D. M. Brinton (Eds.), Pronunciation myths : Applying second
language research to classroom teaching (pp. 56–79). Ann Arbor: University of Michigan Press.
Chapter 5
Results, Analysis, and Discussion
5.1 Introduction
The following chapter presents the key results derived from the methods employed,
analyses their meaning, and discusses the implications. Firstly, the results are
presented, with extracts from the semi-structured interviews provided within the
text so that the reader can easily refer to the salient elements. The results obtained
are then analysed sequentially in terms of each Research Question. The discussion
section of the chapter then explores the key findings and explores issues pertinent
to intelligibility-based suprasegmental instruction in terms of syllabus goals and
content (along with the guidance provided by the course leader), assessment, the
use of pronunciation teaching materials, conclusions and suggestions derived from
teacher practices, and error correction. Where relevant, reference is made to the
literature explored in Chaps. 2 and 3 throughout the discussion section.
Although the course is a ten-week course, all of the sessions in week ten, apart
from the Monday sessions (which have two 90 min sessions dedicated to reading,
listening and seminar exam preparation) involve non-teaching activities, i.e., summa-
tive and formative final assessments. The following sections describe the elements of
the scheme of work that are relevant for pronunciation instruction: syllabus content
of seminars, presentations, and listening; and the guidance provided in the asso-
ciate lecturer supplement and the student handbook, the seminar marksheet, and the
seminar guidelines and standardization document.
The suggested materials for each week have been organized into PowerPoint presen-
tations by two members of staff over the years in order to make lessons easier for staff
to administer and more visually stimulating for the learners. After week one, learners
were given reading materials for subsequent seminars (and the final exam) in order
to better prepare them for seminar discussions. Every seminar session in every week
has free speaking discussion sections which are designed to allow students to practise
speaking about topics in groups of three or four as they would in the exam. Week
nine contains this discussion section only, as it is designed to be a practice exam. In
addition to these discussion tasks, five of the PowerPoints begin with a presentation
of essential vocabulary. I was interested in this section because the observed lessons
also contained this vocabulary session. The vocabulary items are not annotated in any
way in order to show word stress or the number of syllables. Whether teachers chose
to present this information or model the pronunciation in any way was of interest,
and is dealt with in the observation section of the results. Every week (apart from
weeks nine and ten) has a seminar language section (week two has two sections), also
given to students as a handout, which presents functional language, such as language
for agreeing and disagreeing, that can be used in a seminar discussion. Of the nine
handouts, only two are annotated in any way. One of the week two handouts has the
stressed words marked in yellow and arrows to denote whether the intonation rises,
falls or remains flat. The week four handout, on chairing and concluding a discus-
sion, also has intonation arrows, but no annotation to show sentence stress. There are
therefore seven of the nine worksheets that bear no annotations, so it seems that there
are only isolated examples of suprasegmental instruction embedded in the seminar
syllabus and no examples of segmental pronunciation exercises.
In terms of the listening tasks that form part of the seminar class syllabus, there are
20 in total that accompany the seminars from weeks one to eight: twelve video tasks
and eight audio tasks, and of these, there are two videos that have no audio. Aside
from the extracts that support the seminars, the other listening extracts are exam-
ples of predominantly native speaker models. Nine extracts are of SAE (Standard
American English) and four are SSBE (Standard Southern British English). Of the
remainder, there is one example of Northern English, one London variety, and one
Australian English variety. There is also one listening extract with multiple speakers
(one standard Scottish, one Spanish, and one SSBE). Aside from the Spanish speaker
of English, the only other example of non-native accents is that of Chinese speakers
of English in week one.
In terms of listening task type, none of the listening extracts specifically involve
pronunciation; nine extracts focus on general listening skills, and two on note-taking.
Of the two videos that have no audio, one has information on the screen to read and
the other has no task instruction. There is one listening where learners have to check
whether the information given is the same as the predictions they have made, one
listening for gist, one where learners identify who is speaking, and one where their
handout contains incorrect information that they have to correct. There is also a video
5.2 Notes on the Course Documents and Induction Documents 125
of a previous student seminar that learners are expected to watch in week one (this
is the one example containing Chinese speakers of English). They are also asked to
evaluate the performances of the students by referring to a simplified version of the
exam marksheet (Appendix M) and/or the marksheet itself (Appendix A).
discourse markers and learner body language. Week three contains another listening
gap-fill activity that focuses on content: the structure of an introduction. Once again,
there is no specific focus or any annotations in terms of pronunciation features.
Week four contains two pronunciation-focused activities on chunking. The first
involves learners attempting to identify the word that will bear the prominent stress in
the sentence from a listening extract. There is a short practice warmer, where students
try to say particular sentences and place stress on the different words. This may be a
useful activity, but there is no guidance given in the materials, or indeed throughout
the course, in terms of how teachers might explain or show what a stressed word
is and what is distinctive about it. The second task focuses on chunking, and also
involves learners recording their own production. This seems to be a valuable task
in terms of aiding pronunciation.
The next week in the syllabus that contains any pronunciation element is week six
(week five is dedicated to the production of flash cards and PowerPoint slides). Week
six contains two chunking elements and one intonation element. The first chunking
part is similar to the chunking task in week four: Students have to divide a passage
into chunks and then read the passage, leaving pauses between the chunks. However,
once again, there is no focus on content before the chunking exercise. The second
chunking exercise is a little different as it attempts to show how chunking different
parts of the sentence can change the meaning of the utterance.
The intonation section has two parts (Powell, 1996, p. 42). The first part asks
learners to read a list of sayings, vary the intonation, and pay due attention to stress.
Nuclear stress is denoted by capitalization and stressed words are in bold. There is no
guidance in terms of how intonation should be used, and it is assumed that teachers
will know how to advise students. Another consideration is the applicability of the
particular sayings chosen and to what extent learners will be able to utilize any into-
nation patterns they might manage to observe when giving their own presentations.
Examples include Luck is what happens when preparation meets opportunity and
Humour is always the shortest distance between two people (Powell, 1996, p. 42).
The second intonation task at least gives guidance to the learners (and teachers) about
intonation patterns when positive information is contrasted with negative information
(rising for positive; falling for negative). This also seems to be something that might
be fairly easy to transfer to learners’ own presentations if the teacher draws attention
to such positive and negative contrasts in them. It is a little optimistic to believe that
learners will do the exercise and then begin applying it to their own presentations
and everyday speech without instructor guidance.
Week seven contains a PowerPoint recap of the ingredients of a good presentation.
There is one slide that deals with the delivery of a presentation that touches on
pronunciation. The guidance given on the slide is that learners should use natural
spoken English, use pauses for emphasis, and that their delivery should be clear,
simple, and fluent. It is difficult to gauge what learners will take from such general
advice.
Both weeks seven and eight contain a peer review checklist designed to be used
when learners review their peers’ presentations. However, in practice, there were
no presentation lessons in week eight because two of the days in that particular
5.2 Notes on the Course Documents and Induction Documents 127
week were holidays (one university holiday and one bank holiday). Week nine also
includes a self-evaluation checklist that students can use with their own presentations
if they record them. The peer feedback sheet has broad categories, such as good
pronunciation and good fluency, which learners should mark yes or no. Once again,
it is difficult to imagine how learners will interpret these broad categories.
For the remainder of the presentation syllabus, weeks eight and nine are reserved
for peer practice (independent study) and for students to complete and evaluate their
slides in class. Week ten is reserved for assessments. Unlike seminars, presentations
are only formatively assessed (as was the case in 2017). Classes involving instruction
centred around presentations generally involve at least some kind of pronunciation
instruction. It therefore might be significant that the assessment of presentations is
formative rather than summative. Before 2017, the assessment of presentations was
summative. This was changed, largely due to feedback from the British Council and
the external examiner that there were too many summative assessments.
For the specific part of the course dedicated to listening in preparation for the listening
exam, aside from one small segment at the end of the 90 min session in week two,
which involved exercises on word stress and sentence stress, the listening sessions
were either listening for detail exercises or exercises to practise note-taking. There
were some weeks when this 90 min session did not take place (weeks one, eight,
nine, and ten). This was due to a variety of reasons (e.g., week nine was set aside for
presentation practice; week ten was set aside for assessments). Of the weeks where
this session took place, the listening extracts, with one exception, were all of a native
speaker variety (either SAE or SSBE). There was a pilot listening test in week five,
which focused on note-taking and listening for detail.
The course’s website contains some speaking and listening course materials, intended
for learner self-study, and resources designed to support seminar instruction. These
consisted of reading materials in preparation for the seminar, seminar language (that
students are provided with during the lessons in any case), and listening materials.
These supplementary listening activities consist broadly of two types: the TED talk
type, with a US standard pronunciation, which could sometimes be quite long; and six
minute BBC extracts, almost always with SSBE pronunciation. The former, the TED
talks, had little or no scaffolding (except automatically generated YouTube subtitles),
and sometimes had quite advanced vocabulary (particularly one on the complex
topic of biodiversity), whereas the latter did have a number of exercises to guide the
128 5 Results, Analysis, and Discussion
The supplement does not seem to contain any information in terms of course goals.
There is a small section which deals with the scheme of work, detailing the course-
books used and stating that lecturers can use alternative materials if they cover the
same syllabus aims. Conversely, the student handbook does state general and specific
aims. The general aims state that students will be able to participate in typical patterns
of study at undergraduate or postgraduate level by the end of the course, will develop
study skills, become more familiar with the university context, and develop EAP
knowledge. More specifically, under the speaking section of the course description,
three elements are present, and one of these elements specifically mentions pronun-
ciation. It is stated that one of the three aims of the speaking part of the course is
to enable students to understand important aspects of English pronunciation. The
pronunciation goals therefore seem to be quite vague as it is uncertain which aspects
of pronunciation are important for the course. Also, it is interesting that the emphasis
is on student understanding of aspects of pronunciation. Nothing is stated about the
students being able to use these features. Perhaps this is covered in one of the other
elements, which states that students will be effective communicators in presentations
and group discussions. However, what this effective communication entails is not
clearly stated.
One of the most noticeable aspects of the pronunciation category of the seminar mark-
sheet is that, like the vocabulary category, there are only two descriptors (Appendix
A: Seminar Marksheet). The grammar and interactive communication categories
usually have three descriptors each. This could mean that there will be more varia-
tion between markers because choosing one descriptor has more of an effect on the
overall mark in the pronunciation category. One of the descriptors seems to be more
focused on segmental aspects, for example, some mispronounced words may affect
communication, whereas the other focuses on suprasegmental features although this
equal division is a little blurred as some descriptors mention mispronunciation of
words or longer stretches. These longer stretches could also include suprasegmental
features (as could words). This is an area I asked the course leader (referred to
as CL) about, who stated that one descriptor refers to segmentals and the other to
suprasegmentals.
5.2 Notes on the Course Documents and Induction Documents 129
The use of language in the descriptors could also be problematic. There appears to
be very little difference between some mispronounced words may affect communica-
tion and mispronunciation of words or longer stretches can cause a little strain (see
marksheet pronunciation descriptors for the 40–50 and 50–60 categories in Appendix
A: Seminar Marksheet). It may also be difficult for raters to distinguish between few
mispronounced words and some mispronounced words (the descriptors of the 50–
59 and 60–69 scales on the marksheet). Some scales contain intelligibility-based
descriptors, whereas others seem to be based on comprehensibility (see the idea of
strain in the 40–49 descriptor). This leads to a little confusion in the scales. For
example, when we compare some mispronounced words may affect communication
(50–59 scale) with mispronunciation of words or longer stretches can cause a little
strain (40–49 scale), the different scales seem incongruent because the words affect
communication seem to suggest inhibiting understanding much more than cause a
little strain. If learner output causes strain, the interlocutor can understand the output
with a little difficulty, unlike output which may affect communication. Use of the
modals can and may also seems far from clear, and one of the descriptors uses the
term natural; the use of this term was criticised by Harding (2017) and Trofimovich
and Isaacs (2012) not only because it is vague, but also because it may imply a native
speaker target.
The course induction lasted three days (04/07/18–06/07/18). Because the induction
encompassed all aspects of the course, it was clear that there would be perhaps only
a few references to pronunciation during the staff induction. If there was any topic
covered that could remotely be connected to pronunciation instruction, I made notes
of what was discussed. The same was true of staff meetings. These references to
instruction are detailed in this section.
One of the induction sessions (What Students Hear: 4/7/2018) included a segment
that was intended to raise teacher awareness of the difficulties learners face due to
their low level listening skills. The CL suggested that teachers should adjust their
speech rate (speak slower) and use visuals and body language to aid understanding
although how body language could be used was not discussed. Aside from this, there
was also a discussion concerning the optional listening materials, with some teachers
and the CL pointing out that some of these listening extracts (particular extracts taken
from BBC Radio 4) were too difficult for learners. However, this session was curtailed
because the previous session had overrun, and apart from the matters noted above,
there was no in-depth guidance provided in terms of exactly how body language
might be used to reinforce suprasegmental instruction.
The provision of listening materials was touched on in a different session of the
induction (5/7/2018). The session was centred around the stages of an EAP lesson
and the things that most EAP teachers do. One of the recommendations in the session
was that teachers should provide students with a rationale for the particular activity.
I took the opportunity to ask what the rationale was in terms of listening because it
seemed that the listening goals were unclear: This meant that teachers were unaware
of what type of questions would appear in the final exam at that beginning of the
course. I was provided with a rather unclear reiteration of the importance of providing
a rationale, but received no further guidance on what type of listening test types would
be included in the exam. It is difficult to provide a rationale for listening task types if
it is not possible to say whether these task types will be examined. Learners may be
motivated to try harder when given task types that they know will appear in the final
exam. If teachers do not know such information, this rationale cannot be provided.
5.3 Action Research Results 131
On the same day, the CL stated that there were opportunities to incorporate pronun-
ciation instruction in any lesson, even lessons focused on writing. However, there
was no clear statement on the type of instruction that could or should be incorporated
(segmental or suprasegmental) or how this would be achieved. It seems that there is
an assumption on the part of the CL that teachers know how to incorporate pronunci-
ation instruction into their lessons. This may or may not be a correct assumption, but
it is clear that there is no particular suggestion concerning which type of instruction
should take prominence. Also, during staff induction, and particularly bearing in
mind that there were four new members of staff on the course, the CL stated that it
was perfectly possible to follow exactly what was suggested in the syllabus and that
no further adaptation or addition of new materials was required.
During the seminar exam standardization session (5/7/2018), all teachers watched
videos of previous seminar practice exams. Teachers firstly examined the marks
that were given to three learners, along with a rationale provided in each category
(including pronunciation). It was noticeable that there were two examples in the
rationale of one learner receiving a lower score due to the substitution of /θ/ with /s/.
Also, there was no real explanation provided of the meaning of the words used in
the descriptors. A lack of a clear statement concerning which sounds could interfere
with understanding is apparent in the rationale, combined with the noting of I sink*
instead of I think as an error.
This focus on segmentals with a relatively low functional load continued during
one of the induction sessions when the CL raised Chinese learners’ erroneous produc-
tion of the /θ/ and /ð/ sounds, stating that it is a problem that affects communication,
and that it is something that teachers should aim to “fix”. It was also stated that it
was easy to remedy such errors over the ten-week course.
On July 20, and at regular intervals throughout the course, the CL emphasized
that he did not write the syllabus and that another member of the leadership team
was responsible for this. It seems a little unusual that the person responsible for the
implementation and administration of the course took no part in the design of the
course syllabus.
There was one CPD training session which was focused on speaking. In particular,
the focus was on the importance of learner presentations in fostering oracy skills.
The training session leader expressed a great deal of surprise that learner presen-
tations were only assessed formatively, particularly in view of the fact that many
undergraduate and postgraduate courses require learners to give presentations. This
is particularly true of business-oriented courses, which are the destination courses
for many of the learners on the pre-sessional.
132 5 Results, Analysis, and Discussion
The record I kept of my own teaching over the course shows that, in my own opinion,
the syllabus sometimes suggested using materials that were unsuitable, either because
the language used was dense or complex, or because the topics for discussion were
not engaging. In order to compensate for the shortcomings of the syllabus and the
suggested materials, I did try to use alternative materials. For example, I devised
my own seminar worksheet which incorporated a focus on word stress, based on
a topic that I felt was more engaging that had simple, direct questions. I also used
parts of the coursebook (Hewings, 2012) that were not included in the syllabus.
There were sections on word stress in two-word combinations, chunking, and a
more engaging discussion section on cultural differences. Furthermore, I devised a
chunking exercise that followed the advice of Brazil (1994, p. 3): to concentrate on
the content and meaning of the language before chunking. I used an exercise from
Patsko (2014), and found it easy to implement and a useful introduction to chunking
for learners. Although I was able to use these materials on some occasions, time
pressure meant that I was not always able to implement a viable alternative to the
syllabus materials.
Some of the listening materials also proved less than ideal due to a number of
factors. Primarily, the vast majority of listening extracts involved either SSBE or SAE
varieties. Once again, I did devise one listening exercise myself, about Chinese invest-
ment in the local area, which involved a standard Northern English accent, a Chinese
speaker of English, and examples of much more colloquial Northern English. Not
only did this extract (taken from the regional TV news) feature more diverse accents,
it also sparked a good deal of curiosity because many of my students’ destination
courses were business-related. I felt it also interested them to hear another person
from their own country who had settled in the local area, could speak English well,
and had become successful. However, devising these activities is time consuming,
and I was unable to do this on more than one occasion due to time pressure. Also,
because learners (and teachers) were unaware of the question types in the exam, it
was difficult to know which type of listening activities to focus on.
The lack of clear listening goals seemed more apparent when, halfway through the
course, it was decided that a pilot listening test that was entirely optional for learners
would be introduced. The listening pilot yielded extremely low scores, primarily
due to a new note-taking section where students had to take notes on a listening
extract before they were given the question sheet. This test was radically different
in nature to the final exam, and served to increase the anxiety of many learners due
to the low scores received. It was later revealed that the listening test was intended
to be a pilot test for the autumn pre-sessional course. Nonetheless, it is difficult for
listening elements of the course to support pronunciation skills when the goals of
those elements are unclear.
In terms of my own positive experiences in attempting to provide instruction, I
was able to implement some of the recommendations of Baker and Burri (2016). In
5.3 Action Research Results 133
particular, I found the system of recording the number of syllables and the location
of primary stress easy to implement and helpful in fostering consciousness-raising.
I found it much more successful than simply reciting the correct pronunciation and
getting learners to repeat this in whole class mode. Learners recording their own
production also assisted this consciousness-raising. Also, bearing in mind Baker and
Burri’s (2016) recommendations, I believe that I used much more non-verbal commu-
nication to signal pronunciation features, such as prominent stress and intonation,
and that this provided scaffolding to learners which assisted uptake.
A less successful element was the language sheets (for example, language
requesting clarification) that were distributed to learners to assist them in seminar
discussions. It is perhaps unrealistic to expect learners to immediately adopt this
language and use it in free discussion. In hindsight, a role-play stage would have
assisted learners in using this language. Once again though, this would have required
extra planning. Additional guidance on how to facilitate the assimilation of this
language into learners’ discourse would have been useful.
Finally, I noted that the presentation element of speaking instruction, which does
contain some important suprasegmental pronunciation elements, such as chunking
and intonation, was not taken as seriously by students as it was in previous years. This
may simply be because the speaking learner group I taught was less well motivated.
However, it could also be connected with the fact that learner presentations are no
longer the subject of summative assessment.
Teachers routinely double mark one of the seminar speaking exams with their partner
teacher (particularly if one of the groups consists of four learners). After one of
these sessions, the only non-native speaker teacher on the course informed me of
a conversation that occurred with the double marker. There was a disagreement
about whether one of the learners was intelligible. The native speaker double marker
asserted that the learner would not be understood, citing native speaker status as a
reason why this judgement was more persuasive. Although not directly connected
with instruction itself, this particular anecdote suggests that the view of the supremacy
of the native speaker is still apparent, at least with this particular teacher.
Over the duration of the course, learners are given a learning journal to fill in
periodically. The intention is to encourage learners to reflect on the work they have
completed and to plan the next stages of their learning. Experiences involving pronun-
ciation learning can be included in this journal. However, during the current research,
as in other years, I found that, despite my attempts to encourage and persuade learners
to complete their journal and keep it up to date, very few learners did so (usually
those of a higher level). Discussions with colleagues revealed similar experiences.
The majority of learners on the pre-sessional are from a Chinese educational back-
ground where ideas of self-reflection and independent learning are not as widely
134 5 Results, Analysis, and Discussion
accepted as in the UK, and perhaps it is this that is the obstacle to the learning
journal being more successful.
Also, throughout the course, there are skills for life sessions (organized by non-
teaching staff) designed to help learners assimilate to their new environment. One
element of these sessions is a conversation club where learners have the opportunity
to meet and converse with both native speakers and non-native speakers. I believe
only one conversation club session took place over the length of the course although
the initiative is one that continues on a regular basis once learners have been accepted
on their degree courses. Perhaps more of the conversation club activities during the
course would be beneficial.
The next section details areas related to pronunciation instruction that are evident
in the four observed lessons (myself, Mark, Bruce, and Olivia). The results are
presented in pairs because these pairs of teachers taught the same lesson on the same
day. Other areas of interest that might have a peripheral connection to pronunciation
instruction are noted in the second section of each pair. Finally, the checklist data of
all participants are investigated.
Both lessons begin in the same way: following the syllabus and introducing the
new vocabulary. The main difference is that Mark elicits the meaning of the words,
whereas I also elicit the word stress by saying the word, encouraging answers, and
writing the word on the board using the Baker and Burri (2016) system. When
learners are attempting to guess the stress, they say the words aloud. I also tend to
use hand movements to try to show more clearly where rises in intonation occur
and where the main stress is in the new vocabulary. The second stage of the lesson
is an identical listening for detail task involving a video with SAE English. One
difference in implementation is that I ask the students to try to predict the answers
before watching. The introduction of the discussion language worksheets is broadly
similar although there are one or two minor differences (for example, I attempt
to explain what intonation is). Also, the two seminar tasks are almost identical,
the major differences being that Mark allowed two minutes preparation before the
discussion, whereas I did not (learners were already aware of the discussion questions
as they had been given these in the previous week). Another small difference was
that Mark seemed to participate a little more with the particular groups during the
5.4 Results: Observations and Teacher Self-completion Checklist Data 135
seminar discussion, whereas I tended to monitor from a distance, only assisting with
language or pronunciation rather than contributing myself.
There are then two different listening exercises that are not part of the scheme of
work. Mark’s listening is a video consisting of a variety of UK accents, whereas my
own listening is a SAE video. Both extracts were listening for detail although my
own was also followed by a chunking exercise. There are two additional elements in
Mark’s lesson: a short (1 min) SAE video listening for detail exercise, which is also
not part of the syllabus, and a recap of earlier work where students attempt to recite
a famous quotation, stressing the correct words in the quotations.
Throughout the lessons, there are few examples of error correction. In my own
class, correction mainly consisted of asking students to repeat the word once again
in whole class mode if I did not think the pronunciation was sufficiently intelligible.
This happened on four occasions. In Mark’s class, there are two examples where a
learner says something inaccurately and Mark models the correct pronunciation of
the word approximately. The learner then repeats the word more accurately. There
is another example of the word floods that is pronounced incorrectly. Again, Mark
models the correct pronunciation, but the learner does not repeat the word on this
occasion.
Both Mark and myself stress the importance of using the discussion language at
various points throughout the lesson. However, in both lessons, the extent to which
this language was used was negligible although it must be stated that it is difficult to
hear the audio of the students’ discussion during the free seminar discussion mode.
Learners change places for different seminar tasks in both classes to enable them
to communicate with other learners. Indeed, this is true of all the observed classes.
Mark has three seminar discussion phases (5 min/10 min/10 min), whereas I have
two seminar discussions of 10 min duration each. Although the task duration is not
the same as that in the exam (15 min), I assume that Mark, like myself, gradually
increased the seminar task length as the course progressed towards the final exam.
Of Mark’s three seminar phases, only one of them is followed by feedback in terms
of the content of the discussion (i.e., reporting back in whole class mode). In my
own class, the first seminar is followed by feedback from the seminar groups, but
the second one is not.
Another point of interest is that the video extracts used sometimes consist of
videos where learners cannot see the person who is speaking. This may well make
the listening exercise a little more difficult for the students. A summary of the content
of my own and Mark’s lesson can be seen in Table 5.1.
136 5 Results, Analysis, and Discussion
Table 5.1 A summary of the content of Mark’s/my own video recorded lessons on 1/08/2018
Stages of the lesson Mark Myself
Introduction to new Elicits word meanings of target In whole class mode, says the
vocabulary vocabulary in whole class word, elicits the stress and
mode. Students do not repeat or notes it on the board with
say the target vocabulary almost all of the new
vocabulary. Students repeat the
word. Elicits word meanings of
target vocabulary
Listening exercise: video. Describes task, shows the video Describes task, shows the
Listening for detail (SAE) and then elicits answers video and then elicits answers.
Writes the stress pattern of the
new word ‘malnourished’
Language for chairing and Presents discussion language, Explains what intonation is and
Concluding worksheets models the language, and then relates it to their own language.
students repeat the language in Presents discussion language,
unison models the language, and then
students repeat the language in
unison. Elicits sentence stress
Free Seminar Discussion Sets task and monitors. 2 min Sets task and monitors.
(groups of three or four planning before the task. Reminds students to use
students) During monitoring, asks one discussion language towards
student to speak louder and the end. Points to one group
reminds them to conclude the where one student said very
discussion at the end. Asks little and revised earlier work
students if they used the on language for interrupting
discussion language. They say politely. Gets feedback in
yes whole class mode
No feedback on the content of
the discussion afterwards
Free Seminar Discussion Sets task and monitors. 2 min Sets task and monitors. During
(groups of three or four planning before the task. monitoring, points out the
students) During monitoring, participates falling intonation of the last
in some discussion with the item in a list and uses hand
groups. Gets feedback in whole gestures to show the fall. No
class mode. Emphasizes that feedback on the content of the
students must use the discussion afterwards
discussion language. Asks
students if they used the
discussion language. They say
yes
Listening exercise: listening Warmer questions (unusual
for detail, followed by sports), followed by listening
chunking task for detail questions and
feedback, before proceeding to
the short chunking exercise
(continued)
5.4 Results: Observations and Teacher Self-completion Checklist Data 137
When new vocabulary is introduced in the early stages of both lessons, Bruce does
not model the pronunciation, and the learners do not produce the new vocabulary.
Only the meaning is elicited. This is also true in Olivia’s class, apart from the two
vocabulary items: secular and sarcasm, which Olivia says and the class repeats in
unison. Olivia explains the word sarcasm and does provide an example, but does
not produce the intonation that would accompany sarcastic language. It should be
remembered that Olivia’s group is weaker than Bruce’s, so perhaps it was felt that
it was too difficult for the learners to grasp. There is one video exercise that is used
in both classes and is part of the syllabus. However, this particular exercise does not
seem to support pronunciation instruction in any way, not least because there is no
audio. It is rather intended as a warmer for the discussion phase. Learners read the
information from the clip and view the images. Bruce uses a second video (SAE),
which is a listening for detail exercise.
Specific, targeted pronunciation instruction is only really attempted by Olivia,
who, when dealing with language used to assist turn-taking, models the expres-
sions, which are then repeated by the learners in unison. In particular, Olivia draws
138 5 Results, Analysis, and Discussion
learners’ attention to the rising intonation in many of the expressions used, and makes
learners repeat the expressions if the intonation is not correct. Learners seemed to
articulate these expressions well when repeating them. However, as in Mark’s and
my own class, it is far from clear that learners will be able to use them spontaneously
in a conversation. There is certainly little evidence of this in the language that is
discernible in the seminar free discussion phases of any of the observed classes.
In terms of error correction, there are few, if any, discernible examples of pronun-
ciation error correction in either lesson. There is one point in Olivia’s lesson when a
learner says the word mutton a little unclearly, and the teacher repeats the word so
that the learner is aware of the pronunciation. Aside from this example (and Olivia’s
intonation drills), there are no noticeable pronunciation error corrections in either
lesson. This may be because neither teacher believed their students’ spoken language
to be unintelligible. In particular, this may well be true of Bruce’s class because of
the learners’ more advanced language skills (based on the learners’ IELTS scores).
The importance of using discussion language is emphasized in Olivia’s class.
Bruce does not introduce the discussion language worksheet in class, eschewing the
intonation elements present in the taking your turn worksheet, but focuses very much
on what the students should be doing during the seminar, initiating a pre-seminar
task where learners discuss this. It may well be the case that the specific language
worksheet was given earlier as homework, or perhaps a more macro approach is
taken, either because the learners are of a higher level than those in Olivia’s class or
because, as was observed in Mark’s and my own class, it seems difficult to transfer
the worksheet language so that learners actually use it during the seminar discussion.
There is a slightly different approach to the seminar discussion tasks. Olivia includes
more tasks although these are usually shorter, whereas Bruce employs fewer, longer
tasks. Perhaps this is because Bruce’s class are a higher level, so it is easier for
the learners to converse for a longer period of time. Also, there appears to be a
different emphasis in terms of monitoring and feedback after the seminar tasks.
Bruce participates much more whilst monitoring, whereas Olivia tends to intervene
rarely. Feedback after seminar tasks differs in that Bruce has a reporting back stage
where seminar members reveal their findings to the whole class, whereas Olivia
usually has no reporting back stage and simply moves on to the next task (see Table
5.2).
The checklist data consist of three checklists completed by Mark, myself, and Bruce,
and two completed by Olivia. For each of the four teachers, there is one checklist that
5.4 Results: Observations and Teacher Self-completion Checklist Data 139
Table 5.2 A summary of the content of Bruce’s/Olivia’s video recorded lesson on 25/07/2018
Stages of the lesson Bruce Olivia
Short warmer speaking activity: Learners discuss their favourite
groups of three food and the last thing they ate
Introduction to new vocabulary Elicits word meanings. Of the eight vocabulary items,
Learners do not say the words the teacher models two words
that the learners then repeat in
unison. Learners do not say the
other words. Elicits word
meanings
Video exercise: there is no audio. Whole class feedback follows A short two minute discussion of
Reading/visuals only the task what the learners know about
India and whole class feedback
precedes the task. Brief whole
class feedback follows the video
Listening exercise. Listening for A short discussion of the
detail: video (SAE) listening in groups of three
follows the task, which is them
followed by feedback in whole
class mode
Free Seminar Discussion (groups Elicits what learners should do During monitoring, occasionally
of three or four students) during the task (e.g., listen to participates in some discussion
others). During monitoring, with the groups. No whole group
participates in some discussion feedback when the task is
with the groups. Gets feedback complete
in whole class mode
Language for discussion: taking Presents discussion language,
your turn models the language, and then
students repeat the language in
unison. Points out and models
the intonation of questions
Free Seminar Discussion (groups During monitoring, occasionally
of three or four students) participates in some discussion
with the groups. No whole group
feedback when the task is
complete, apart from the teacher
commenting that the discussion
was better
A short free Seminar Discussion During monitoring, occasionally
(groups of three or four students) participates in some discussion
with the groups. No whole group
feedback when the task is
complete
Free Seminar Discussion (groups During monitoring, occasionally
of three or four students) participates in some discussion
with the groups. No whole group
feedback when the task is
complete
(continued)
140 5 Results, Analysis, and Discussion
was completed after the video recorded observation to enable comparison: Mark and
myself on 01/08/2018, and Bruce and Olivia on 25/07/2018. However, as has already
been stated, the video recorded observations contain seminar discussion elements
where it is difficult to discern what the teachers and learners are saying because
students are interacting simultaneously. In addition to the four teachers (including
myself), two other teachers completed one checklist each, giving a total of thirteen
checklists (details of the number of checklists and when they were filled in can be
found in Table 5.3). The two additional teachers, who filled in one checklist each, are
referred to by the pseudonyms Alice and June. This section begins with an analysis
of types of error correction from the sum total of all the checklists and the particular
segmentals corrected, followed by an examination of the type of error correction
used by Mark, myself, Bruce, and Olivia. Finally, a comparison is made between
the observation and the checklist that each of the four teachers completed after the
observation.
From all checklist data, the most commonly stated correction occurred with into-
nation, with correction declared in 10 of the 13 checklists. Next comes sentence
stress: 7 times, word stress: 5 times, and vowel and consonant correction: 4 times
each. In terms of the sum total number of corrections, intonation is the most common
(21), followed by word stress (16), sentence stress (10), consonants (9), and vowels
(7). There is therefore a total declared correction of 16 segmentals, compared to 47
suprasegmentals. Of the segmentals corrected, specific examples were only given on
6 occasions: 5 consonant corrections and 1 vowel correction. These were /p/ and /b/
problems, mentioned twice, once by Bruce, who also notes that this problem was
apparent with Arabic speakers. Problems with /t/ and /Ѐ/ were also noted, as was
5.4 Results: Observations and Teacher Self-completion Checklist Data 141
confusion between /d/ and /t/. The erroneous production of /θ/ was also declared
on one occasion. In terms of vowel correction, this occurred because of confusion
between /i / and /I/. The teachers who did estimate the success of their interventions
corrections occurred during the seminar phases of the lesson, but were not picked
up on the audio. She also states that intonation was corrected on one occasion. In
fact, Olivia’s observed lesson does contain a section where an intonation feature is
introduced in a listen and repeat drill, and this occurs on more than one occasion, so
it appears that Olivia may have actually underestimated the number of corrections
made. Alternatively, she may not have viewed these listen and repeat exercises as
corrections. My own view is that the drilling she employs does sometimes involve
error correction because the drilling is repeated if she is dissatisfied with the language
that the students produce. She also does explain that the questions “go up at the end”
when she asks learners to repeat them once again. This advice is accompanied by a
gesture she uses with her hands and arms to further reinforce the upward intonation.
In this respect, this type of drilling seems rather different to that of Mark’s. A sentence
stress correction is also reported, but is difficult to observe in the video recording. It
is also reported that learners recorded their own output during the observed lesson,
but this was not visible, so perhaps this was simply an error.
Bruce reports three vowel and three consonant corrections in the video recorded
lesson. None of these corrections are audible, and once again, may well have been
made during the seminar phases of the lesson. The word stress and intonation correc-
tions reported were also not audible. Although these corrections may have been made
during the seminar phase whilst monitoring (particularly as Bruce was much more of
a participant during monitoring than other teacher participants), it is certainly clear
that no pronunciation instruction takes place in Bruce’s class during whole class
mode. This could be due to the teacher considering pronunciation instruction less
important because of the higher level of his class. This is explored further in the
semi-structured interviews.
The following section presents the results of the seminar assessment phase. The
variation in pronunciation marking is compared to the other seminar marking cate-
gories. This is then followed by an examination of the particular descriptors chosen
by the teachers. The section concludes with the results from the post-marking short
interview.
Table 5.4 shows the scores given by Mark, Bruce and Olivia to the three students
appearing on the left, the middle, and the right of the screen (L, M, and R). It can
be seen that there is quite a wide variation in the marks given, with the greatest
5.5 Teacher Participants’ Assessment of Student Pronunciation 143
Table 5.4 Scores given by the three teacher participants to the three candidates (L, M, and R) in
the pronunciation category
Candidate Mark Bruce Olivia
L 50 53 45
M 60 60 45
R 50 60 45
Source Author’s own
Table 5.5 Variation in marks given by the three teacher participants to the three candidates (L, M,
and R) in the remaining categories
Marking category Marks given Variation in marks
Grammar L:53/45/45 M: 55/53/45 R: 45/50/45 10
Vocabulary L: 50/45/50 M: 55/55/50 R: 50/45/45 5
Interactive communication L: 47/55/53 M: 57/63/53 R: 45/47/45 10
Source Author’s own
discrepancy in the marks given for candidates M and R. The difference in the marks
awarded by Bruce and Olivia for these two candidates is 15 in each instance, with
Bruce awarding both candidates 60, and Olivia awarding the candidates only 45. In
terms of IELTS equivalents, a mark of 45 represents a score of 6, and a mark of
60 represents a score of 7. In terms of the overall speaking assessment score, such
a discrepancy would represent a difference of 3.75. This is quite a large variation
to appear from one category. Although there are variations in marking in the other
assessment categories (grammar, vocabulary, and interactive communication), there
is no variation in marks as great as that in the pronunciation category.
In order to illustrate this point, Table 5.5 shows the variation in the marks given
in the grammar, vocabulary, and interactive communication categories. It should be
pointed out that the variation of 15 marks occurs with two candidates in the category
of pronunciation, yet a variation of 10 marks only occurs with one candidate in the
grammar and interactive communication categories, the variation being much lower
in the vocabulary category. Table 5.6 shows the particular pronunciation descriptors
chosen by the teachers for the three candidates. This illustrates how the teachers
arrived at the score that was awarded to the candidates.
Seminar grading follow-up short rationale interviews yielded little further infor-
mation. Mark mentions the difficulty of distinguishing between pronunciation and
grammar errors. For example, he mentions one candidate saying woman instead of
144 5 Results, Analysis, and Discussion
Table 5.6 Pronunciation descriptors marked by the three teacher participants for the three
candidates (L, M, and R)
Candidate Teacher participants
Mark Bruce Olivia
L • Inconsistency or • Tempo, rhythm and • Inconsistency or
inaccuracy in tempo, intonation usually inaccuracy in tempo,
rhythm and intonation consistent but rhythm and intonation
cause a little strain communication may be cause a little strain
affected by inaccuracy
• Some mispronounced • Some mispronounced • Mispronunciation of
words may affect words may affect words or longer
communication communication stretches can cause a
little strain
M • Tempo, rhythm and • Tempo, rhythm and • Inconsistency or
intonation generally intonation usually inaccuracy in tempo,
consistent and consistent but rhythm and intonation
communication not communication may be cause a little strain
affected affected by inaccuracy
• Some mispronounced • Few mispronounced • Mispronunciation of
words may affect words but not affecting words or longer
communication communication stretches can cause a
little strain
R • Inconsistency or • Tempo, rhythm and • Inconsistency or
inaccuracy in tempo, intonation usually inaccuracy in tempo,
rhythm and intonation consistent but rhythm and intonation
cause a little strain communication may be cause a little strain
affected by inaccuracy
• Some mispronounced • Few mispronounced • Mispronunciation of
words may affect words but not affecting words or longer
communication communication stretches can cause a
little strain
Source Author’s own based on the marksheet descriptors chosen
category. So it seems that, for Bruce, the amount of language produced, if insuffi-
cient, should affect the interactive communication mark, not the pronunciation mark.
This is a difficult area because although the logic of this interpretation appears to be
sound, there could be circumstances where the candidate says very little, but with
good pronunciation. It might seem unfair that another candidate takes more risks,
says much more, and receives the same pronunciation grade, even if they receive
better marks in the interactive communication category.
The results for this particular phase of the research are divided into three sections.
The first section describes the general results, involving the goals of instruction,
segmentals and suprasegmentals, teacher guidance, and materials. The second section
addresses the results in terms of assessment, and the final section provides details
of a shorter follow-up interview that was conducted specifically to elicit the CL’s
opinion on features of lessons that I had noticed during the teacher observations.
The views of the CL seem somewhat ambivalent in terms of the goal of pronunciation
instruction. In several statements during the interview, he states quite firmly that
intelligibility is the goal of instruction and that learners should be intelligible to
other students in the learner group on the pre-sessional. This is quite significant
as most learner groups on the course in 2018 (and also in previous years) had the
same L1, consisting of Chinese speakers. If Chinese learners are intelligible to other
Chinese learners, this does not necessarily mean they will be intelligible to native-
speaker learners or other non-native speaker learners that they will likely interact
with on their destination courses.
Interviewer: So do you want them to be intelligible or comprehensible, or…what are we
striving towards? Are you thinking about native speaker-like.
Course Leader: No (definitive). They don’t need that to get onto those courses anyway.
They’re not going to be, not from an IELTS 5, possibly 5.5 when they arrive here, they’ll be
nowhere near native speaker level.
Interviewer: So it’s more about intelligibility…intelligible to (pause).
Course Leader: To the people in the group, which will hopefully make them more intelligible
in the future, to students on their course.
On the other hand, when discussing whether the substitution of the /θ/ or /ð/ sound
for /s/ or /z/ should be corrected, which would seem to suggest a native speaker target
of instruction, the view seems less clear. At some points during the interview, there is
146 5 Results, Analysis, and Discussion
Another area of interest is that a justification for correcting such sounds is that it is
easy, even though it is acknowledged that it may not be effective at all (“but they never
change”). Indeed, there does seem to be some doubt as to whether these corrections
of segmentals during the course, and pronunciation instruction in general, will prove
to be effective over the duration of the course. This raises an important question: If
it is not possible to improve learner pronunciation over the duration of the course
in one or more aspects that are the subject of summative assessment, why should
those aspects be assessed at all? Assessment will merely reflect the learner’s ability
at the beginning of the course. From the following exchanges, as well as the previous
exchange, it seems evident that, apart from intonation and general fluency, there are
generally few improvements in learner use of suprasegmentals, and no noticeable
improvement in the pronunciation of segmentals. It is therefore difficult to understand
why 50% of the descriptors in the pronunciation category on the marksheet focus on
segmentals.
Interviewer: But you don’t think that will change by the end (of the course).
Course Leader: It might not do no, but it’s something that they bear in mind-the students.
You can remind them of these things, I mean, whether you can make them change is another
thing, isn’t it?
Interviewer: What are the things that we can change in ten weeks?
Course Leader: We can improve, not their pronunciation, but their confidence to speak-
that’s not pronunciation. We can, as I say, rising intonation at the end of a sentence. We
can get them thinking about that, if they don’t already know about that, but in reality, when
someone is speaking, I’m not sure they’re probably thinking like that. They’re thinking about
what’s the next word and how do I join together a sentence to express what I want to say.
There’s too much to think about, and that probably comes in time.
Course Leader: It’s…it’s an error, isn’t it. And it’s something…it’s not standard English, but
I think (stress on I), you know, if, within a sentence, you understand what they are saying…
it depends, doesn’t it? If within a sentence, there are no other errors… It depends on the
amount of errors within one utterance, doesn’t it? And whether that one error is affecting
comprehensibility or not. If there are loads of errors, then it will affect comprehensibility,
but one little thing like that… I mean ‘sing*’ is more interfering than ‘fing*’, isn’t it? ‘Sing*’
is a different word, so it might be confusing for people.
This exchange is interesting for a number of reasons. Firstly, the initial idea that
it is an error seems to give way to a view that although it is “not standard”, it should
not be regarded as an error if it appears in isolation. Also, it is far from clear that
the idea that a substitution with /s/ interferes more with comprehensibility than a
substitution with /f/ carries any validity. One could assume that it would be less
comprehensible to native speakers, which would imply that the target of instruction
is to appropriate a native speaker-like pronunciation. In any case, the argument that it
is less comprehensible, presumably because certain UK accents do substitute /θ/ for
/f/ (e.g., Estuary English), is not entirely convincing because many native speakers
are also aware of accents that do substitute the sound by using /s/, and listener
familiarity with a particular variety of English can make it easier for native speakers
to understand that variety (Browne & Fulcher, 2017, p. 39; Gass & Varonis, 1984).
Films and television programmes in the UK have long featured numerous examples
of French or German speakers using the /s/ or /z/ instead of the /θ/ or /ð/: Peter Sellers,
in The Pink Panther series of films, and Poirot immediately spring to mind. Indeed,
the stereotype of a French speaker accent is so well established in British society that
it would seem unlikely that this substitution is less comprehensible. In another part
of the interview, the CL seems to acknowledge this.
Course Leader: You wouldn’t get a French person to change…when you hear a French
person, it sounds charming when they are using ze* French accent…and it sounds great, but
you wouldn’t want them to change, and it’s not necessary-it may not be necessary to change.
Also, the idea that the /s/ substitution could have more of a serious effect on
comprehensibility because it would lead to the formation of a different word does
not take into account the crucial importance of context. If the conversation was not
about singing, it would be highly unlikely for the interlocutor to assume that the
speaker meant sing instead of thing.
In terms of the materials specified in the syllabus, the CL mentions pronunciation
being “touched upon” throughout the course, refers to the language focus worksheets
that are present in almost every week of the presentation syllabus, and argues that
building on this language throughout the course can be an effective way of improving
intonation and sentence stress. Also, although it is acknowledged that it may be easier
for teachers to focus on single sounds due to their salience (Levis, 2018, p. 29), he
is unsure of the effectiveness of this approach and favours the more suprasegmental
focus of the syllabus.
Course Leader: It’s easy to focus on individual sounds. It’s very easy, isn’t it? But, yeah,
whether it’s effective, I don’t know, unless it’s actually interfering with the pronunciation
of the word. It’s…I think in the seminar sessions, we do, there are those sheets aren’t there,
148 5 Results, Analysis, and Discussion
that we do each week: language focus, which do focus on intonation and stress: intonation
and sentence stress. I think all of it, it’s kind of touched on throughout the course, and I
assume what the tutors are doing…in week one this is what you are focusing on. Then, from
week one onwards, the tutor focuses on that, and that’s left on the wall for students to ponder
and think about. And they get a copy anyway, don’t they. And it’s just something you keep
building into the course, “remember this we did three weeks ago about the rising tone at the
end of a yes/no question?”
Another interesting aspect of this exchange is that the rising intonation of polar
questions is mentioned. This has been shown to be far from clear as the intonation of
polar questions can be quite variable (Fries, 1964; Lee, 1980; Levis, 1999a). When
asked for his personal opinion on which particular element of pronunciation was the
most important, he chose sentence stress.
Course Leader: Sentence stress is probably quite an important thing to focus on because
they are going to be speaking in sentences and it…I don’t know. Sentence stress possibly.
The main listening goal is identified as being able to take notes (which is also
identified as the main goal in the Student Handbook). According to the CL, there is
no listening designed to improve pronunciation skills on the course although he does
add the caveat that the listening skills section of the book “touches on pronunciation”.
The coursebook (Hewings, 2012) does have four lecture skills sections that were used
in 2017. Use of these lecture skills sections is optional, but it is highly unlikely that
any teachers used these sections as they were one of the major sources of complaint
raised by teachers in previous years; most teachers expressed the view that these
listening extracts were far too difficult for their learner groups. It is also possible
that the CL was referring to other specific pronunciation sections on word stress
and intonation in the coursebook. However, in 2018, the number of these sections
included in the syllabus was substantially reduced.
Interviewer: Again, about the goal: is there some kind of pronunciation goal linked in with
the listening or not? I mean, what are the main goals in listening? I’m thinking about, you
know, from the start of the course, what do we want them to be able to do by the end? I
mean…
Course Leader: Without having the listening learning objectives in front of me, I can’t
exactly remember what they are. To be able to listen to input from lectures, for example. To
be able to take notes. Listening and writing together. You either listen and speak or listen
and take notes.
Interviewer: What about listening to improve pronunciation? Is that kind of in there or not?
Course Leader: In the course? No. That’s something that the tutor will touch on as you go
through the course. The listening skills section of the book, I mean, that always touches on
pronunciation issues…it’s whether it’s done by the tutor. I mean, there’s always something
on pronunciation.
Finally, I believed it could be of interest to ask about whether teachers have enough
guidance in terms of pronunciation instruction. The CL points to the guidance in the
teacher’s book, but concedes that much of the pronunciation material provided by
Powell (1996) is accompanied by little or no guidance. It is also pointed out that
guidance could be provided by the teacher’s teaching partner or be assisted in some
5.6 Semi-structured Interview with the Course Leader 149
Also, when reflecting on teacher confidence and identifying and rectifying errors,
it is acknowledged that there could be a good deal of variability in terms of how
confident teachers are when approaching pronunciation instruction.
Course Leader: To identify it is (easy), but to fix it is difficult, but it’s easy to identify and
focus on that, but yeah, the larger suprasegmentals…er, I don’t know, I think it depends on
the individual…some people tend to like…they might feel very confident with pronunciation
and happily just weave it into any session; other teachers may not do.
Course Leader: Yeah, it is. It’s marked as 50-50. We don’t decide that… I don’t decide that
personally, but that’s what it is. Whether it should be 50-50, I don’t know.
It is also acknowledged that only having two descriptors could lead to difficulties
in marking because all of the various suprasegmental elements are contained within
only one descriptor. He confirms that the marksheet did have more than one descriptor
for the suprasegmental category in the past. In the same exchange, the CL also states
that it is easier to mark down the segmental category because “you can hear how the
sounds interfere.” If it is true that it is easier to mark down segmental errors because
they are more noticeable (mentioned by Levis, 2018, p. 29), it may mean that some
teachers could concentrate on these aspects of pronunciation a little more than the
suprasegmental section.
Course Leader: Yeah…so it is easier to mark down because you can hear how the sounds
interfere, but perhaps we need to remind tutors more what this top section means, that it is
related to all these other things, but it seems there’s a lot in the top section, but very little
in that last section because that’s just about sounds, it’s easy, but this covers every other
aspect of pronunciation, which is quite a lot really. But it used to be bigger. It used to be a
lot bigger…
Interviewer: Maybe it should be a bit bigger then.
Course Leader: Well maybe. It used to be intonation, stress…
Interviewer: Maybe it could be two to one: two suprasegmental ones and one segmental
one.
Course Leader: Yeah.
However, the CL does not seem entirely convinced that more importance should
be afforded to suprasegmentals on the course.
Interviewer: So let me recap a little. In the marking scheme, there’s a kind of equal focus
between single sounds and other sounds.
Course Leader: Yeah.
Interviewer: And you think that should be the case? Or should there be more stress on,
focus on intonation, rhythm, all of those things…sentence stress?
Course Leader: I don’t know because (if) you pronounce some words incorrectly, you can
use all the correct intonation and sentence stress, but if you were pronouncing every word
incorrectly, then there won’t be any communication.
One of the important topics addressed was the guidance given to tutors in terms of
marking the pronunciation categories, particularly in terms of discriminating between
segmental errors that may interfere with intelligibility and those that generally do
not. When discussing the guidance provided to teachers in this area, reference was
made to the document all teachers were given during induction: seminar guidelines.
This document was examined in the document analysis Sect. 5.2.7.
Course Leader: Well, you would hope in the standardization process, that it says. I’m not
sure what the guidelines…you know, the guidelines that go along with this when we do the
standardization in the induction, it says. …I’m not sure, I can’t remember now whether it says
when there are these individual sounds that don’t really interfere that much, they shouldn’t
be included…it depends how many there are, doesn’t it? If there are multiple sounds…if
5.6 Semi-structured Interview with the Course Leader 151
one phoneme is pronounced incorrectly. If there are ten of them in one utterance, then it will
interfere…
Course Leader: It depends, it all depends on the amount of errors, doesn’t it…whether
something is comprehensible or not, I’d have thought.
Firstly, the point made regarding error frequency does not generally apply to lower
functional load errors. These errors (if classified as such) do not impact comprehen-
sibility, even if the number of errors increases (Munro & Derwing, 2006). Secondly,
in terms of the guidance document, as was explained in Sect. 5.2.7, the seminar
guidance document does not contain any guidance at all in terms of which segmental
errors could interfere with intelligibility. The guidance given merely consists of
the descriptors from the categories themselves and definitions of one or two of the
language items used in the descriptors, not which type of segmental errors may be
important for intelligibility. The CL also points out the importance of standardization
at another point in the interview.
Interviewer: There’s this…you know, the guidance you gave us in induction on the…stan-
dardization. Is that…I’m just thinking where teachers know where to go from that sort of,
you know, whether the /θ/ sound is maybe not that important? You know, do they all have
the same…
Course Leader: Yeah. Maybe one person would consider that an awful error and another
person not. You know, they need to be standardized, don’t they…you know, are these errors
that interfere or not?
I was able to raise one of the potentially problematic areas when marking pronun-
ciation: the confusion between the pronunciation and the interactive communication
categories. I was interested in discovering how the CL interpreted the two categories,
and whether he thought there was any danger in teachers confusing the two cate-
gories. His response was that there should not really be any confusion between the
two categories.
Course Leader: Yeah, well, I guess it’s (interactive communication) more about ideas, isn’t
it, and contributions. And this (pronunciation) is about how you say it. That’s about what you
say. What ideas you bring to the table and how well you discuss it. Do you use interactive
strategies, and that’s (pronunciation) how you say it, and whether any of the errors that you
possess affect your communication. I think interactive communication is more about the
ideas. So I don’t think it crosses over much. I don’t think it crosses over at all really.
After reviewing the video recorded lessons, bearing in mind the different approaches
that teachers took towards introducing new vocabulary (i.e., drilling was sometimes
employed and sometimes not), I took the opportunity to present the CL with particular
vocabulary items from Mark’s lesson and my own lesson, and from Bruce’s and
Olivia’s lesson. I wanted to discover what the CL’s expectations were, and particularly
whether he thought that a presentation and oral practice of these items was needed.
152 5 Results, Analysis, and Discussion
Although the vocabulary items in the session that Mark and I taught do seem
to be more academic in nature, it is unclear why less academic words should not
be spoken by learners when they are introduced. The words may still be used, and
are intended to be used during the seminar discussion phase, so whether they are
academic or not seems largely irrelevant. However, it should also be noted that, after
initially suggesting that learners saying the words might not be necessary, the CL
also states that “it’s always useful to get them to say words anyway.”
When speaking about the new vocabulary from Mark’s/my own lesson, the CL
states that he would deal with the word stress of the vocabulary items.
Interviewer: Would you do the word stress with those or not (show words from Mark’s/my
own lesson)?
Course Leader: Yeah. I mean I’d get someone to say it, yeah…or I’d say it and ask them
where’s the word stress, just quickly, have it written on the board. Have it on the visualiser,
just note the word stress. I mean, it’s always good to remind them all the time about that sort
of stuff.
Also, when discussing the focus on language worksheet that appears at a later stage
of the Bruce/Olivia lesson (taking your turn), the CL states that he would model the
intonation (as Olivia did). In the same exchange, he confirms that the words from
the earlier stage of the same lesson (ban, secular, Hindus, revere, vegetarian, vegan,
sin, and sarcasm) need not be said by the learners.
Interviewer: But would you model the intonation there or not?
Course Leader: Yeah. Pop in the questioning intonation at the end “can I say something”
(rising intonation on ‘something’).
Interviewer: And that one as well yeah?
Course Leader: Yeah, because it’s got the arrows going up and down. I mean, it’s good to
do that, to model it. I always go through that when I’m doing it. I don’t do much speaking
5.6 Semi-structured Interview with the Course Leader 153
because I do the reading and writing on the autumn pre-sessional, but I would do. I would
get them to say that.
Interviewer: But not necessarily those words (points to words from Bruce/Olivia class).
Course Leader: No, no. I think the main point is to get them to interact with each other,
isn’t it: to use these phrases.
Finally, I asked whether the materials for the seminar were easy to use. The CL
replied that the materials are a guide, but did not really answer the question of whether
they were easy to use.
Interviewer: Do you think it’s quite easy to use those materials?
Course Leader: The whole thing?
Interviewer: Well, just generally what’s in the seminar.
Course Leader: I think it’s just a guide really. Have language all around the class as you
do it. Put a big A3 copy of it, and then just keep coming back to it all the time. It’s easy
enough to use, you’ve just got to remember to use it. If you display it, it’s something that’s a
visual reminder for them rather, because they won’t bring their papers in, will they? Students
always forget.
This section details the results from the teacher interviews. The section begins with
a brief description of the teacher training that teachers have received and their early
pronunciation teaching experiences. The results then move on to teacher confidence
and guidance, teacher goals in terms of instruction and assessment, opinions on
the syllabus, error correction preferences, declared practice priorities, and finally,
suggestions for how the course could be improved.
All three teachers have been teaching on the same pre-sessional course for a number
of years and had some experience of EFL or ESL teaching before they embarked on
their EAP teaching career. They all began by taking the preliminary certificate course
(either Trinity or CELTA). Mark followed this up with a Postgraduate Diploma in
TESOL, as did Bruce, and Bruce and Olivia have both completed a PGCE. Olivia
has completed a PhD in the study of a foreign language. Although this is not an
EFL qualification, it is worth noting as this experience may have some impact on the
instructor’s view of what instruction is provided and how that instruction is provided.
All three teachers generally report early positive teaching experiences although
experiences of teacher training, in terms of preparation for providing pronunciation
instruction, do vary. Mark could not recall any particular techniques that he was
taught, and argues that his present day instruction is more a result of personal expe-
rience and trial and error. Although Bruce states that his teacher training did not
154 5 Results, Analysis, and Discussion
involve training to teach pronunciation, and that it was fairly rudimentary in this
area, he also argues that the background knowledge gained in phonology was useful
in that it boosted his confidence and gave him the ability to incorporate the IPA in
lessons. Olivia argues that teacher training cannot prepare teachers in general, and
that experience is the key factor which informs teachers’ knowledge base. Also, she
expresses a firm dislike for pronunciation instruction in general.
Interviewer: What about, do you think it prepared you in any way for teaching pronunciation.
Olivia: No. I hate teaching pronunciation. Sorry. (laughter). Once I was teaching, I was
trying to… I was teaching a CPE group in Poland, when I worked in Poland, and the director
of the school, she was one of the students in the class, and she asked me how to write on the
board the phonemic letters for ‘colon’, and so… I just made it up and I just thought “Oh…”
(laughter). Yeah, yeah. So no, I don’t teach pronunciation. I hate teaching pronunciation.
Both Mark and Bruce express some confidence in providing pronunciation instruction
although Bruce does state that he is less confident when dealing with word stress
and that there should be more CPD, particularly on the instruction of intonation. He
also states that pronunciation instruction seems to be a neglected area in teaching
and CPD. Mark and Olivia echo this opinion that the guidance received could be
more comprehensive. Olivia states that she is confident in providing instruction, but
doubts its usefulness in some contexts.
Interviewer: Ok. Final question: so, do you think you would…would it be helpful to have
more guidance on pronunciation instruction or not?
Mark: Yep. Probably wouldn’t harm anything, yes. A bit more explicit guidance.
Interviewer: Do you think there’s enough guidance on how to use the materials?
Mark: Well no. It’s just kind of given to us and it’s expected that we know how to use them,
which is reasonably fair as we’re all trained teachers and most of us have done it before, year
after year. There isn’t much explicit guidance.
Interviewer: Yeah. Do you think there’s enough guidance.
Olivia: No. No there isn’t, no. It’s probably one of those subjects isn’t it, you know,
pronunciation, that people just think “I’ll just leave that” (laughter), I don’t know.
None of the three teachers were able to identify the goals of the course in terms of
pronunciation instruction.
Interviewer: If you’re thinking about pronunciation goals, yeah, what do you think the
pronunciation goals are on the course?
5.7 Teacher Interviews 155
Mark: I think the basics, probably… Do we have overall goals? (laughter) I wouldn’t know
explicitly what they were really, so I guess it’s anything to do with presentations and seminars,
so using good presentation and seminar language, and pronouncing it clearly with some
intonation and sentence stress.
Interviewer: Do you think…What do you think the goal of the course is in terms of
pronunciation? Is it stated anywhere in documents or…
Bruce: I’m pretty sure it isn’t. I mean, I suppose you could… I don’t know, you could copy
and paste the criteria of markers from a C (40-50 scale descriptor) and say this is the goal
of the course to make sure that everybody is happy with this sort of thing, but it’s definitely,
definitely not clearly stated there.
Interviewer: What do you think the goals of the pre-sessional are in terms of pronunciation?
Olivia: I don’t think there are any. Sorry (laughter).
Interviewer: I’ve been trying to find the goals and I don’t know what they…
Olivia: I mean, if you think about just pronunciation, I suppose it’s just to sort of like…
make them more fluent so that they’re going to be understood on the course. Maybe that’s
what it should be. I don’t know if that is the goal of this course, but you know, I just think
trying to get them so that other people can understand them.
It can be observed that, in the absence of clear explicit goals, Mark assumes some
rather general goals. The goal of pronouncing seminar language clearly could entail
a variety of interpretations, particularly in terms of whether native-like production
or intelligibility is the key goal. Olivia’s assumption is that of intelligibility as a goal
so that learners are able to communicate. Bruce does not offer his own interpretation
of what the goals could be, but states, not altogether seriously, that in theory, the
goals of the course could be gleaned from an inspection of the marking descriptors.
It does seem clear that an absence of course pronunciation goals could possibly lead
to teachers adopting inconsistent goals in both instruction and marking.
All three teachers argue that intelligibility is their overriding goal of instruction.
Bruce’s contribution is a typical representation of the stated approach of the teachers.
Interviewer: Would you say that the goal is to make students sound a little bit like native
speakers. I’m not saying that they can be, but is that the kind of goal or is it just that they
can be understood?
Bruce: I think it’s that they can be understood. Yeah, definitely because I don’t know, I
mean, I suppose this is from my own experience-I’m not sure if I sound like a typical native
speaker. In the staff room there’s probably 10 different variations of a native speaker. And
so, you might be a native speaker, but I suppose we all have slight differences. It gives us
our charm I guess, in that sense, and I think it’s the same for them.
Olivia also points out that native speaker production as a goal is unrealistic.
However, when asked about the correction of the /θ/ sound (for example, ‘I sink*’),
although Olivia states that she would not correct such a student error, both Mark and
Bruce disagree. Mark states that he would sometimes correct it, and Bruce states that
he definitely would.
Mark: I think sometimes you would, sometimes you wouldn’t because it depends on how
much time you have…what mood I’m in…where the students are at…if there’s really a
student with like the one thing that they need to tighten up on.
156 5 Results, Analysis, and Discussion
Bruce: Yeah, I think so, definitely. I mean, even if it’s…That’s the problem sometimes
because, if you, if you’re so used to a mistake then maybe it sort of washes off you, but I
guess it benefits nobody by not correcting them, if that makes sense.
However, when asked if this error does in fact interfere with intelligibility, he
states that it probably does not, but adds that the conversation might “stop a little”
as a result of it. This idea that lower functional load segmental errors could affect
comprehensibility has little support from research evidence (Munro & Derwing,
2006).
Bruce: It probably doesn’t. I mean, because we’ve had that much sort of, I don’t know,
what’s the word? ‘Immersion’ or you’ve heard it so many times that it doesn’t, but I’m just
thinking about, for example, a person on the street-they’re not going to have trained ears, I
guess, that we have. And so we definitely could correct that because I suppose it’s, I don’t
know, if they’re talking to someone and I don’t understand you, it’s not good for their kind
of self-confidence is it? And something as simple as that, which I guess, you use it a lot. But
if you can’t get past ‘I think’ sort of, the conversation kind of stops a little bit, doesn’t it?
Mark states that he would give a lower mark to a candidate who was substituting
the /θ/ sound with /s/, but does acknowledge that this seems to be inconsistent with
the overriding goal of intelligibility. He also offers the explanation that, perhaps
merely because it is a deviation from standard English, and therefore identifiable,
could be one reason why it might be given a lower mark. This idea that teachers
might concentrate on lower functional load segmentals due to the fact that they are
noticeable for native speakers is also discussed in Levis (2018, p. 29).
Interviewer: But what about if you’re marking it then. You’re marking two students who
have a very similar pronunciation, but one says ‘I sink*’ and one says ‘I think’, would there
be any difference in the mark or…?
Mark: Yeah probably. ‘I sink*’, it probably, it would get marked down. Probably not great
teaching. It’s a bit harsh in punishing somebody, I don’t know. That is, that is probably what
I did in your…
Interviewer: But I was just thinking that, because, if it’s not that important and you’re
marking it down…
Mark: Yeah. There’s a slight contradiction there, isn’t there? But I think when you’re marking
though, you’re just looking out for things that you can’t help, actual things that are wrong
and yeah… (inaudible)
Mark reaffirms that his main goal of instruction is intelligibility, despite the
exchange seeming to suggest that, at least in this one area, a native speaker model
is being followed. Olivia, like Mark, suggests that she would give a lower mark, but
then states that there is no descriptor to do that, presumably because these errors do
not significantly interfere with intelligibility or comprehensibility. If that is the case,
the desire to award a lower mark seems inconsistent with intelligibility as the goal
of instruction.
5.7 Teacher Interviews 157
Interviewer: If there’s one, if there were two students who, for example have the same level
and is one saying ‘I think,’ and one is saying I ‘sink,’ you’d probably give them the same
mark for pronunciation.
Olivia: No. Probably not. I’d probably give the other one a lower mark, but then we don’t
have the descriptor to do that. It doesn’t really seem as if those descriptors are really tailored
in detail enough. There’s not nearly enough detail to mark down for those small differences.
There may be many other things that are much worse.
Bruce’s answer is a little more nuanced. He argues that his marks would be
determined by the reactions of the student’s interlocutors in the exam, i.e., whether
they understood the utterance or not. This does seem to be more consistent with
intelligibility as a goal of instruction. However, this way of judging intelligibility
may also have flaws: Some learners may be in a seminar exam group with learners
with the same L1, whereas others may be in mixed L1 groups. This also applies to
the level of the learners: Some may be in groups with lower level learners who might
struggle to understand them.
Bruce: Yeah. It is, it’s a difficult one, and I would sort of, try and base it on maybe the
reactions of the other students, so maybe mark it down a little bit, but depending on the
confusion that it raises, if any. And so, if someone asks you “sorry, I don’t know what you
mean” or they have a puzzled look, or they say something differently to what the other people
say, then I might start to think “OK, this is, kind of, it is affecting other things as well” and
then that is where I would maybe start to think about scrutinising it a bit more.
The issue of comprehensibility is touched upon by both Mark and Bruce: Mark
states that it is quite important for learners to be easy to understand, whereas Bruce’s
contribution seems to emphasize the importance of intelligibility rather than compre-
hensibility. The point that Bruce makes about some native speakers being difficult
to understand is an interesting one, and is a reminder of the rather uncertain notion
of a native accent in general, particularly in the UK, where there are many regional
variations in accents.
Bruce: I think that’s the thing isn’t it. Maybe some native speakers take a lot of effort to
understand. I think you’ve got to try to give them that benefit of the doubt a little bit, if that
makes sense. I guess, it is a speaking exam and pronunciation is important, but the important,
I suppose, what’s more important is being able to take part in the thing. I mean, that other
people understand you.
All three participants argue that the descriptors used in the pronunciation category
can be difficult to use. Mark appears to suggest that it is the overall impression of the
performance that is the primary consideration, and that the descriptors are then used to
justify this overall impression. Bruce also points out the subjective nature of marking,
particularly drawing attention to specific language used in the descriptors. Olivia
suggests that the descriptors from the CEFR (Council of Europe) could perhaps be
exploited to provide more consistent marking because the descriptors are too vague.
158 5 Results, Analysis, and Discussion
Mark: It’s so vague-that can pass or fail. That’s what we do, “No that’s terrible,” and then
kind of use the mark scheme…I think (inaudible) buying your idea first and then use the
mark scheme to justify. Well, not quite as vague as that.
Bruce: Yeah, I mean it’s just, that’s the difficult thing isn’t it? ‘A few’, ‘some’, whatever. It’s
just that ‘some’ for you might be ‘a few’ to me, and that’s, I think that’s maybe the hardest
part.
Olivia: It’s not well-thought-out. I think they need to look at the Common European Frame-
work for languages really and use that, you know. There’s plenty they could take from
there.
Although both Mark and Olivia suspect that there could be quite varied inter-
pretations of many of the terms in the descriptors among teachers, Bruce believes
that these inconsistencies are usually overcome during the standardization process.
However, he does also suggest that more standardization would be beneficial.
Another aspect of marking touched upon by Olivia is the difference between the
pronunciation category and the interactive communication category on the mark-
sheet. Whereas the CL viewed interactive communication as an assessment of the
content of what is said and the pronunciation category as an assessment of how some-
thing is said, Olivia takes the view that the former category assesses the communica-
tive effect the performance has on the interlocutors, and the latter involves micro
elements.
Olivia: No, I think they are very different because one is… Well the pronunciation is more the
kind of micro level for the individual, but then…and how they’re producing, in a subjective
way, these sounds, whereas the interactive level is how that’s affecting people.
All three teachers raise doubts concerning the syllabus in terms of the level at which
it is pitched. One particular area that is raised is that the vocabulary items used, in
both seminar speaking activities and listening activities, are too complex for some
learners. Bruce points out that complex vocabulary in seminar classes can sometimes
be problematic, even for his higher level class, whereas Mark points to a general lack
of listening materials.
Bruce: I’ve noticed that a couple of times, even with my group that is quite strong, a couple
of the questions [are asked], yeah, sort of, OK. There’s just silence, so I say “okay what’s
wrong?” And they say “we don’t know what this means.” OK. So… You have to explain it.
And I think that, I don’t know, if you think about maybe how long it would take. It would
be 10 minutes or something. It’s not a massive sort of input of time, but it’s just …
Bruce also points to the complex vocabulary present in some of the listening
exercises that are in the syllabus. The example he gives is the Steve Jobs video.
Bruce: And I suppose if they can’t understand the vocabulary and grammar, then everything
else is gone because they don’t know if it’s well organized because they don’t know of the
topic and maybe they might be mispronouncing words, but they don’t know because maybe
they don’t know the words as well.
5.7 Teacher Interviews 159
Bruce argues that the syllabus tends to support higher level learners, but seems
lacking for those learners at a lower level. However, he does state that the seminars,
apart from being pitched at an inappropriate level, are quite well prepared, and that
he generally adheres to the syllabus.
Mark states that the materials on sentence stress (used during presentations
lessons) are useful, but also argues that the general listening materials that are used
(not the ones used in the presentation sessions) do not really support pronunciation
instruction. Also, although an explicit reason is not provided, Mark states that he
rarely uses the chunking exercises. These chunking exercises that form part of the
materials for academic presentations consist of short, isolated extracts. Olivia argues
that these extracts can also contain outdated information, that the listening materials
do not support pronunciation instruction, and that the chunking exercises should be
placed within a context.
Interviewer: What about the listening materials? Do you think they kind of support
pronunciation in any way.
Olivia: No. Not at all. I’m sorry (laughter).
Interviewer: Do you think they should?
Olivia: If you link chunking to a particular speech, for example, then that would be useful
because you can see what the point is, you’ve got an illustration, you’ve got a live example.
Interviewer: Yeah, the content first before the…
Olivia: Yeah. Yeah. So if you did it like that, then it would be useful, but just separately,
without any context, then…no.
Both Bruce and Olivia doubt the value of the video clip recommended in
the syllabus that they used during their video recorded observation, with Bruce
suggesting that although unsatisfactory from a listening point of view, it may have
been the only particular material that could be found on that topic. Olivia points out
that many listening extracts seem unnaturally slow and of an RP variety.
Olivia: Well, I mean, I was listening to the one today, on cities, and it was so slow
and so received pronunciation. And it was just really not realistic. You know, it was
just rubbish…you’d be better off listening to, I don’t know, not radio 4 because that’s
difficult…just people with different accents at a usual speed or a bit slowed down.
All three teacher participants cast doubt on whether the discussion language work-
sheets distributed to learners are useful. One or two of these sheets contained intona-
tion arrows to assist learners when using the particular phrases. Mark raises the issue
of learners simply losing these worksheets, and all participants argue that, even if
learners repeat these phrases in class, they do not go on to use them during the actual
seminar discussion (which is what they were designed for). The same point is made by
Olivia. This difficulty in transferring the language, along with the correct intonation,
obviously has implications for the effectiveness of pronunciation instruction.
Interviewer: Do you think they can use it though? I mean, I don’t know if…
Mark: Well, repeating some of them parrot fashion, but that’s not the same as producing
spontaneously…
Interviewer: And during the seminar?
160 5 Results, Analysis, and Discussion
Mark: I mean, I don’t know. I mean, I’d say probably not, to be honest.
Interviewer: Do you like the scheme of work? Does it work In terms of pronunciation stuff
and those chunking things? Do they work…the intonation things…do they work?
Bruce: I don’t think they do. I don’t think the students take it very seriously.
Interviewer: So you were modelling that, and my question is: Do you think that they use
that?
Olivia: No (laughter). That’s what I meant before. No, they didn’t use it at all. You can see
it was really worth it individually (heavy sarcasm, followed by laughter).
Olivia: No. The only thing that I’ve done is, like, intonation. And it was when I was a bit more
inspired at the beginning of the course and the students were still listening a bit, and then,
we did, yeah, we did the kind of intonation of “So do you agree?” (with rising intonation).
So, sort of, like a rise in intonation for questions and I’d get them to repeat it, and I thought
that was quite funny, and it was quite funny, but then I’d get, you know …say “Now you’ve
got to use the same…the same intonation during the seminar discussions,” and immediately
they didn’t do it. So I just wonder what the point is sometimes.
In terms of how closely teachers follow the scheme of work, Bruce states that he
generally tends to follow the scheme of work for seminars (which may imply that he
follows the presentation syllabus more loosely). Mark, as mentioned earlier, states
that he does not use the chunking exercises so much (from the presentation element),
and also states that he does not use the sheets containing seminar language as these
are usually lost by students. Instead, he uses one general sheet for the whole course.
He also states that he adheres to the syllabus “very loosely.” Olivia’s responses do
not show how closely she follows syllabus, but she does seem somewhat critical of
it, particularly in the area of listening.
Presentations were discussed in the interviews with Mark and Bruce, who both
state that the fact that the assessment of presentations is only formative is problematic
because learners are not motivated to try hard and receive no reward for their presen-
tation performance in terms of a mark (they only receive general comments). Bruce
also points out that on the business course where he himself studied, presentations
were much more common than seminars, and that presentations should therefore
have some kind of summative assessment on the pre-sessional (where many learners
will go on to study business-related degree courses).
Interviewer: Do you think that dropping the presentations as a summative assessment was
a mistake or…? You did presentations today.
Mark: Yeah. Yeah, I think it’s a shame that they put in a lot of work and then it’s not, it
doesn’t really mean anything. Some of them have done a lot of work and some have done
no work at all, and it doesn’t really matter. Yeah. So yeah, I think it should be something…it
should be, maybe, perhaps 25% was a bit high, but I mean 20%, 15%, yeah.
Interviewer: So there should be some kind of assessment.
Mark: I think so, yeah. I think that was a common consensus wasn’t it? …In the meeting
(teachers’ meeting with the external examiner).
Interviewer: Do you think that helps them with pronunciation a bit: the presentation element
of the course?
Mark: Yeah, because hopefully they’re going to make sure they know at least all the words
in their presentation; they’ve checked how to pronounce them and…
5.7 Teacher Interviews 161
Bruce: I mean, from my experience on a business courses, I didn’t do any seminars. I did
four or five presentations. And so, I guess depending on the course, the presentation skills
will probably make up a much bigger part, and so I’m not sure why we’ve sort of gone the
other way.
Bruce: I mean, I never, we never used to sort of have it in previous years where they didn’t
take it very seriously, whereas now you can tell that some people are like…Maybe even as
well… Maybe even the formative sort of writing as well: maybe having that count towards
it. I don’t know… something.
In terms of pronunciation error correction, both Mark and Bruce state that they correct
more when students are working in small groups rather than in whole class mode.
Mark points out that it is more sensitive to provide correction in small groups. He
also states that he says the word himself if learners are struggling to pronounce it.
Bruce argues that he does more error correction in groups because more language is
produced when learners work together in groups, and there is therefore more language
to correct. Bruce also states that the modelling of new vocabulary, followed by whole
class repetition is largely unsuccessful because it is difficult to hear what the learners
are actually saying. However, Olivia states a preference for correction in whole class
mode, citing exactly the same reason given by Mark for correction in small groups:
sensitivity. I presume that Olivia is referring to the type of modelling and repetition
that Bruce alludes to, but feels that it is possible to hear the learners articulating the
words correctly, and that they feel more relaxed when repeating the words in unison
with their peers. Teachers seem to be acutely aware of the need to be sensitive when
approaching instruction, but this desire does lead to different approaches and outlooks
because there are different assumptions about which correction circumstances make
learners feel safer and more relaxed.
Interviewer: Yeah. But would you sometimes give feedback to groups, you know, when
they’re doing their seminar thing?
Mark: Yeah…Yeah, I walk around and give feedback immediately and correct somebody.
Mark: Yeah, generally if they mispronounce a word, I’d probably just say that word again. I
wouldn’t like, force them: “No. You need to repeat this five times now”… and so I just kind
of, If I’m kind of getting halfway through and they’re really struggling, I just kind of say the
word loudly for everybody. I don’t know how great teaching this is for everybody because it
shows that they made a mistake, but I think just saying like, if it’s like a really long difficult
word… I don’t think there’s any harm in just saying those difficult words for them, and then
minor mistakes you probably just let that go. If you’re correcting them all the time, they’ll
never say anything because they say so little anyway.
Bruce: It depends. I mean, for example, if I’m going around monitoring a small seminar
and generally, sort of, maybe there and then, or sort of after they finished that little bit. I
mean it’s not so much. I guess in the whole class there’s probably less or they’re giving you
less, whereas in the smaller groups they’re producing more and so there’s more corrections
needed, if that makes sense.
162 5 Results, Analysis, and Discussion
Bruce: Yes. I know what you mean. It depends. I mean, if you try to do it as a whole class,
then you’re going to get 15 people saying… everyone pronounces it incorrectly. I can’t tell
what you’re saying sort of thing.
Olivia: It could be whole class. I think it depends, you see, I don’t like the students to feel
embarrassed by the mistakes, so I often tend to do it as a whole class, otherwise I might do
it individually, just in a tutorial, but I tend to do that as a whole class so that they don’t feel
embarrassed in front of their mates.
On introducing new vocabulary items, Mark states that he might ask learners to repeat
some items in order to practise the pronunciation of those items, whereas Bruce points
out the difficulty in actually hearing the pronunciation of individual learners when
they repeat the items in unison, whilst also acknowledging the difficulty of hearing
the production of those words in group phases of the lesson. This does seem to be
a problematic area: trying to balance the need to hear learners pronouncing the new
vocabulary items with the limited time that is available for speaking.
Bruce: I think sometimes in the seminars, we’ve done this list of vocabulary before listening,
and what we find when we’re going around, I’ll go and ask them “What does this word mean?”
And you’re asking the…and I think that’s probably a better way, but it’s difficult to sort of
get round all of them…
Bruce identifies intonation and rhythm as the most important areas of instruction.
At a later stage in the interview, sentence stress is also mentioned as an important
element, but not word stress or weak forms. Bruce mentions that weak forms do not
“come up very often.” Presumably, this means that they do not really interfere with
intelligibility.
5.7 Teacher Interviews 163
Olivia’s statements on practice are radically different from Mark and Bruce from
the outset. Although she states that she hates teaching pronunciation, from the
description of learners “parroting” teacher output, it appears that she is describing
a particular type of pronunciation instruction that she detests. Later in the inter-
view, Olivia mentions that she does cover word stress, but doubts its effectiveness.
She mentions that, as was sometimes apparent in the observed lesson, she does get
learners to repeat difficult words in class. It is also stated that she tends to do this
when learner production is unclear, perhaps suggesting that intelligibility or compre-
hensibility are the goals of instruction. She identifies word stress, sentence stress,
intonation, and rhythm as important elements of instruction on the course rather than
segmentals.
Olivia: Yeah, word stress, yeah, but. Yeah, I mean, I find that some students don’t do it
anyway. They just sit there and stare into space.
Interviewer: So you think, generally I guess, you’re thinking that not much time should be
spent on pronunciation anyway.
Olivia: Not if it’s not necessary.
Interviewer: How do you know when it’s necessary? How do you guess that?
Olivia: Because if the students are not clear… If they’re kind of mumbling or if they’re
really getting all the words mixed up…but then that’s…it’s kind of like, if they’re here, then
they should have a level of English which should combat that really. Okay, okay, so there
are always words that you don’t know how to pronounce, even as a native English speaker
sometimes I have to look in a dictionary to find out how to pronounce it, but you know,
there should be a certain level for pronunciation as well as, you know, it’s tied in with the
speaking.
Interviewer: Yeah. So you generally avoid these, kind of like, single sound listen and
repeat…
Olivia: Yeah, because I feel if you do word stress and you do sentence stress and things like
that, that should incorporate it.
Aside from the teacher comments regarding simplifying the language used in the
materials, reintroducing summative assessments for presentations, and having a
greater variety of accents, Bruce provides a number of suggestions in terms of how the
course could be improved. One of the points raised is that learners should be encour-
aged to be independent learners a little more by providing pronunciation resources on
the university e-learning website for self-study. There is also the suggestion that the
pronunciation sessions could be more learner-focused, with the instructor acting as a
monitor. Furthermore, Bruce suggests that the discussion topics themselves could be
more learner-focused and concentrate on topics that interest the learners in order to
encourage them to participate. This seems a feasible suggestion, particularly as the
pre-sessional course is a course of English for general academic purposes, i.e., there
is little subject specific content. How the interests of the particular classes are deter-
mined is a more problematic consideration, but if learners are more motivated, and
164 5 Results, Analysis, and Discussion
This section provides details of the results of the teacher questionnaire. Biographical
data are presented first, followed by an examination of the other results: firstly the
closed questions, followed by the open ended questions.
The total number of teachers on the three courses (five, ten, and fifteen-week courses)
was 26. Of these teachers, a total of 23 completed the questionnaires: 7 of the 9
teachers on the five-week course, 11 of the 12 teachers on the ten-week course (I
did not complete the questionnaire myself), and all 5 of the teachers on the fifteen-
week course. There were 11 male teachers and 12 female teachers; the date of birth
of 9 teachers was between 1949 and 1969, 8 teachers’ date of birth was between
1972 and 1992, and 6 teachers chose not to declare their date of birth. In terms of
teaching experience, the majority of teachers had a good deal of experience teaching
on the pre-sessional: 10 teachers had up to 6 years experience, and 13 teachers had
between 7 and 18 years experience. Only two teachers possessed the lower level
Trinity TESOL certificate or equivalent with no higher level qualification, with the
remainder (save for one no data response) qualified with the higher level Diploma
TESOL/TEFL and/or MA qualifications. The vast majority of the staff therefore
possessed level 7 qualifications or above according to the UK Ofqual Regulated
Qualifications Framework.
The teachers were presented a list of 20 statements and asked whether they agreed
or disagreed with them (on a scale: from strongly disagree to strongly agree). Their
responses were then collated and divided into agree/disagree answers (see Fig. 5.1).
1 13 10 0
2 6 15 2
disagree
3 5 18 0
4 7 16 0 agree
5 15 8 0 don't know
6 19 4 0
5.8 Teacher Questionnaires
7 13 10 0
8 14 8 1
9 3 19 1
10 19 3 1
11 5 18 0
12 7 15 1
13 12 5 6
14 1 20 2
15 7 15 1
16 2 15 6
17 3 15 5
18 10 12 1
19 14 8 1
20 3 20 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Fig. 5.1 Teachers’ attitudes towards different aspects of pronunciation (in absolute numbers). Source Author’s own: The statements (1–20 on the vertical axis)
are listed on the next page; the horizontal axis represents the number of teacher responses to the statements (N = 23)
165
166 5 Results, Analysis, and Discussion
research
3
from another person
reading
6
0 5 10 15 20
by intonation and sentence stress. However, both rhythm and weak forms received
less favourable scores, suggesting that teachers view these as less important areas
(5/8/9 for rhythm and 5/9/8 for weak forms). Although word stress, sentence stress,
and intonation received fairly high scores from teachers, it should also be noted that
both vowel sounds and consonant sounds received a rating of 1–3 from a significant
minority of teachers (7).
There were 8 no data responses to the open ended question number 4, regarding
how the pre-sessional could be improved in terms of pronunciation instruction. Of the
15 responses, the most common was the suggestion that more time should be devoted
to pronunciation instruction (6). Also, two teachers suggested a designated session
for pronunciation instruction, and a further two teachers proposed more guidance
and training. Other suggestions (one teacher per suggestion) included the provision
of self-study pronunciation materials for learners online, longer tutorials to enable
one to one support for learners, an adjustment of the materials to reflect the students’
level, and a communicative component to improve learners’ confidence. However,
one teacher suggested that the pronunciation instruction provided was sufficient.
Open-ended question number 5, regarding the difficulties teachers face when
providing pronunciation instruction, had 17 responses (6 no data). Some teachers
mention multiple factors as obstacles to providing instruction, but by far the most
common mentioned is a lack of time, with 10 teachers referring to this (one of the
teachers specifically refers to a lack of preparation time). Other responses include a
lack of training or guidance (2), a lack of materials/activities (1), learner pronunci-
ation of vowels (1), the seminar marking criteria (1), large classes (1), an emphasis
on strategies (1), difficulty in incorporating instruction into listening tasks (1), and
ad hoc provision (1). There was also one teacher who reported that there were no
problems in providing instruction. When examining responses to both suggestions
for improvement and problems faced by teachers, it is clear that, from those teachers
who responded, time is the key consideration.
170 5 Results, Analysis, and Discussion
Russian
Arabic
Japanese
87 Thai
This final section provides the results of the questionnaire data collected from the
learners. The most salient points are addressed, firstly in terms of biographical data,
followed by the closed question statements in question 1, and closed questions 2 and
3. Open-ended data results are then categorized and presented.
A total of 156 out of a possible 161 students answered the question regarding their
native language. Of these students, 87% were Chinese speakers, 6% were Arabic
speakers, 3% were Thai speakers, 3% were Russian speakers, and 1% Japanese
speakers (Fig. 5.4).
Students had been learning English for an average of 9 years. Approximately one
third of the learners had up to five years experience of learning English, one third
between six and ten years, and another third had been learning English for eleven
years or more. In terms of time spent in an English speaking country, for the vast
majority of respondents study on the pre-sessional course was their first experience
of being in an English speaking country (78% reported two months at the time the
questionnaire was completed).
A summary of the results from question 1 can be found in Table 5.7.
5.9 Learner Questionnaires 171
Table 5.7 Question 1: The learners’ attitudes towards pronunciation (in percentages)
Disagree Agree Don’t Know
1 I am confident about my English pronunciation skills (I 18 81 1
think other people can understand my pronunciation)
2 I am more confident about my pronunciation now than 12 87 1
when I started the course
3 I want to improve my pronunciation skills in English 3 96 1
4 I want my teacher to teach pronunciation 8 90 2
5 I want my teacher to correct my pronunciation 3 96 1
6 I like it when my teacher corrects my pronunciation in 12 87 1
front of the class (when the class can hear what he/she
says)
7 I like it when my teacher corrects my pronunciation 14 85 1
privately (e.g., in tutorials) (The class can NOT hear what
she he/says)
8 I like it when my teacher corrects our pronunciation as a 11 88 1
group in class. (He/She does NOT focus on me
individually)
9 I want to listen to American English speech 37 61 2
10 I want to listen to British, Australian, Canadian, 22 77 1
American or other native-English speech
11 I want to listen to Indian, Chinese, Arabic, French, 50 48 2
Spanish or other English-speech models
12 I want to speak English with a British, Australian, 11 88 1
Canadian, American or other native-English accent
13 I want to speak English with my native language accent. 41 57 2
(E.g. Chinese, Arabic, Korean, etc.)
14 My pronunciation has improved over the course 9 89 2
15 The pronunciation activities we did on the course were 13 84 3
interesting
16 The pronunciation activities we did on the course will 5 95 0
help me on my course in the future
17 The pronunciation activities we did on the course will 4 95 1
help me to speak to other people outside the classroom
18 The listening activities in class were interesting 15 83 2
19 The listening activities in class will help me to understand 11 88 1
others outside the classroom
20 The textbook and/or materials that the teacher uses are 12 87 1
helpful for improving my pronunciation
21 My teacher usually corrects me when I pronounce words 9 90 1
incorrectly in speaking classes
22 My teacher usually corrects us if we do not stress the 9 90 1
words in a sentence correctly in speaking classes
(continued)
172 5 Results, Analysis, and Discussion
the classroom (88%). Learners also generally agreed that the materials used by the
teacher were helpful in improving their pronunciation (87%).
Answers regarding correction sometimes provided rather contradictory answers.
90% of the respondents report that their teacher usually corrects their pronunciation
in speaking class, and a similar number of learners also report the correction of
word stress errors in both speaking and writing classes. The vast majority of learners
who completed the questionnaire also report that their teacher usually corrects their
intonation errors in speaking classes (89%). Of course, one problematic area here
is that learners may be unaware that they are making intonation or sentence stress
errors (Derwing & Rossiter, 2002). It may therefore be the case that the teacher does
not usually correct these errors or corrects only some of them. However, the finding
that learners perceive that such corrections usually take place may indicate that they
usually expect these correction to take place. Somewhat contradictory answers are
apparent in questions regarding the circumstances of error correction. Although many
learners report that correction takes place privately, on a one to one basis (74%), and
also in groups, without focusing on the individual learner (87%), a similar percentage
do report that correction takes place in front of the whole class (81%). Of course, it is
possible that all of these types of corrections are made by the teacher in class, but it
seems unusual that the percentages are so high for each category. Another explanation
is that the learners may simply be providing the answers that they perceive they are
expected to give. This possibility is underlined by the fact that the majority of learners
agree with almost every statement. Finally, the majority of learners agree that their
teacher asks them to watch English TV programmes or listen to the radio (93%) and
asks them to record their own language production (88%). Of course, we do not know
if learners pay heed to such advice although the percentage of learners who report
that they check the pronunciation of new words on the internet (92%) does seem to
suggest that, in this respect at least, learners have been encouraged to become more
independent learners.
In terms of frequency of instruction, when questions 2 and 3 are compared, they reveal
an average of 25% actual instruction time reported, compared with an average of
31% reported as the ideal percentage of time that should be devoted to pronunciation
instruction. Although both of these figures are unrealistic when we consider that
there are few lessons devoted to speaking and listening when compared to writing
(for example, some learners stated that over 50% of instruction time should be spent
on pronunciation), it is perhaps significant that learners state a higher figure for
the ideal percentage of time. Perhaps more significantly, when an analysis of the
groups is conducted, those groups that report having less instruction (5, 7, 8, and
9) have higher figures for the ideal time that should be spent on it. So it seems that
those learners who receive less pronunciation instruction would like much more of it
(according to their reports) and that learners generally would like more instruction.
174 5 Results, Analysis, and Discussion
Table 5.8 Relationship % should be how long have you been learning English
between the length of time spent on Up to 5 years 6 to 10 years 11 and more
learning English and the pronunciation years
opinion of the percentage of
time that should be spent on Up to 5% 5 4 0
pronunciation From 6 to 10% 4 16 9
From 11 to 20% 18 32 29
From 21 to 50% 69 36 50
51% and more 4 12 12
Total 100 100 100
Source Author’s own; Cramer’s V = 0.24—moderate relationship
A cross tabulation analysis (conducted using SPSS software) of the length of time
studying English also reveals that those learners who have been studying English for
a longer period of time would like more pronunciation instruction (Table 5.8).
Figure 5.5 illustrates the activities learners nominate as the ones their teachers use
that are the most helpful for improving pronunciation. Listening was mentioned by
31% of the learners. Other prominent categories included speaking in groups (22%),
presentations and seminars (17% each), and reading aloud (12%).
35
31
30
25
22
20
17 17
15
12
10
7 7
5 3 3 2 2 2 1 1 1
0
Fig. 5.5 What activity has your teacher used that is most helpful for improving your pronunciation?
Source Author’s own
5.10 Analysis 175
5.10 Analysis
What follows is an analysis of the results in terms of the Research Questions I sought
to answer. The Research Questions are dealt with in turn, and this analysis is then
followed by a discussion of the results/analysis.
The course documents distributed to teachers during induction do not clearly state
the pronunciation goals of the course, and the student handbook also shows a lack
of clearly stated goals. From an inspection of the course documents, it is clear that
there is no segmental instruction in the syllabus, and that the instruction that does
exist is of a suprasegmental variety. However, the seminar syllabus seems to contain
few suprasegmental elements.
The CL interview further reinforces the opaque nature of the goals, with intelligi-
bility being stated as the goal of instruction, but other statements made which seem
to suggest that errors occurring with the low functional load item /θ/ could interfere
with intelligibility. Although the interview provides equivocal results on the issue, it
is noticeable that the only pronunciation instruction advice given during the induction
was that this particular segmental error should be corrected. It was also stated that
this correction could “fix” the erroneous pronunciation. This would seem to suggest
that native speaker-like production is the goal that is being set for instructors.
It is clear that there is much more of a focus on pronunciation instruction in the
presentation syllabus, and it is also clear that the major pronunciation goal is supraseg-
mentals. However, a number of exercises taken from the main coursebook relating
to suprasegmental instruction that were apparent in the syllabus in previous years
were removed from the syllabus in 2018. The reasons for this are unclear because the
CL did not devise the syllabus and was therefore unable to clarify particular motives
for this change. Also, the syllabus materials could be more effective if guidance
were provided on how particular exercises could be better implemented. Scaffolding
opportunities for learners seem to be limited with both seminar and presentation
materials. It is clear from the CL’s statement that it is perfectly possible for teachers
to simply follow the syllabus, and that they do not need to adapt the materials or devise
new materials, that the assumption is that the syllabus and materials are adequate for
providing pronunciation instruction. Like Mark in the semi-structured interviews,
I feel this is an overstatement of the efficacy of the syllabus. Notes from my own
action research also suggest that the materials often require modification in order to
provide suprasegmental instruction that is more meaningful: The chunking exercises
did not have a general comprehension section before the chunking (as recommended
by Brazil, 1994, p. 3). Also, it seems unclear how chunking tasks will lead to students
176 5 Results, Analysis, and Discussion
chunking their own speech in the same way, or how teachers might guide students to
achieve that end. Furthermore, the lack of guidance supplied along with the materials
in terms of annotations, how the materials might best be implemented, and the lack
of verbal guidance throughout the course show that although some suprasegmental
instruction is embedded in the syllabus, its implementation might not be so straight-
forward. The CL does acknowledge that the materials themselves do not contain
guidance, but argues that teachers will probably liaise with colleagues to discuss
how best to utilize the tasks. However, questionnaire data suggest this is unlikely:
Very few teachers state that they have learnt what they know about pronunciation
instruction through colleagues.
One simple example of how this lack of guidance can lead to different instruction
is evident in the observed classes. Of course, variation is inevitable because teachers
are individuals, with their own viewpoints, beliefs, and attitudes. However, it does
seem strange that learners sometimes do not say the new vocabulary items that are
introduced to them. It is also unusual that the CL suggests that quite complex new
vocabulary might not need to be spoken by the learners because it is not of an
academic nature.
Similarly, all observed teachers noted the difficulty in transferring the discussion
language from the seminar worksheets to the authentic discussions that take place
among learners in the free seminar discussion phase of the lesson. For example,
instructors may introduce phrases such as can I come in here? with intonation arrows
to assist learners so that they can use this language for interrupting, but the difficulty
lies in promoting authentic usage in the setting of a seminar. Once again, no guidance
is provided in the materials on how this could be achieved.
Another key example of how the syllabus might not be promoting suprasegmental
instruction is the fact that presentation classes, which are by far the most significant
element of the course promoting pronunciation instruction, only progress to a final
formative assessment. There is no summative examination for presentations. All
three teachers who took part in the research point out that this can lead to a lack of
motivation for learners. I also observed this phenomenon with my own class.
The materials for learners to evaluate presentations seem a little unrealistic. The
first week of the syllabus, containing the Steve Jobs video presentation task, was a case
in point. Learners were expected to either use the presentation marksheet to evaluate
the presentation or use a simplified evaluation sheet. Firstly, it is an extremely difficult
task for learners because the language on the marksheet is difficult to understand and
time-consuming for the teacher to explain. Secondly, the type of presentation that
learners are required to give is different in nature to the Steve Jobs one: They are
not selling a product; they are giving an academic presentation. Thirdly, although
the language in the video is challenging, a more serious problem is the language
used in the presentation criteria key features worksheet. It may be too ambitious
to expect students to evaluate the pronunciation of Steve Jobs because learners are
expected to take notes on pronunciation accuracy, features of intonation, stress, and
L1 interference. The final category is obviously redundant as the video is of an L1
speaker, but it seems unrealistic to expect learners who have a 4.5 or 5.0 score in
listening to be able to evaluate these other pronunciation features in an L1 speaker in
5.10 Analysis 177
the first presentation lesson. It may be difficult for learners to make value judgments
on intonation and stress, particularly in their first lesson.
Peer assessment of pronunciation is encouraged, but the categories that are
included for learners to check their peers’ pronunciation include use natural spoken
English, use pauses for emphasis, and that their delivery should be clear, simple, and
fluent. It is difficult to ascertain what learners with an overall IELTS level 5 score
will understand by the words fluent or natural, but it is unlikely that such terms
will be useful. Similarly, the self-assessment of pronunciation is encouraged through
the use of a self-assessment checklist (for use with presentations). Self-evaluation
can certainly be an important tool for realizing the improvement of suprasegmental
pronunciation. However, not only are the categories far too general, it is also far too
simplistic and rather odd to ask students to make a binary choice. The self-evaluation
sheet has the following questions: Did you pronounce all the words correctly? Did
you stress important words? and Did you use intonation to show interest?
The listening goals of the course seem to adhere to a native speaker goal. Almost
all of the listening extracts were SAE or SSBE varieties. It seems a little odd that
the accents represented not only fail to include many non-native varieties of English,
but that, given the context of instruction (a university in the North of England), the
native varieties include only one very short listening extract using Northern English
over the duration of the course. There does not seem to be any focus on segmental
elements, which may seem a little surprising as the vast majority of learners are
Chinese speakers, and may therefore experience common problems with certain
segmentals that could interfere with intelligibility.
The listening extracts that are intended for self-study are all either SAE or SSBE. It
is also unclear how many learners actually use these resources. It is perfectly possible
to take part in the seminar speaking activity without having completed the listening
support resources. It may therefore be the case that only the more highly motivated
learners use these. Also, bearing in mind that the vast majority of learners are from
a culture where independent learning is less common than in the UK (China), it
may have been more beneficial to encourage independent learning a little more by
checking learners’ work and providing feedback rather than it being a take it or leave
it option for learners.
that these errors cause strain and interfere with comprehensibility. Because a mixture
of both intelligibility measures and comprehensibility measures are used in the exam
marksheet, it could be said that a native speaker-like goal is being pursued to a certain
extent, particularly because there is no common threshold of intelligibility in all
descriptors. The marksheet’s even division between segmentals and suprasegmentals
suggests an estimation of the equal importance of the two, but this is not mirrored
by the syllabus, which includes some suprasegmental tasks but no segmental ones.
Another problematic area is the use of expressions of quantity in the marksheet
descriptors. In the teacher interviews, Bruce talks of the difficulty in interpreting the
difference between a few and some. The fact that there are only two descriptors in the
pronunciation category also means that any differences in interpretation will have a
more dramatic effect on the overall mark given in the pronunciation category. Also,
the teacher and CL interviews seem to show a subtle, but important difference in
the interpretation of the function of the interactive communication category. Perhaps
these issues of defining the key elements of the marksheet could be one of the reasons
why the teacher questionnaire data (question 1: statement 18) show that a substantial
number of teachers (10) did not find it easy to mark the pronunciation category.
The teacher interviews show that the marking goals of the teacher participants vary.
For example, Mark states that he probably would give a candidate who substituted /θ/
with /s/ a lower mark, suggesting native speaker production as a goal. However, Bruce
and Olivia’s responses indicate a goal of intelligibility. The teacher questionnaire
data also seem to suggest that there is some variability in this respect. Although
not a majority, a significant number of teachers agree with Mark. This seems to
show a division in teacher assessment goals that could lead to inconsistent marking
that might not accurately reflect how intelligible the candidate’s pronunciation is to
others.
The potential for variation in marking is revealed in the marks given by the three
teacher participants in the seminar assessment stage of the research. Of course, there
is an element of subjectivity in marking, and particularly when marking speaking,
but the variation in the marking of the pronunciation category is quite striking when
compared to the other categories. This may be a reflection of the problematic mark-
sheet containing too few descriptors and imprecise descriptors, combined with the
standardization guidance regarding low functional load segmentals and lack of guid-
ance regarding the marking goals. In sum, it appears that the number of descriptors,
the vague, undefined language that is sometimes apparent within those descriptors,
and the lack of a clear statement of pronunciation goals during course standardization
might be combining to produce the different marker interpretations.
The non-native speaker teacher who related the views of the double marker (native
speaker) when there was a disagreement about the perceived intelligibility of a learner
during seminar marking is also of note. The native speaker teacher’s claim to superi-
ority in making such a judgment because of native speaker status is quite revealing:
If intelligibility was the goal for this particular teacher, why would a native speaker
be more qualified in making such a judgment? This would also seem to suggest
that this particular teacher viewed native-like production as a goal. It also suggests
that the view of native speakers as ‘expert’ judges of intelligibility is still present
5.10 Analysis 179
(at least in this one case), and that there may be an unequal power dynamic at play
between native and non-native teachers similar to that outlined by Matsuda (2019)
and Phillipson (2016).
The analysis of course documents related to pronunciation assessment, the inter-
view with the CL, the teacher seminar grading task and follow up interviews, the
semi-structured teacher interviews, and the teacher questionnaires seem to show that
the course goals related to pronunciation assessment are rather opaque. The lack of
clear goals and guidance for teachers may therefore be contributing to the provision
of assessment that is not always based on intelligibility as a goal, is not sufficiently
detailed due to the lack of descriptors, and perhaps does not accurately reflect the
crucial role of suprasegmentals in communication. There seems to be little guidance
given throughout the course in terms of which particular sounds affect intelligibility,
meaning variation in marking is more likely.
When asked in the interviews, none of the three observed teachers could identify the
course’s pronunciation goals. In the absence of clear course goals, it appears that
teachers form their own goals of instruction. All three observed teachers state that
they do not strictly adhere to the syllabus, and they are critical of some aspects of
it. For example, Bruce points to the dense language used in the materials intended
to facilitate and encourage discussion, something that I was also acutely aware of
with my own weaker learner group. Also, although all three teachers declare that
the goal of instruction is intelligibility, two of the teachers state that, very much as
suggested by the CL in induction, they would correct the substitution of /θ/ sound
with /s/. Once again, the correction of this relatively low functional load segmental
seems to suggest that, to a certain extent at least, native speaker-like production is
being pursued as the target of instruction. In the questionnaire, the vast majority
of teachers declared intelligibility as the target of instruction. However, bearing in
mind that the two observed teachers, Mark and Bruce, also declared intelligibility
as the target of instruction, in spite of appearing to adhere to different goals in some
respects, it could be the case that the same phenomenon is being observed with
teachers in the questionnaire. They may be declaring intelligibility as a goal without
fully understanding its meaning in terms of instruction.
Checklist data show that the corrections declared by the three teacher participants
are generally ones involving suprasegmentals. However, on the basis of only very
limited observation, it is difficult to confirm whether this is true in practice. The
observations do confirm that there seem to be few corrections of segmentals, but
also shows that there is considerable variation in terms of the quantity and manner
in which suprasegmental instruction is employed.
180 5 Results, Analysis, and Discussion
5.11 Discussion
Although the syllabus indicates that where pronunciation instruction does exist, it
is of a suprasegmental nature (consisting largely of chunking, word stress, sentence
stress and intonation exercises), whether the institutional goals of instruction are
closer to what Levis (2005) termed the intelligibility principle or the nativeness
principle is far from clear. The listening materials suggested in the syllabus consist
of, save very few exceptions, SSBE or SAE native varieties. Although it is perhaps
a little surprising that native accents are so dominant in the listening materials,
given the teaching context (a Northern English city), it is even more surprising
that there is only one example of a relatively short listening extract that contains a
speaker with Northern English pronunciation. If learners are successful on the pre-
sessional course (as the overwhelming majority are), they will be spending between
one and three years in the city, and apart from their academic study, they will be
surrounded by Northern English pronunciation. At least a few more examples of
Northern English extracts would be welcome and may serve a purpose in helping
learners to understand the everyday English that is all around them. This could
assist in minimizing social and psychological barriers to language acquisition by
minimizing what Schumann (1978) termed transition anxiety, and as a result, it could
aid assimilation. This seems particularly relevant when we consider Deterding’s
research (2005) that emphasizes the value of listening training to improve learners’
receptive skills in terms of native English accents. In light of various research and
viewpoints emphasizing the international nature of English (Jenkins, 1998, 2000;
Kachru, 1992) and the need to expose learners to a variety of types of English
(Thomson, 2018), the lack of non-native listening extracts is also noteworthy. Of
course, learners at University A will primarily need to communicate with their tutors,
5.11 Discussion 183
fellow learners, and members of the public in the city, and it is likely that the majority
of these interlocutors will be native speakers. However, some of their fellow learners
(and some of their tutors) will, in all likelihood, not be native speakers. In an attempt to
redress this imbalance in the syllabus, I devised a listening exercise of my own which
was based on a local TV news report. It included standard Northern English (the local
newsreader and a local council leader), a Chinese speaker of English, and one or two
examples of more colloquial local speech. Learners found this listening engaging and
were motivated by the task; they liked the fact that they could see parts of the city,
hear local people speaking, and see one of their compatriots who had successfully
established a business in the city. Providing learners with role models of successful
non-native English speakers is beneficial (Cook, 1999), and this consideration of
learners’ feelings and their new learning environment can be a valuable means of
enhancing motivation (Arnold, 2011; Gadd, 1996; Liu & Jackson, 2008; Stevick,
1980). However, the major difficulty is that it is time consuming to devise such
activities, and in the context of the marking load and other preparation tasks, it is
unrealistic for teachers to prepare more than a few of these tasks.
It seems that the extent to which the teachers adhere to the syllabus is variable. All
three teachers report some problematic areas of the speaking syllabus and therefore
felt the need to depart from it or adapt it. For example, Mark follows the syllabus
“very loosely,” and Olivia does not use all of the discussion language sheets that are
suggested in the syllabus. Although Bruce is dissatisfied with the overly complicated
language in the materials, he generally approves of the seminar materials and usually
follows the syllabus.
One of the most significant findings from the teacher questionnaire data is that over
half of the staff (who are generally highly qualified and experienced) feel that teacher
training did not prepare them for teaching pronunciation. This echoes the findings of
Baker (2011), Couper (2017), Foote et al. (2011), and Macdonald (2002). When we
consider that, of the teachers who responded, the majority declared teacher training
as their primary source of knowledge for pronunciation instruction, the data seem
even more significant. These data, coupled with the clear desire of the vast majority
of teachers to have more training in pronunciation instruction, seem to show that,
despite the majority reporting that they are confident with teaching both segmentals
and suprasegmentals, there is clearly space for more training in this area. Another
major finding, that the greatest obstacle to the provision of instruction is a lack of
time, shows that perhaps this need for training is even more pressing, and materials
could be adapted and new ways found to incorporate pronunciation instruction into
a range of lessons (even, for example, in writing classes).
Although the goals of the course are rather unclear, the vast majority of teachers
state that intelligibility should be the target of instruction. However, although not a
majority, a significant number of teachers agree that they would give a lower mark
to a student who substituted /θ/ with /s/. Perhaps this is another example of what was
observed in Sifakis and Sougari (2005): that some teachers agree with the principle
of intelligibility as a goal, but also feel pressurized to provide instruction to meet the
needs of exams. It could also simply be the case that these teachers assume that the
substitution of the dental fricative interferes with intelligibility or comprehensibility,
184 5 Results, Analysis, and Discussion
perhaps as a result of the only pronunciation instruction advice they were provided
with throughout the course (that this substitution could and should be remedied). The
estimation of the relatively low importance declared for weak forms does support the
idea that teachers appear to be adhering to a goal of intelligibility though as weak form
errors do not generally interfere with intelligibility (Jenkins, 2000; Roach, 1983). This
is also supported by the teacher interview data, where participants viewed certain
suprasegmental elements as being important for instruction but not weak forms.
Learners overwhelmingly view the listening and speaking activities on the course
as helpful, and a minority of learners state that reading aloud activities [outlined
by Ricard (1986) and Underhill (1994, p. 202)] were helpful. Also, the majority of
learners state that their teacher asks them to record their own speech and encour-
ages them to watch TV programmes or listen to radio broadcasts in English. This
suggests that self-assessment, as well as self-reflection and independent study, are
being encouraged [advocated by Baker and Burri (2016) and Scarcella and Oxford
(1994, p. 227)]. This can only be beneficial for the provision of meaningful pronun-
ciation instruction. This self-reflection is also supported in some small way by the
learning journal that learners complete over the duration of the course. However,
because self-reflection in educational settings might not be familiar to many learners,
it could be facilitated by encouraging learners to discuss elements of their journal in
groups to ensure that they fill in the journal and further practice their speaking and
listening skills.
Learners also seem to display overwhelmingly positive attitudes towards instruc-
tion, stating that although they are confident, that they want to improve, and that
they approve of all types of error correction. They also believe that the lessons on
the course have been beneficial. One significant finding though, one which is similar
to the learners’ positive attitudes towards pronunciation instruction and its impor-
tance in Couper (2003) and Baker (2011), is that learners would generally like to
receive more instruction than they do. Also, as in Baker (2011), Derwing (2003),
Timmis (2002), and Waniek-Klimcza, Rojczyk & Porzuczek (2015), learners gener-
ally express a preference for native models of English. However, although learners
should clearly benefit from listening to a range of native varieties, they should also
be exposed to at least a few examples of L2 English, and more importantly, should
not be taught with the aim of getting them to sound like native speakers because
this may be counter-productive and largely unattainable (Derwing & Munro, 2015;
Deterding & Kirkpatrick, 2006; Gramley & Patzold, 2004; Levis, 2018).
Although there is a stated goal of intelligibility, during the staff induction, the CL
stated that the substitution of /θ/ with /s/ (for example, ‘I sink*’) was “a problem that
affects communication” and a problem that teachers should “aim to fix.” Bearing in
mind Brown’s (1991) and Munro and Derwing’s (2006) description of this particular
example of a segmental error as one that carries a relatively low functional load,
5.11 Discussion 185
it is difficult to see how such an error could have a particularly noticeable effect
on intelligibility and therefore why it should be prioritized. This focus on the same
relatively low functional load item was also apparent in the speaking assessment tasks
that staff completed during speaking standardization. The model marksheet that staff
consulted to guide them contained a rationale for the marks given. This rationale did
mention this item as an error. Again, this is something that seems difficult to reconcile
with intelligibility being the overriding goal.
During a more in-depth discussion of the CL’s beliefs in the semi-structured
interview regarding the substitutions of /θ/ and /ð/, the view of whether these substi-
tutions would affect intelligibility seemed unclear. At some points the view is that
it may interfere with intelligibility if it is accompanied by other errors, that the CL
personally does correct it when teaching, and that it is important to raise aware-
ness of it. However, on other occasions the CL points out that the same substitution
can occur with French speakers of English, and that it is not viewed as an error
in such cases. Also, when the matter of English accents which regularly substitute
the /θ/ was discussed, the CL argues that fing* (thing) is not as interfering as sing*
because sing is a different word. Although Brown (1991) does view the latter error
as one that has a higher functional load (5 on the 10 point scale), the substitution
is by no means one that carries a high functional load, and the CL’s view does not
seem to take into account the important point made by Jenkins (2000), Marks (2002,
p. 157), and Munro and Derwing (2006) that segmental errors often do not interfere
with intelligibility because the meaning of the utterance is clear due to context. The
important issue here is that instruction and the guidance on how instruction should be
provided should be based on evidence (Derwing & Munro, 2015; Levis, 2018), and
the evidence suggests that low functional load substitutions should not be a priority
(Brown, 1991; Jenkins, 2000; Munro & Derwing, 2006). There is an acceptance by
the CL that clearer explanations during staff induction and standardization of how
certain sounds can interfere with intelligibility would be helpful in ensuring that
more uniform instruction and assessment takes place. However, in light of the CL’s
rather equivocal responses, coupled with the fact that the only advice provided to
teaching staff in terms of pronunciation instruction focused on the dental fricative,
it is doubtful whether such advice would be based on research evidence.
In terms of assessment, although the research only involved three staff members, so it
is rather difficult to draw firm conclusions, the different marks awarded for pronun-
ciation by the three teachers for three student seminar performances are striking.
There is certainly much more variation in marking than in any of the other cate-
gories (grammar, vocabulary, and interactive communication). It seems that, as the
CL suggests, one of the reasons for this could be the course descriptors. Although,
as Harding (2017) points out, too many descriptors can cause unreliable marking,
too few descriptors may also be the cause of unreliable marking. Harding (2017)
186 5 Results, Analysis, and Discussion
recommends between three and five descriptors, and two descriptors seem insuf-
ficient, particularly when the differences between descriptors in different marking
bands are centred around word meanings that are difficult to define. Of course, as
Thornbury (2017) points out, marking pronunciation is highly subjective, and inter-
preting the meaning of words used in descriptors can be problematic. Teachers can
have different definitions or conceptualizations of key terms (Bohn & Hansen, 2017),
and some teachers may also mark non-standard pronunciation of low functional load
substitutions as errors due to their salience rather than their communicative value
(Levis, 2018, p. 29). Although Mark did note that one of the candidates did not
produce /θ/, but substituted it for /s/, he also stated that he could understand what the
candidate was saying. So for this particular candidate at least, it appears that Mark
evaluated this segmental error as one carrying a low functional load. However, it is
clear from the teacher questionnaire data that not all teachers share this view. It is
precisely for this reason that the discrimination of high and low functional load errors
should be explained fully to teachers and that the key terms used should be defined.
Furthermore, contrary to the CL’s suggestion, low functional load segmental errors
do not generally interfere with comprehensibility, even when they occur on multiple
occasions (Munro & Derwing, 2006). Also, it seems that the point noted by Bruce in
the interviews, that there is a difficulty in interpreting measures of frequency in the
marksheet (e.g., some/a few) is problematic, as is the lack of adherence to an intel-
ligibility measure in all descriptors (see Council of Europe, 2018; Piccardo, 2016,
p. 15).
Different interpretations of the roles of the four marking categories are also
apparent, with the CL and Olivia providing slightly different explanations of how they
view the pronunciation and interactive communication categories. Once again, clear
guidance on which particular aspects of language are being assessed in each cate-
gory would be useful. Another possibility would be to adopt the solution proposed by
Harding (2017) and combine the two overlapping categories into a delivery category.
All three of the observed teachers state that the fact that the presentation assess-
ment is purely summative in nature is problematic. The reason that the summa-
tive assessment of presentations no longer takes place is that the external examiner
suggested that the course had too many assessments. However, it may be the case
that because presentations no longer count towards learners’ overall final mark, they
are not taken so seriously by learners. My own action research confirmed the reduc-
tion in learner motivation reported by Bruce, Mark, and Olivia. All three teachers
believe that some form of summative assessment would prove beneficial. This is
significant because much of the pronunciation work on the course takes place during
the lessons devoted to presentations. Therefore, if learners are less well-motivated
5.11 Discussion 187
in these lessons, they may not participate fully and might not grasp the opportu-
nity to improve their pronunciation skills. Returning to the summative assessment of
presentations would obviously be a difficult organizational decision to make, but it
could well be beneficial in terms of improving motivation and pronunciation skills.
The important role of presentations in higher education, particularly on business
courses, is noted by Bruce during his interview, and it does seem that if presenta-
tions are examined, learners may be more motivated to participate and evaluate their
own video recorded performances, something that is recommended by both Heron
(2018, 2019) and Ritchie (2016). Such presentation skills can also be instrumental
in helping to promote learners’ oracy skills (Engin, 2017).
The listening activities employed could also be improved in other ways. There
are some listening materials from Powell (1996) related to chunking, sentence stress,
and intonation which can broadly be described as being similar to the listening
discrimination exercises advocated by Bowen and Marks (1992, p. 25) and Foster
(in The University of Sheffield, 2017). Of these activities, Bruce points out that
the sentence stress activity is particularly useful. However, there are few exercises
where learners group words or sentences according to stress patterns or locate stressed
syllables [suggested by Field (2005)]. Perhaps one of the less satisfactory aspects
of the listening exercises connected with pronunciation features in Powell (1996)
and the more general listening exercises is not only that they, with very few excep-
tions, present standard native models of English, but also that the language presented
is exactly of the sanitized and scripted type described by Shockey (2003, 2011)
and Wagner and Toth (2017). This was also noted by Olivia, who described the
extracts as consisting of a slow delivery of standard pronunciation. Furthermore,
during my own action research, I noticed that many of the videos were recordings
where learners could not see the speaker. When we consider the interaction situations
that learners will encounter on their destination courses, because the pre-sessional
course is so short, I believe it would be advantageous to have video recorded mate-
rials where learners can see the speaker in order to enable more scaffolding and
prepare learners for the authentic listening situations they will encounter in their
future academic settings (lectures, seminars, and presentations are all situations
where learners can see their interlocutors). By including videos where learners can
see the person communicating, the social dimension of interaction that is very much
at the centre of sociocultural theory (Vygotsky, 1962, 1978) could be included so that
learners could utilize non-verbal cues to aid comprehension. The general listening
activities on the course do not seem to be aimed at improving pronunciation. The CL
confirmed that the major aim of the general listening tasks is to improve note-taking:
an important skill that learners will need when they begin their courses. However, the
tasks offer learners little support in terms of utilizing pronunciation features to assist
this. Olivia is particularly critical of the listening, claiming that it does not support
pronunciation instruction, and the questionnaire data show that the vast majority of
the teachers disagree with the statement that the listening materials on the course
support pronunciation.
The listening pilot that took place, although intended as a pilot for the autumn
pre-sessional, further reinforced the emphasis on note-taking. However, one part of
the pilot task seemed unsatisfactory because learners were expected to take notes
on the listening before being provided with the questions in the subsequent section
of the test. This seems an unusual approach to the testing of listening and does not
appear to be compatible with the idea of purposive listening. Whatever the view on
the efficacy of such tasks, it is unlikely that they will serve to support pronunciation
instruction in any meaningful way.
190 5 Results, Analysis, and Discussion
In terms of actual practice, the observation and checklist data were useful in exam-
ining the different approaches taken to the instruction of suprasegmentals during
classes. In particular, how new vocabulary was introduced, and how error correction
took place did differ in the four classes (including my own). Firstly, the introduction of
new vocabulary sometimes did not involve learners articulating the new words. Some
of these words were challenging for learners, so this is perhaps a little surprising.
Also, if we accept that it is both desirable and beneficial to teach word stress (Council
of Europe, 2018; Dauer, 2005; Field, 2005; Levis, 2018; Piccardo, 2016; Waniek-
Klimczak, 2015), it is surprising that these vocabulary items are sometimes presented
without the provision of visual cues to assist learners in identifying the stress. Gener-
ally speaking, the use of non-verbal cues to assist instruction (advocated in Baker
& Burri, 2016; Loewen & Philp, 2006) was only apparent in one of the observed
classes (Olivia). In my action research in the current study, I attempted to incorporate
the recommendations of Baker and Burri (2016) and found the system of recording
word stress easy to incorporate into lessons and easily understood by learners. Also,
I found that by being more aware of how non-verbal communication can be used, I
was able to assist learners in identifying pronunciation features, such as word stress
or rising intonation. This non-verbal communication was also apparent in Olivia’s
lesson, where it was used to demonstrate rising intonation. Considering the impor-
tance of scaffolding and that learners may require assistance in bridging the gap
between what they already know and target production (Vygotsky, 1978), more use
of both non-verbal communication and the system of recording word stress could
be beneficial. More scaffolding could also be applied to listening activities too, for
example, by activating schema where the activities suggested in the syllabus appear
to be far beyond the level of comprehensible input.
Of course, one of the problems that can sometimes arise is that learners have
little or no prior knowledge of the subject area. In this regard, Bruce’s suggestion
that seminar topics might be generated by learners themselves, with appropriate
guidance, might be worth exploring further. This suggestion seems, in some small
way, to be consistent with the principles of DOGME lessons (Meddings & Thornbury,
2009; Thornbury, 2000), and such a syllabus change would have the advantage of
enhancing motivation and exploiting learners’ prior knowledge of subject areas. The
feasibility of implementing such learner-focused lessons is far from clear though,
particularly when learners come from cultural backgrounds that are more traditional
and less learner-centred. Nevertheless, it is certainly worthy of further consideration,
and further research is perhaps needed in this area.
In terms of practice, one of the problematic areas noted by myself and the three
observed teachers was the difficulty in trying to encourage learners to use the discus-
sion language worksheets authentically in the free seminar discussion phases of
speaking lessons. Some of the language worksheets contained useful information,
for example, regarding the rising intonation used when interrupting politely (can I
just say something here?), as well as functional language which can be utilized in
5.11 Discussion 191
free discussions. It was noticeable that, for example, even when Olivia modelled
this language and intonation successfully, learners seldom, if ever, went on to use
the language in free discussion. All three teachers commented on this difficulty of
transfer to free practice, and it is a problem that I also encountered. Once again,
one of the issues is that there is little time to consider these problems and imple-
ment solutions once the course is underway and teachers have to fulfil a whole range
of tasks. One possible way to facilitate transfer would be to incorporate a guided
practice phase involving a role play situation where learners have to use the target
language. Baker (2014, pp. 153–154) and Baker and Burri (2016) have pointed to
the importance of guided practice in pronunciation instruction, and supplementing
the discussion language worksheets with an appropriate role play task could be an
effective way of bridging the gap between the discussion language and its use in free
practice.
The modelling and drilling of discussion language, along with the appropriate
intonation, did take place in some of the observed lessons. However, as Cook (2008,
pp. 81–82) points out, for repetition to be successful, adequate feedback needs to be
provided. Some lessons, for example, the one previously mentioned, led by Olivia,
did contain feedback on intonation, and the feedback was provided specifically in
connection with the intonation, that is, it was of a form-focused nature advocated
by Baker and Burri (2016) and Saito and Lyster (2012). Hattie (2009) and Hattie
and Timperley (2007) point out that very general feedback rather than focussed
feedback, might not be so effective. Such targeted feedback was rarely evident in the
observed classes, and it was sometimes far from clear that learners were pronouncing
the target language in a sufficiently intelligible manner: There often seemed to be
a concentration on the language itself, without sufficient attention paid to whether
that language was produced intelligibly. A certain amount of flexibility is obviously
required: Teachers on the course are well qualified and experienced, so they should
be able to judge how appropriate the pronunciation materials are for their partic-
ular group of learners. For instance, Bruce concentrated on the definitions of new
vocabulary items and elected not to ask his learners to say the particular words,
presumably because his learners were more advanced than those in other classes and
would perhaps have more knowledge of word stress. It could also be the case that, as
in Derwing and Rossiter (2002), the challenge of attempting to provide communica-
tive instruction to a multilingual learner group, combined with a lack of guidance
on how pronunciation instruction can be incorporated into such lessons led to a lack
of focus on pronunciation instruction. Of course, it is also perfectly feasible that
because the learner group was more advanced than other groups, this focus was not
deemed necessary as learners were generally intelligible. In general terms though,
clearer form-focused feedback would be beneficial for learners.
Checklist data show a roughly even split between error correction reported on an
individual level or in small groups, and correction in the whole class. Observations
show that Bruce tends to favour correction in small groups, but as he himself points
out, it can sometimes be difficult to hear and correct important errors if they take place
at the group level because of a lack of time. However, as Bruce also points out, drilling
particular words or phrases in the whole class may sometimes be counter-productive
192 5 Results, Analysis, and Discussion
The pronunciation errors that were corrected, according to the checklist data, suggest
that there is a firm emphasis on suprasegmentals, at least in terms of error correction.
The three most common, in order of frequency, were intonation, word stress, and
sentence stress. It is interesting that word stress was corrected more than sentence
stress. Perhaps, as Mark suggests, this is because this feature can be easily assimilated
with other aspects of language learning, such as grammar and vocabulary. Field
(2005) describes word stress as a moderately important feature of pronunciation, yet
the teacher questionnaire reveals it as the one that teachers on the course believe is
the most important. This may be because teachers find it easier to teach, or perhaps
because they have had more experience of dealing with word stress on teacher training
courses. However, it is certainly true that teachers seem to broadly agree with Dauer
(2005), Field (2005), and Levis (2018) that word stress is a suprasegmental feature
worthy of instruction time.
The segmental corrections from checklist data reflect the different needs of the
multilingual class taught by Bruce. For example, Bruce reports the substitution of /p/
for /b/ by the Arabic speakers in his class, an error that is specific to his class because
it is the only one that comprises learners other than Chinese L1 speakers. Bruce’s
awareness of such errors might also show how knowledge of contrastive analysis
(Lado, 1957) or error analysis (Corder, 1967), presumably gained during TESOL
diploma studies, could be useful in increasing teacher awareness of certain errors
that might occur with speakers of particular languages. This raises an interesting
5.11 Discussion 193
point concerning Chinese learners and the guidance provided regarding common
errors that might be specific to them. Although the pre-sessional course has existed
for several years, and has usually consisted of predominantly Chinese speakers, little
or no attention has been paid to errors that may typically occur with this group of
learners. The only mention of specific segmental errors is the substitution of the /θ/
with /s/ that the CL touched upon during induction. As mentioned earlier, if particular
segmental errors that typically occur with Chinese learners are raised, it does not seem
to be productive to concentrate on segmentals that have a relatively low functional
load. However, in terms of error correction, the checklists show that instructors do
not generally share the view that this area is something that “can be fixed” because
there is only one instance of the correction of this sound. Nonetheless, it may not
be the best course of action, and seems inconsistent with having intelligibility as the
goal of instruction, to concentrate on this relatively low functional load segmental
substitution, bearing in mind the evidence that suprasegmentals can have a greater
effect on intelligibility (Derwing et al., 1997; Derwing & Rossiter, 2003; Hahn, 2004),
and that adherence to native speaker production as a goal can be counter-productive
(Derwing & Munro, 2005, 2011, p. 317; Wagner & Toth, 2017, p. 87).
References
Chan, M. J. (2018). Embodied pronunciation learning: Research and practice. The CATESOL
Journal 30(1), 47–68. Retrieved 28/11/ 2019 from https://files.eric.ed.gov/fulltext/EJ1174234.
pdf.
Cook, V. J. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33(2),
185–209. https://doi.org/10.2307/3587717
Cook, V. J. (2008). Second language learning and language teaching (4th ed.). London: Hodder
Education.
Corder, S. P. (1967). The significance of learners’ errors. International Review of Applied Linguistics,
5(4), 161–170. https://doi.org/10.1515/iral.1967.5.1-4.161
Council of Europe (2018). Common European Framework of Reference for Languages: Learning,
teaching, assessment: Companion volume with new descriptors. Retrieved February 13, 2019,
from https://rm.coe.int/cefr-companion-volume-with-new-descriptors-2018/1680787989.
Couper, G. (2003). The value of an explicit pronunciation syllabus in ESOL teaching. Prospect,
18(3), 53–70. Retrieved January, 28, 2019 from https://openrepository.aut.ac.nz/bitstream/han
dle/10292/1524/18_3_4_Couper.pdf?
Couper, G. (2017). Teacher cognition of pronunciation teaching: Teachers’ concerns and issues.
TESOL Quarterly, 51(4), 820–843. https://doi:10.1002/tesq.354
Dauer, R. M. (2005). The Lingua Franca Core: A new model for pronunciation instruction? TESOL
Quaterly, 39(3), 543–550. https://doi.org/10.2307/3588494
Derwing, T. M. (2003). What do ESL students say about their accents? Canadian Modern Language
Review, 59(4), 547–566. https://doi.org/10.3138/cmlr.59.4.547
Derwing, T. M., Diepenbrook, L. G., & Foote, J. A. (2012). How well do general skills ESL textbooks
address pronunciation? TESL Canada Journal, 30(1), 22–44. https://doi.org/10.18806/tesl.v30i1.
1124new-descriptors-2018/1680787989
Derwing, T. M., & Munro, M. J. (2005). Second language accent and pronunciation teaching: A
research-based approach. TESOL Quarterly, 39(3), 379–398. https://doi.org/10.2307/3588486
Derwing, T. M., & Munro, M. J. (2011). The foundations of accent and intelligibility in pronunciation
research. Language Teaching, 44(3), 316–327. https://doi.org/10.1017/s0261444811000103
Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidence-based perspectives
for L2 teaching and research. Amsterdam: John Benjamins.
Derwing, T. M., Munro, M. J., & Wiebe, G. (1997). Pronunciation instruction for ‘fossilized
learners’: Can it help? Applied Language Learning, 8(2), 217–235. https://eric.ed.gov/?id=EJ5
67518.
Derwing, T. M., & Rossiter, M. J. (2002). ESL learners’ perceptions of their pronunciation needs
and strategies. Science Direct, 30(2), 155–166. https://doi.org/10.1016/S0346-251X(02)00012-X
Derwing, T. M., & Rossiter, M. J. (2003). The effects of pronunciation instruction on the accu-
racy, fluency and complexity of L2 accented speech. Applied Language Learning, 13(1), 1–
17. https://www.researchgate.net/publication/234570703_The_Effects_of_Pronunciation_Instru
ction_on_the_Accuracy_Fluency_and_Complexity_of_L2_Accented_Speech.
Deterding, D. (2005). Listening to Estuary English in Singapore. TESOL Quarterly, 39(3), 425–440.
https://doi.org/10.2307/3588488
Deterding, D., & Kirkpatrick. (2006). Emerging South-East Asian Englishes and intelligibility.
World Englishes, 25(3), 391–409. https://doi.org/10.1111/j.1467-971X.2006.00478.x
Engin, M. (2017). Contributions and silence in academic talk: Exploring learner experiences of
dialogic interaction. Learning, Culture and Social Interaction, 12, 78–86. https://doi.org/10.1016/
j.lcsi.2016.11.001
Field, J. (2005). Intelligibility and the listener: The role of lexical stress. TESOL Quarterly, 39(3),
399–423. https://doi.org/10.2307/3588487
Foote, J. A., Holtby, A. K., & Derwing, T. M. (2011). Survey of the teaching of pronunciation in
adult ESL programs in Canada, 2010. TESL Canada Journal, 29(1), 1–22. https://doi.org/10.
18806/tesl.v29i1.1086
References 195
Lee, W. R. (1980). A point about the rise-endings and fall-endings of yes-no questions. In L.
R. Waugh & C. H. van Schooneveld (Eds.), The melody of language intonation and prosody
(pp. 165–168). Baltimore: University Park Press.
Levis, J. M. (1999a). The intonation and meaning of normal yes/no questions. World Englishes,
18(3), 373–380. https://doi.org/10.1111/1467-971X.00150
Levis, J. M. (1999b). Intonation in theory and practice, revisited. TESOL Quarterly, 33(1), 37–63.
https://doi.org/10.2307/3588190
Levis, J. M. (2005). Changing contexts and shifting paradigms in pronunciation teaching. TESOL
Quarterly, 39(3), 369–378. https://doi.org/10.2307/3588485
Levis, J. M. (2018). Intelligibility, oral communication, and the teaching of pronunciation.
Cambridge, UK & New York, NY: Cambridge University Press.
Lewis, M. (1993). The lexical approach. Hove: Language teaching publications.
Lewis, M. (2002). Implementing the lexical approach: Putting theory into practice (2nd ed.). Boston:
Heinle.
Liu, M., & Jackson, J. (2008). An exploration of Chinese EFL learners’ unwillingness to commu-
nicate and foreign language anxiety. The Modern Language Journal, 92(1), 71–86. https://doi.
org/10.1111/j.1540-4781.2008.00687.x
Loewen, S., & Philp, U. (2006). Recasts in the adult English L2 classroom: Characteristics, explicit-
ness, and effectiveness. The Modern Language Journal, 90(4), 536–556. https://doi.org/10.1111/
j.1540-4781.2006.00465.x
Macdonald, S. (2002). Pronunciation-views and practices of reluctant teachers. Prospect, 17(3),
3–16. https://hdl.handle.net/1959.14/323083
Marks, J. (2002). Pronunciation in the lexical approach. In M. Lewis (Ed.), Implementing the lexical
approach (pp. 156–162). Boston: Heinle.
Matsuda, A. (2019). World Englishes in English language teaching: Kachru’s six fallacies and the
TEIL paradigm. World Englishes, 38(1–2), 144–154. https://doi.org/10.1111/weng.12368
Meddings, L., & Thornbury, S. (2009). Teaching unplugged: Dogme in English language teaching.
Peaslake: Delta Publishing.
Munro, M. J., & Derwing, T. M. (2006). The functional load principle in ESL pronunciation
instruction: An exploratory study. System, 34(4), 520–531. https://doi.org/10.1016/j.system.2006.
09.004
Patsko, L. (2014, March 3). Identifying and practising thought groups [Blog post]. Retrieved January
29, 2019, from https://elfpron.wordpress.com/2014/03/03/identifying-and-practising-tone-units/
Phillipson, R. (2016). Native speakers in linguistic imperialism. Journal for Critical Education
Policy Studies, 14(3), 80–96.
Piccardo, E. (2016). Council of Europe: Common European Framework of Reference for Languages:
Learning, teaching, assessment phonological scale revision process report. Retrieved from
Council of Europe: https://rm.coe.int/phonological-scale-revision-process-report-cefr/168073
fff9
Powell, M. (1996). Presenting in English: How to give successful presentations. Hove: Language
Teaching Publications.
Ricard, E. (1986). Beyond fossilization: A course on strategies and techniques in pronunciation for
adult learners. TESL Canada Journal, Special Issue 1, 243–249. https://doi.org/10.18806/tesl.
v3i0.1009
Ritchie, S. M. (2016). Self-assessment of video-recorded presentations: Does it improve skills?
Active Learning in Higher Education, 17(3), 207–221. https://doi.org/10.1177/146978741665
4807
Roach, P. (1983). English phonetics and phonology. Cambridge: Cambridge University Press.
Saito, K., & Lyster, R. (2012). Effects of form-focused instruction and corrective feedback on L2
pronunciation development of/ô/ by Japanese learners of English. Language Learning, 62(2),
595–633. https://doi.org/10.1111/j.1467-9922.2011.00639.x
Scarcella, R. C., & Oxford, R. L. (1994). Second language pronunciation: State of the art instruction.
System, 22(2), 221–230. https://doi.org/10.1016/0346-251X(94)90058-2
References 197
Schumann, J. H. (1978). The Acculturation Model for second language acquisition. In R. C. Gingras
(Ed.), Second language acquisition and foreign language teaching (pp. 27–50). Arlinton, VA:
Center for Applied Linguistics.
Shockey, L. (2003). Sound patterns of spoken English. Oxford: Blackwell.
Shockey, L. (2011). Understanding L2 and the perspicacious Pole. In J. Arabski & A. Wojtaszek
(Eds.), The acquisition of L2 phonology (pp. 27–36). Bristol: Multilingual Matters.
Sifakis, N. C., & Sougari, A.-M. (2005). Pronunciation issues and EIL pedagogy in the periphery:
A survey of Greek state school teachers’ beliefs. TESOL Quarterly, 39(3), 467–488. https://doi.
org/10.2307/3588490
Sonsaat, S. (2017). The influence of an online pronunciation teacher’s manual on teachers’ cogni-
tions. (Doctoral dissertation, Iowa State University, US). https://doi.org/10.31274/etd-180810-
5849
Stevick, E. W. (1980). Teaching languages: A way and ways. Rowley, MA: Newbury House.
The University of Sheffield. (2017, July 12). Ten practical ideas for teaching pronunciation in EAP
[Video webinar]. Retrieved 28/01/2019 from https://uni-sheffield.adobeconnect.com/pno7lgb00
vgh/?launcher=false&fcsContent=true&pbMode=normal
Thomson, R. I. (2018). High variability [pronunciation] training (HVPT): A proven technique
about which every language teacher and learner ought to know. Journal of Second Language
Pronunciation, 4(2), 208–231. https://doi.org/10.1075/jslp.17038.tho
Thornbury, S. (2000). A Dogma for EFL. IATEFL Issues, 153(2). Retrieved January 28,
2019, from https://esol.britishcouncil.org/sites/default/files/attachments/informational-page/Ado
gmaforEFL.pdf
Thornbury, S. (2010, December 5). R is for repetition [Blog post]. Retrieved from https://scottthor
nbury.wordpress.com/2010/12/05/r-is-for-repetition/
Thornbury, S. (2017, May 28). An A-Z of ELT: I is for intelligibility [Blog post]. Retrieved January
28, 2019, from https://scottthornbury.wordpress.com/2017/05/28/i-is-for-intelligibility/
Timmis, I. (2002). Native-speaker norms and international English: A classroom view. ELT Journal,
56(3), 240–249. https://doi.org/10.1093/elt/56.3.240
Underhill, A. (1994). Sound foundations: Learning and teaching pronunciation. Oxford: Macmillan.
Vygotsky, L. S. (1962). Thought and language. Cambridge MA: MITT Press.
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes.
Cambridge: Cambridge University Press.
Wagner, E., & Toth, P. D. (2017). The role of pronunciation in the assessment of second language
listening ability. In T. Isaacs & P. Trofimovich (Eds.), Second language pronunciation assessment
(pp. 72–92). Bristol: Multilingual Matters.
Waniek-Klimczak, E. (2015). Factors affecting word stress recognition by advanced Polish learners
of English. In E. Waniek-Klimczak & M. Pawlak (Eds.), Teaching and researching the pronuncia-
tion of English studies in honour of Włodzimierz Sobkowiak (pp. 189–204). Switzerland: Springer
International Publishing.
Waniek-Klimczak, E., Rojczyk, A., & Porzuczek, A. (2015). Polglish’ in Polish eyes: What English
studies majors think about their pronunciation in English. In E. Waniek-Klimczak & M. Pawlak
(Eds.), Teaching and researching the pronunciation of English studies in honour of Włodzimierz
Sobkowiak (pp. 23–24). Switzerland: Springer International Publishing.
Wong, R. (1987). Learner variables and prepronunciation considerations in teaching pronunciation.
In J. Morley (Ed.), Current perspectives on pronunciation: Practices anchored in theory (pp. 13–
28). Alexandria, VA: Teachers of English to Speakers of Other Languages.
Chapter 6
Conclusion and Recommendations
for Future Practice
6.1 Conclusion
This research project set out to discover the pronunciation goals on a ten-week
pre-sessional EAP course in the UK and the extent to which summative assess-
ment, teacher instruction, attitudes, and beliefs reflect these goals. Learner attitudes
towards pronunciation instruction were also explored. The mixed methods approach
that was applied at various stages of the course revealed a number of key find-
ings. What follows is a summary of the research, followed by recommendations for
pronunciation instruction on future pre-sessional EAP courses.
Firstly, although suprasegmental instruction accounted for almost the entire
content of the pronunciation element of the course, the goals of the course were
not clearly evident in course documents or clearly provided in the guidance given
to teachers. The issue of whether intelligibility or native speaker production was the
overriding target of instruction was far from clear. In particular, almost all listening
materials suggested in the syllabus consisted of either SAE or SSBE accented speech,
much of which was of a scripted, unnatural nature. The fact that the local accented
speech of the area surrounding University A is of a Northern English type suggests
that the models provided may not be the most effective in improving learner intelli-
gibility of the Northern English pronunciation they are exposed to outside the class-
room. It is also not effective in helping to improve learner intelligibility of non-native
English speech.
The issue of whether low functional load substitutions should be viewed as errors is
another example of the less than clear course goals. The equivocal views of the course
leader, the lack of clearly stated goals in course documents and during staff induction
and training sessions, and the guidance provided during marking standardization
sessions combine to provide goals that seem rather confusing. Definitions of what
intelligibility could actually mean in practical terms are generally absent, so it is left
to teachers to provide their own definitions of these key concepts.
The mixed methods approach I have employed enabled me to gain an overall view
of the course goals, along with a snapshot of teacher practices and information
on teacher beliefs in terms of pronunciation instruction, particularly with regard to
suprasegmental instruction and the goals of instruction. What follows are tentative
recommendations for similar courses of EAP instruction.
The first recommendation is that the pronunciation goals of the institution
providing instruction should be explicitly stated and clearly presented to teachers.
Because of the status of English as a Lingua Franca, learners who go on to use
English on their academic courses will almost inevitably interact with other non-
native speakers of English. Intelligibility should therefore be the goal of instruction
rather than native-like production. Although this may seem a rather obvious recom-
mendation, institutions should be aware that merely stating that the goal of instruction
is intelligibility without examples might lead to misunderstandings and confusion.
Examples of segmental errors that do not fundamentally interfere with intelligibility
should therefore be provided (e.g., when /θ/ and /ð/ are substituted with /s/ and /z/).
There should be an awareness that is communicated to teachers that contextual clues
might mean that low functional load segmentals will have little impact on intelligi-
bility or comprehensibility (see Munro & Derwing, 2006). Because of the evidence
of the crucial role suprasegmental instruction can play, even over a relatively short
period of time (Derwing et al., 1997, 1998; Derwing & Rossiter, 2003; Hahn, 2004),
an explicit statement of the primacy of suprasegmental instruction, also reflected
in syllabus content, is desirable. Because learners are likely to encounter a variety
of English accents during their degree courses, pre-sessional courses should also
endeavour to provide clear listening goals and both native and non-native listening
materials that reflect this variety (Deterding, 2005; Jenkins, 2000; Thomson, 2018).
Including some examples of the local native accent in listening extracts would
probably be beneficial in helping learners comprehend people in the local area where
they will be based for at least a year, and hopefully assist their assimilation to the
new surroundings. Also, with the work of Shockey in mind (2003, 2011), examples
of natural speech should be included in listening extracts because this will often be
encountered by learners. This could also be reinforced by universities attempting
to encourage interactions between the learners and the local population as there is
evidence that such interaction can improve learner fluency (Derwing et al., 2006).
Derwing et al. (2008) suggest that this can be facilitated by providing learners with
small talk language so that learners will feel more at ease with simple interactions
with the local population. Of course, the scope for this interaction to take place on
short courses of instruction is limited, but perhaps there is room for improvement
in this area. The conversation club initiative that usually takes place at University A
once learners have been accepted on their destination courses is certainly a laudable
one that could be beneficial on many EAP courses. Bearing in mind the fact that
the vast majority of learners on the course at University A were in the UK for the
202 6 Conclusion and Recommendations for Future Practice
first time, a situation that may well be replicated on many UK pre-sessional courses,
efforts should be made to encourage interaction between the learners and the local
population, as well as other learners at the university.
In the case of University A, a further obstacle to the transmission of course goals
was that no one person was responsible for both the administration of the course
and the syllabus. This might have made it more complicated for the course leader
to transmit these goals. Course leaders should therefore have direct input into the
syllabus and the goals of the course, or at least be fully aware of what the goals of
instruction are. This would also make course leaders accountable for syllabus input.
In terms of testing, considering the work of Saito et al. (2017) and Bohn and
Hansen (2017), and also bearing in mind the variations present in the teachers’
seminar grading, course administrators should make it clear to markers that the
possession of a native accent is not a key consideration, and reinforce intelligibility
as the key assessment criteria. CEFR descriptors (Council of Europe, 2018) could be
utilized as a basis for assessment, containing intelligibility-based criteria across the
scales, and bearing in mind Harding’s suggestion (2017), marksheets should have
between three and five descriptors. Although the descriptors should contain both
segmental and suprasegmental elements, it is my view that, because of the value of
suprasegmental instruction in terms of improving learner intelligibility over relatively
short courses of instruction, the primacy of suprasegmentals should be apparent, with
the number of suprasegmental descriptors exceeding the segmental ones. Because
accentedness is not so important for intelligibility (Derwing & Munro, 2015), intel-
ligibility (across all scales) should be the focus in the descriptors, with any reference
to foreign or native accent removed. As Harding (2017) states, clarity is important
when using terminology in the descriptors, and markers should be provided with
precise definitions and explanations of the key terms in order to facilitate standard
and accurate marking. In particular, where segmentals are referred to in descriptors,
they should only be considered in terms of how segmental errors affect intelligibility,
and this should be made clear to the teacher markers. Comprehensibility measures
should also be employed, but the measure of intelligibility across all rating scales is
vital.
It could also be useful to implement some pedagogical suggestions emanating
from pronunciation research in EAP. In particular, feedback on pronunciation should
be of a specific rather than general nature (Hattie, 2009; Hattie & Timperley, 2007).
Teachers should therefore be made aware that general comments might not be the
most effective way of encouraging accurate and intelligible production. An awareness
that learner uptake of target features can be improved by accompanying this targeted
feedback with gestures and non-verbal cues could also be advantageous (Baker &
Burri, 2016; Foster, in The University of Sheffield, 2017).
The instruction of word stress could be facilitated by using the Baker and Burri
(2016) system of notation. This would enable learners to not only recognize particular
word stress patterns, but also to keep a record of them when they encounter new
vocabulary, and therefore assist independent learning. If learners become accustomed
to using such notation at the beginning of the course of instruction, it could be
beneficial, for example, when they use new vocabulary in their subject area when
6.2 Recommendations for Pronunciation Instruction on Future … 203
preparing presentations for the pre-sessional course, and more importantly, during
their future studies. During my own action research, I found that using the system
of notation and using particular hand movements to show primary stress location
was easy to incorporate into lessons, and seemed to be beneficial, with relatively low
level learners (B1) finding the notation system easy to grasp.
The encouragement of self-assessment and peer assessment can be important in
fostering independent learning and student-centred lessons: important considera-
tions for pre-sessional courses. However, sufficient scaffolding should be provided
to facilitate both types of assessment (Dlaska & Krekeler, 2008). This scaffolding
can include learners recording their own production on smartphones and analyzing
certain suprasegmental elements, such as nuclear stress and chunking (Celce-Murcia
et al. 1996). These features can then be analyzed through learners rehearsing parts
of a model presentation and then incorporating them into their own presentations
(Goodwin, 2001, 2005). The scaffolding of peer feedback can also be enabled by a
careful elicitation from learners of an evaluation of the target pronunciation that has
been produced by their peers. For learners from certain cultures, it may prove difficult
to persuade them to evaluate another student’s language production, but encourage-
ment to do so should be provided, perhaps by using checklists as a means of focusing
attention on particular suprasegmentals. Such checklists could also perhaps be used
when students are evaluating suprasegmental elements of their language if they record
their own production. Of course, teachers need to guide learners through this process
in a sensitive manner by making suggestions, asking questions, and giving encour-
agement. Also, the checklists themselves should be clear for learners, without being
unnecessarily vague (e.g., the term good pronunciation is far too general).
Scaffolding can be provided in other ways too, particularly in terms of chunking.
Attempts should be made to provide a context for the particular chunking by providing
a short comprehension activity before the listening activity (Brazil, 1994, p. 3). In
this way, learners will be able to comprehend the extract, and the chunking activity
will be given a context. There may be other pronunciation materials too, which
course administrators might adapt in order to provide a communicative context so
that learners can more easily become aware of the communicative function of the
suprasegmentals they are being introduced to.
Finally, and bearing in mind the above suggestions on how suprasegmental instruc-
tion can be better implemented, the clear desire for more teacher training apparent
in the group of teachers involved in the study may well be present on other EAP
pre-sessional courses. The assumption that because teachers are well-qualified and
experienced, they are fully equipped to provide suprasegmental instruction in the
most effective way possible may not hold true because of a number of factors.
Firstly, some teachers on EAP courses might be experienced in the EFL sector,
but not in the EAP sector. This could mean that they may experience difficulties in
trying to provide pronunciation instruction in this new context, particularly if their
previous experience was not with adult learner groups and largely involved segmen-
tals. Secondly, much research into suprasegmental instruction is relatively recent,
so many teachers might have completed their training at a time before this research
was widely known. Finally, pronunciation instruction might only comprise a small
204 6 Conclusion and Recommendations for Future Practice
element in teachers’ diploma and MA level studies. Bearing in mind that most learners
in the study expressed a desire for more pronunciation instruction, and that teachers
in the study mentioned a lack of time in which to provide instruction, administrators
of pre-sessional programmes could endeavour to devote more time to instruction, and
if that is not possible, search for new ways to combine suprasegmental instruction
with other teaching materials. This is particularly important given the fact that many
learners on EAP programmes in the UK are Chinese speakers of English; Derwing
et al. (2006, 2008) illustrate the importance of explicit pronunciation instruction for
Chinese learners of English.
On many, if not all EAP courses, learners are taught how to give a short presen-
tation. Aside from the fact that presenting is a skill that is often required in higher
education (Heron, 2018, 2019; Ritchie, 2016), presentation-focused lessons may
be used to foster learner confidence: an important asset in learners’ future educa-
tional and employment careers. Such presentation lessons can also allow learners to
further practice the application of chunking, sentence stress, and intonation in their
spoken output, and also improve their oracy skills in academic settings (Engin, 2017).
An added bonus is that, if at least some learners consent to being video recorded,
their performances can be recorded and used for analysis by future learners. Even
if learners do not consent to their presentations being filmed, they may feel more
comfortable about their performance being recorded for their own use only. They
may then analyze and make improvements on their performance in their own time
outside the classroom. However, consideration should be paid to the positive wash-
back effects of assessment. If presentations are the subject of summative assessment,
it seems more likely that learners will be motivated to study, and therefore more likely
to acquire the suprasegmental features that will improve intelligibility.
It should be stressed that the above recommendations are tentative in nature and
that further research into suprasegmental instruction, particularly involving class-
room observation, would be illuminating, as would research that involves close coop-
eration with course leaders and administrators. Also, qualitative research into learner
viewpoints, involving in-depth interviews conducted in their L1 so that they are not
constrained linguistically, could be a valuable way of gaining a valid, in-depth insight
into what learners think about their instruction. Longitudinal follow-up interviews
could also be explored.
References
Baker, A., & Burri, M. (2016). Feedback on second language pronunciation: A case study of EAP
teachers’ beliefs and practices. Australian Journal of Teacher Education, 41(6), 1–19. https://doi.
org/10.14221/ajte.2016v41n6.1
Bohn, H., & Hansen, T. (2017). Assessing pronunciation in an EFL context: Teachers’ orientations
towards nativeness and intelligibility. Language Assessment Quarterly, 14(1), 54–68. https://doi.
org/10.1080/15434303.2016.1256407
Brazil, D. (1994). Pronunciation for advanced learners of English (Teacher’s book). Cambridge:
Cambridge University Press.
References 205
Celce-Murcia, M., Brinton, D., & Goodwin, J. (1996). Pronunciation: A reference for teachers of
English as a foreign language. New York: Cambridge University Press.
Council of Europe (2018). Common European framework of reference for languages: Learning,
teaching, assessment: companion volume with new descriptors. Retrieved February 13, 2019,
from https://rm.coe.int/cefr-companion-volume-with-new-descriptors-2018/1680787989.
Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidence-based perspectives
for L2 teaching and research. Amsterdam: John Benjamins.
Derwing, T. M., Munro, M. J., & Thomson, R. I. (2008). A longitudinal study of ESL learners’
fluency and comprehensibility development. Applied Linguistics, 29(3), 359–380. https://doi.org/
10.1093/applin/amm041
Derwing, T. M., Munro, M. J., & Wiebe, G. (1997). Pronunciation instruction for ‘fossilized
learners’: Can it help? Applied Language Learning, 8(2), 217–235. https://eric.ed.gov/?id=EJ5
67518.
Derwing, T. M., Munro, M. J., & Wiebe, G. (1998). Evidence in favour of a broad framework for
pronunciation instruction. Language Learning, 48(3), 393–410. https://doi.org/10.1111/0023-
8333.00047
Derwing, T. M., & Rossiter, M. J. (2003). The effects of pronunciation instruction on the accu-
racy, fluency and complexity of L2 accented speech. Applied Language Learning, 13(1), 1–
17. https://www.researchgate.net/publication/234570703_The_Effects_of_Pronunciation_Instru
ction_on_the_Accuracy_Fluency_and_Complexity_of_L2_Accented_Speech.
Derwing, T. M., Thomson, R. I., & Munro, M. J. (2006). English pronunciation and fluency in
Mandarin and Slavic speakers. System, 34(2), 183–193. https://doi.org/10.1016/j.system.2006.
01.005
Deterding, D. (2005). Listening to Estuary English in Singapore. TESOL Quarterly, 39(3), 425–440.
https://doi.org/10.2307/3588488
Dlaska, A., & Krekeler, C. (2008). Self-assessment of pronunciation. System, 36(4), 506–516.
https://doi.org/10.1016/j.system.2008.03.003
Engin, M. (2017). Contributions and silence in academic talk: Exploring learner experiences of
dialogic interaction. Learning, Culture and Social Interaction, 12, 78–86. https://doi.org/10.1016/
j.lcsi.2016.11.001
Goodwin, J. (2001). Teaching pronunciation. In M. Celce-Murcia (Ed.), Teaching English as a
second or foreign language (3rd ed., pp. 117–137). Boston, MA: Heinle & Heinle.
Goodwin, J. (2005). The power of context in teaching pronunciation. In J. Frodeson & C. Holten
(Eds.), The power of context in language teaching and learning (pp. 225–236). Boston, MA:
Thomson Heinle.
Hahn, L. D. (2004). Primary stress and intelligibility: Research to motivate the teaching of
suprasegmentals. TESOL Quarterly, 38(2), 201. https://doi.org/10.2307/3588378
Harding, L. (2017). What do raters need in a pronunciation scale? The user’s view. In T. Isaacs & P.
Trofimovich (Eds.), Second language pronunciation assessment (pp. 12–34). Bristol: Multilingual
Matters.
Hattie, J. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to achievement.
Florence, US: Routledge.
Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1),
81–112. https://doi.org/10.3102/003465430298487
Heron, M. (2018). Communicating in a business-like way: Developing oracy skills in higher educa-
tion [Power Point slides]. Retrieved January 29, 2019, from https://www.baleap.org/event/spe
aking.
Heron, M. (2019). Making the case for oracy skills in higher education: Practices and opportunities.
Journal of University Teaching & Learning Practice, 16(2). Retrieved from https://ro.uow.edu.
au/jutlp/vol16/iss2/9.
Jenkins, J. (2000). The phonology of English as an international language. Oxford: Oxford
University Press.
206 6 Conclusion and Recommendations for Future Practice
Munro, M. J., & Derwing, T. M. (2006). The functional load principle in ESL pronunciation
instruction: An exploratory study. System, 34(4), 520–531. https://doi.org/10.1016/j.system.2006.
09.004
Ritchie, S. M. (2016). Self-assessment of video-recorded presentations: Does it improve skills?
Active Learning in Higher Education, 17(3), 207–221. https://doi.org/10.1177/146978741665
4807
Saito, K., Trofimovich, P., Isaacs, T., & Webb, S. (2017). Re-examining phonological and lexical
correlates of second language comprehensibility: The Role of rater experience. In T. Isaacs
& P. Trofimovich (Eds.), Second language pronunciation assessment (pp. 141–156). Bristol:
Multilingual Matters.
Shockey, L. (2003). Sound patterns of spoken English. Oxford: Blackwell.
Shockey, L. (2011). Understanding L2 and the perspicacious Pole. In J. Arabski & A. Wojtaszek
(Eds.), The acquisition of L2 phonology (pp. 27–36). Bristol: Multilingual Matters.
The University of Sheffield. (2017, July 12). BALEAP Webinar: Ten practical ideas for teaching
pronunciation in EAP [Video webinar]. Retrieved January 28, 2019, from https://uni-sheffield.
adobeconnect.com/pno7lgb00vgh/.
Thomson, R. I. (2018). High variability [pronunciation] training (HVPT): A proven technique
about which every language teacher and learner ought to know. Journal of Second Language
Pronunciation, 4(2), 208–231. https://doi.org/10.1075/jslp.17038.tho
Appendix A
Seminar Marksheet
(continued)
Pronunciation Grammar Vocabulary Interactive
communication
69–60% • Tempo, rhythm • A wide range of • Vocabulary range • Can communicate
and intonation structures used sufficient to allow all ideas and
generally with some some flexibility opinions with
consistent and flexibility • Occasional word very few
communication • Error free choice/form difficulties
not affected utterances errors • Uses a range of
• Few frequent interactive
mispronounced • Errors cause no strategies
words but not strain effectively and
affecting with some
communication success to
maintain
discussion and
encourage others
to contribute
• Contributions
always
appropriate to
conversation flow,
showing a good
range of ideas &
flexibility
59–50% • Tempo, rhythm • A range of simple • Vocabulary range • Can communicate
and intonation and complex sufficient all ideas and
usually consistent structures used although lacking opinions with few
but with some a little in difficulties
communication flexibility but flexibility • Uses interactive
may be affected with some • A few word strategies to some
by inaccuracy repetition choice/form good effect to
• Some • Error free errors contribute and
mispronounced utterances encourage other
words may affect common contributions
communication • Errors may cause • Contributions
strain nearly always
appropriate to
conversation flow,
range of ideas a
little narrow or
lacking in
flexibility
(continued)
Appendix A: Seminar Marksheet 209
(continued)
Pronunciation Grammar Vocabulary Interactive
communication
49–40% • Inconsistency or • A range of simple • Vocabulary range • Can communicate
inaccuracy in and complex adequate but a main ideas and
tempo, rhythm structures used little repetitive opinions despite
and intonation but with some and inflexible some difficulties
cause a little repetition and • Some word • Uses interactive
strain inflexibility choice/form strategies with
• Mispronunciation • Some error free errors varying degrees
of words or utterances of success to
longer stretches • Errors in complex contribute and
can cause a little structures remain occasionally to
strain and can cause a encourage others
little strain • Contributions
usually
appropriate to
conversation flow
and context but
sometimes a little
simplistic and
lacking in
flexibility
39–30% • Inconsistences • A limited range • Vocabulary range • Can communicate
and inaccuracy in of simple and adequate but some ideas and
tempo, rhythm complex tends to be opinions but with
and intonation structures repetitive and some difficulties
cause strain • Few error free inflexible which may cause
• Mispronunciation utterances • Word choice/form a few problems
of words or • Errors in complex errors occur for others
longer stretches structures cause sometimes • Uses some
cause strain strain interactive
strategies
• Contributions
sometimes
appropriate to
conversation flow
or to context but
often too
simplistic and
lacking in
flexibility
(continued)
210 Appendix A: Seminar Marksheet
(continued)
Pronunciation Grammar Vocabulary Interactive
communication
29–20% • Some attempt at • Rather narrow • Vocabulary barely • Frequent
appropriate range of mainly adequate and difficulties
tempo, rhythm simple structures rather repetitive contributing
and intonation • Complex and inflexible opinions, even
• Mispronunciation structures • Word choice/form when asked
of words and attempted but errors frequent directly
longer stretches mainly inaccurate • May attempt to
cause strain • Errors cause use some
strain interactive
strategies
• Contributions
may often be
inappropriate to
conversation flow
or context; often
simplistic and
inflexible
19–10% • General lack of • Produces a • Vocabulary range • Great difficulty
control over narrow range of inadequate, very contributing
tempo, rhythm basic structures repetitive and opinions and
and intonation only inflexible ideas even when
interferes with • Complex • Word choice/form asked directly
communication structures rare errors may be • Does not use
• Often difficult to and usually frequent interactive
understand inaccurate strategies
individual words • Errors interfere • Infrequent
and some longer with contributions
stretches communication often
inappropriate,
irrelevant,
simplistic due to
lack of
understanding of
others’
contributions
9–0% • General lack of • Produces narrow • Vocabulary range • Can communicate
control over range of basic inadequate, very occasionally
tempo, rhythm structures with repetitive and using a few words
and intonation frequent inflexible or phrases but
impedes inaccuracies • Word choice/form unable to
communication • Errors impede errors may be participate in
• Often very communication very frequent conversation
difficult to except to answer
understand very simple,
individual words direct questions
and longer • Responses may
stretches cause difficulties
for others
Appendix B
Oral Criteria Seminars Guidelines:
Pronunciation and Interactive Communication
Pronunciation:
tempo: speed of delivery, natural connected speech, (may be affected by speech
being too fast or too slow; there may not be a regular speed of delivery).
rhythm: chunking, word and sentence stress.
intonation:
inconsistencies: both changes in tempo, rhythm and intonation
consistency = the speaker may be difficult to understand at the beginning but
becomes easier to understand as the presentation continues.
intonation and inaccurate tempo, rhythm and intonation.
Interactive Communication
Definition:
Interactive Communication Refers to the candidate’s Ability to Take Part in the Interaction
Appropriately Using Language to Achieve Meaningful Communication.
(Taylor and Galaczi 2011, p 185)
At higher levels this criterion refers to the candidate’s ability to take a proactive part in the
development of the discourse, participating in a range of interactive situations in the test
and developing discussions on a range of topics by initiating and responding appropriately.
It also refers to the deployment of strategies to maintain interaction at an appropriate level
throughout the test so that the tasks can be fulfilled.
(Taylor and Galaczi 2011, pp 185–186).
At lower levels, this includes initiating and responding, the ability to use interactive strategies
to maintain or repair communication, and sensitivity to the norms of turn-taking. Candidates
are given credit for being able to ask for repetition and clarification if necessary. (Taylor and
Galaczi 2011, 185–186).
(Dr. Linda Taylor ppt, Cambridge Speaking Composium). Using a range of inter-
active strategies: Interactive strategies include: maintaining interaction (encouraging
others to contribute, introducing new ideas/questions which come out of the discus-
sion/where there is a break in the talk), repairing misunderstandings (asking for
repetition and clarification and negotiating meaning), sensitivity to turn taking (not
dominating, handing turns to others, recognising when s.o. wishes to contribute etc.).
N.B. No detailed references were supplied for the definitions quoted in the
guidance. However, I was able to trace the source of the information to the url
below on 08/03/2020: https://slideplayer.com/slide/5293088/.
Appendix C
Notes Taken from Induction and Staff Meetings
Throughout the Course
July 4/2018.
1. Support Resources on the university’s Blackboard website contains a lot of
speaking and listening course materials encouraging students to do work at home.
2. Online listening/reading is designed to support the seminar.
3. Like last year, the presentation segment of the course is formative.
4. ‘What Students Hear’ segment of the induction was designed to raise awareness
of adjusting speech rate, using visuals and body language to facilitate under-
standing. Although the session was cut short, we did mention these and other
issues connected with listening, particularly with regards to using material from
radio 4 in the past that was way above the students’ level.
July 5/2018.
1. During the session on EAP objectives, in terms of lesson aims, it was stressed
that teachers should provide a rationale to students for the particular lesson
focus/activity.
I took the opportunity to ask whether a rationale for the listening, in terms of
task/question types will be clear, i.e., will teachers know the question types?
The course leader gave a general answer, reiterating the importance of providing
a rationale, without addressing whether the type of listening exam had been
decided.
2. On the topic of lesson observation, we focused on the EAP principles which do
refer to pronunciation instruction. The course leader emphasized that during a
typical lesson, even a writing lesson, there are opportunities to provide pronunci-
ation instruction. However there was no real explanation of how this would take
place or what kind of instruction could/should be focused on.
3. During seminar standardization we watched an exam from last year and attempted
to grade the students. There was no real explanation of the words used in the
pronunciation category rubric and their meaning.
4. At a later stage in the session, the Course Leader raises the Chinese students’
erroneous production of the ‘th’ /θ/ sound and states that it is a problem that
affects communication, and that it is something that teachers should aim to ‘fix’.
It was also stated that it was easy to fix over the 10 week course. How does this
square with the intelligibility principle?
Staff Meeting: 20th July/2018.
During the Staff meeting the Course Leader mentions that he did not write the
syllabus (scheme of work) himself. Why not? Who did write it?
It was also stated that the Steve Jobs video listening activity was unpopular and
should not have been included in the scheme of work.
Staff Meeting: August 3/2018.
It is announced that there will be some changes in the SOW. Week 5 will include
a seminar practice test in the classroom. Also, listening and reading will be piloted
in the afternoon. Does this mean the course leaders are still undecided about the
listening exam? If so, why? What is the goal of instruction in terms of listening?
This is far from clear.
Appendix D
Action Research: Some Examples of Notes
Taken Regarding My Own Teaching
Date:
Time:
Below are statements and questions about the speaking lesson you have just
conducted. Please fill in the relevant information by either circling the relevant
response or writing in the information requested.
__________________________________________________________________________________
6. How successful was the correction on a scale of 1 to 5, where 1 is not at all successful and
10. How successful was the correction on a scale of 1 to 5, where 1 is not at all successful and
14. How successful was the correction on a scale of 1 to 5, where 1 is not at all successful and
16. How many times did you correct word stress? ________
18. How successful was the correction on a scale of 1 to 5, where 1 is not at all successful and
20. How many times did you correct sentence stress? ________
21. How successful was the correction on a scale of 1 to 5, where 1 is not at all successful and
23. Students recorded their own output (e.g., on their smart phones). YES NO
__________________________________________________________________________________
Appendix F
Interview Guide: Course Leader
1. What Do You Think Are the Main Goals of the Course in Terms of Pronunciation
Instruction? NS like pronunciation or intelligibility/comprehensibility.
2. Are These Goals Officially Stated? if so, Where Can They Be Found?
3. Would you expect teachers to correct an incorrect pronunciation of a ‘th’ /θ/
sound?
4. Would you expect teachers to give learners who incorrectly pronounce ‘th’ a
lower mark? Why? Do you think teachers are aware of that?
5. Do You Think the Marking Criteria for Pronunciation is Clear? Do Teachers
Define ‘strain’ in the Same Way?
6. How Should Teachers Correct Pronunciation? in the Whole class/individually/in
Groups? Why?
7. Do You Think the Materials Adequately Support Pronunciation Instruction?
What about the listening materials (variety of types of English).
8. Teacher Pronunciation Goals
Which particular aspects of pronunciation are the most important on the course?
Why?
Aspect of pronunciation
1. Single vowel sounds
2. Single consonant sounds
3. Word stress
4. Sentence stress
5. Intonation
6. Rhythm
7. Weak forms
2. Pre-sessional Experience/Attitudes
3. Frequency of Instruction
1. How much time per week do you spend on teaching pronunciation? Why?
2. How much time do you think teachers should spend on pronunciation?
3. Do you spend this amount of time on pronunciation? If not, why not?
4. In your opinion, is the amount of time you spend on pronunciation adequate?
Why, why not? What is missing?
4. Type of Instruction
1. Which particular aspects of pronunciation are the most important in your opinion?
Appendix G: Interview Guide: Teachers 225
Aspect of pronunciation
1. Single vowel sounds
2. Single consonant sounds
3. Word stress
4. Sentence stress
5. Intonation
6. Rhythm
7. Weak forms
1 I am confident
about my English
pronunciation
skills. (I think
other people can
understand my
pronunciation).
2 I am more
confident about my
pronunciation now
than when I started
the course.
3 I want to improve
my pronunciation
skills in English.
4 I want my teacher
to teach
pronunciation.
5 I want my teacher
to correct my
pronunciation.
6 I like it when my
teacher corrects
my pronunciation
in front of the class
(when the class
can hear what
he/she says).
7 I like it when my
teacher corrects
my pronunciation
privately (e.g., in
tutorials) (The
class can NOT
hear what he/she
says).
8 I like it when my
teacher corrects
our pronunciation
as a group in class.
(He/She does NOT
focus on me
individually).
9 I want to listen to
American English
speech.
10 I want to listen to
British, Australian,
Canadian,
American or other
native-English
speech.
Appendix H: Learner Questionnaire: Student Attitudes and Experiences 229
11 I want to listen to
Indian, Chinese,
Arabic, French,
Spanish or other
English-speech
models.
12 I want to speak
English with a
British, Australian,
Canadian,
American or other
native-English
accent.
13 I want to speak
English with my
native language
accent. (E.g.
Chinese, Arabic,
Korean, etc.).
14 My pronunciation
has improved over
the course.
15 The pronunciation
activities we did
on the course were
interesting.
16 The pronunciation
activities we did
on the course will
help me on my
course in the
future.
17 The pronunciation
activities we did
on the course will
help me to speak to
other people
outside the
classroom.
18 The listening
activities in class
were interesting.
19 The listening
activities in class
will help me to
understand others
outside the
classroom.
20 The textbook
and/or materials
that the teacher
uses are helpful for
improving my
pronunciation.
21 My teacher usually
corrects me when I
pronounce words
incorrectly in
speaking classes.
230 Appendix H: Learner Questionnaire: Student Attitudes and Experiences
22 My teacher usually
corrects us if we
do not stress the
words in a
sentence correctly
in speaking
classes.
23 My teacher usually
corrects us if we
do not stress the
words in a
sentence correctly
in reading and
writing classes.
24 My teacher usually
corrects our
intonation (the
pitch or melody) of
sentences in
speaking classes.
25 My teacher
corrects my
pronunciation
privately (e.g., in
tutorials) (The
class can NOT
hear what she
says).
26 My teacher
corrects our
pronunciation as a
group in class.
(He/She does NOT
focus on me
individually).
27 My teacher
corrects our
pronunciation in
front of the class
(when the class
can hear what
he/she says).
28 My teacher asks us
to watch English
TV programmes or
listen to English
radio programmes.
29 The teacher
sometimes asks us
to record ourselves
to improve our
pronunciation in
speaking classes.
30 I usually check the
pronunciation of
new words (e.g, on
the internet).
Appendix H: Learner Questionnaire: Student Attitudes and Experiences 231
Other questions:
2. What is the ideal (best) percentage of time that you think should be spent on
3. What percentage of time does your teacher spend on pronunciation in your class? ____%
4. What activity has your teacher used that is most helpful for improving your
pronunciation?
________________________________________________________________________
6. How long have you been studying English? _______ years _________ months.
7. How long have you lived in an English-speaking country? _____ years _____ months.
MALE / FEMALE
1 My Teacher
training prepared
me well in terms
of pronunciation
instruction in
EAP.
2 My early
pronunciation
teaching
experiences
were positive.
3 Pronunciation
instruction is
important on the
pre-sessional
course.
4 Teaching
pronunciation
on the pre-
sessional course
is difficult..
5 Pronunciation
instruction is
only effective
for highly
motivated
learners.
6 The listening
materials on the
pre-sessional
course support
pronunciation
instruction.
7 The listening
materials on the
pre-sessional
course consist
of a variety of
native accents.
8 The listening
materials on the
pre-sessional
course consist
of a variety of
non-native
accents.
Appendix I: Teacher Questionnaire: Teaching Pronunciation—Attitudes … 235
9 Pronunciation
teaching on the
course should
help make
students
comfortably
intelligible to
their listeners.
10 Pronunciation
teaching on the
course should
aim to
eliminate, as
much as
possible, foreign
accents.
11 I’m completely
comfortable
teaching
segmentals, e.g.
single sounds,
minimal pairs.
12 I’m completely
comfortable
teaching all
aspects of
prosody
(suprasegmental
s), i.e., all other
aspects of
pronunciation,
such as
intonation,
stress, rhythm,
weak forms etc.
13 Drilling
minimal pairs is
the best way to
teach
pronunciation
on the pre-
sessional
course.
14 Communicative
practice is the
best way to
teach
pronunciation
on the pre-
sessional
course.
15 I find it easy to
provide
feedback on
pronunciation
during class.
236 Appendix I: Teacher Questionnaire: Teaching Pronunciation—Attitudes …
16 The
pronunciation
feedback I give
is effective in
improving
pronunciation of
single sounds.
17 The
pronunciation
feedback I give
is effective in
improving other
pronunciation
features
(intonation,
rhythm, stress,
weak forms etc).
18 I find it easy to
grade the
pronunciation
category during
seminar
assessment.
19 I would give a
student who
incorrectly
pronounced 'th'
as a 's'(e.g., I
sink*it is true) a
lower mark on
pronunciation
than a student
who pronounced
'th' correctly.
20 I wish I had
more training in
teaching
pronunciation.
(2) How have your learnt what you know about the teaching and learning of
pronunciation? (Please circle all of the categories that apply to you.)
1. teacher training
2. trial and error
3. personal experience
4. visit in a foreign country
5. reading
6. from another person (e.g. colleague)
7. other (please specify)
(3) Which of the above is the most important source of your pronunciation teaching
knowledge?.............
Appendix I: Teacher Questionnaire: Teaching Pronunciation—Attitudes … 237
How important do you think the following aspects of pronunciation are in EAP?
(Please rank from 1 to 7, where 1 means it is ‘the most important’ and 7 means
‘the least important’.)
4. In what ways, if any, could the pre-sessional on which you teach be improved in terms of
pronunciation instruction?
__________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
5. What are the major difficulties you face in teaching pronunciation on your pre-sessional course?
__________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
__________________________________________________________________________________
...............................................
238 Appendix I: Teacher Questionnaire: Teaching Pronunciation—Attitudes …
7. How many years have you been teaching English on pre-sessionals? (Please write in one figure.)
...............................................
8. What are your main qualification in terms of teaching English? (e.g., MA TESOL/Applied
...............................................
1. Female
2. Male
Dear ……………………………….
As you know I am currently studying for a PhD in Applied Linguistics at the
University of Łódź. As part of this course I am doing some research on particular
aspects of teaching speaking. I would like to find out about teacher and student
perceptions of pronunciation instruction and classroom practices in this area. In
order to do so I would like to ask for your consent for the following:
(1) For three teachers to be allowed to video record their classes on one occasion
and for teachers to fill in a self-observation checklist after three lessons.
(2) For the three teachers to grade two student seminars from last year (I received
written consent from the students last year). This will most likely be in week
one or two.
(3) For students to be asked to complete a questionnaire at the end of the course
to discover what their attitudes are towards pronunciation instruction and how
helpful they feel the instruction they received was.
(4) For a short interview with the course director on university premises that will
be recorded (audio only) during the course or during induction.
(5) For three in-depth interviews with teachers to take place in week eight. These
interviews will also be audio recorded.
(6) The video recording of one of my own lessons, self-observation checklists, and
notes to be kept as part of the action research.
(7) Access to course documents/syllabus and consent for me to make notes on
guidance received during induction and CPD sessions.
(8) Interviews with a maximum of three student participants (subject to their written
consent) approximately six months after the course has finished.
The questionnaire will be anonymous, as will the information derived from the
teacher participants’ and their classes during the observations/interviews. They will
also have the right to withdraw from the research at any time. The identity of the
university and its participants will be undisclosed.
The students will receive an information letter about the project and be asked
to sign a consent form and return it me. They will be assured that participation is
voluntary, that students can withdraw from the research at any time and that their
progress and assessment will not be affected in any way by the research.
I would like to ask for your permission for the above activities to be carried out on
university premises. I hope the results of the project will help to improve the lessons
in future pre-sessional courses held at this and other universities.
If you have any queries or concerns, I can be contacted at the email address below.
Thank you very much in advance. If you could please sign the consent form below,
I would be very grateful.
John Hodgetts email: Anonymized.
Appendix J: Course Leader Consent Form 241
Title of research project: Pronunciation Instruction: goals, practices and assessment on a UK Pre-
Please sign the form if you are in agreement with the following statements:
I have been fully informed about the format and procedure of the research project.
I agree to the follow-up interviews of three students six months after the course has finished.
I give permission for video recording of 4 lessons (subject to teacher and student consent): 1
evaluating video recorded seminars) and one interview with the course leader (subject to
I agree to access to, and analysis of work covered by teachers and syllabus content/ syllabus
There will be no mention of the university’s name or the participants in the final research
document.
I understand that I can withdraw from all or part of the research at any stage and that I can
see a summary of the findings after the research project has been completed.
Signed:____________________ Date:_____________
Course Leader
Signed:____________________ Date:______________
Please sign the form if you are in agreement with the following statements:
I have been informed about the format and procedure of the research project.
checklist after three speaking lessons (including the one that is recorded).
I agree to grade two student seminars and provide some reasons for my grading. I
understand that parts of my responses may be recorded and transcribed and that the
information given during the session will be confidential and only used in this research
project.
I agree to take part in one in-depth interview of around 40 minutes. I understand that parts
of my responses will be recorded and transcribed and that the information given during the
I understand that I can see a summary of the findings after the research project has been
completed.
Teacher Participant
Signed:____________________ Date:_____________
PhD Student
Signed:____________________ Date:______________
Appendix L
Consent Form for Adult Students
DECLARATION
that my personal details will remain confidential and that I have the right to withdraw from
……………………………………………….
Pronunciation Student
• Pronunciation (mispronunciation)
• Intonation
(=the way your voice goes up and down when you speak)
• Little L1 interference
(aspects from your own language that may affect your English pronunciation)
• Tempo
(speed of delivery, natural connected speech - may be affected by speech being too fast
or too slow; there may not be a regular speed of delivery)
• Rhythm (chunking, word and sentence stress)
Grammar Student
• Range of structures
(range = the amount of something)
• Complexity of structures
• Accuracy
Vocabulary Student
• Range
• Accuracy of word forms
(word forms e.g. nouns/adjectives/verbs/adverbs)
• Appropriacy
(the fact that a word or phrase sounds natural and is acceptable when used in a
particular situation)