The Role Accent Plays in the Evaluations of 'Native Speakerness' by "Native
Speakers" of American English
A thesis presented to
the faculty of
the College of Arts and Sciences of Ohio University
In partial fulfillment
of the requirements for the degree
Master of Arts
Nandaka M. Kalugampitiya
June 2012
© 2012 Nandaka M. Kalugampitiya. All Rights Reserved.
2
This thesis titled
The Role Accent Plays in the Evaluations of 'Native Speakerness' by "Native
Speakers" of American English
by
NANDAKA M. KALUGAMPITIYA
has been approved for
the Department of Linguistics
and the College of Arts and Sciences by
Peter Githinji
Associate Professor of Linguistics
Howard Dewald
Interim Dean, College of Arts and Sciences
3
ABSTRACT
KALUGAMPITIYA, NANDAKA M., M.A., June 2012, Linguistics
The Role Accent Plays in the Evaluations of 'Native Speakerness' by "Native
Speakers" of American English
Director of Thesis: Peter Githinji
With the broader aim of providing insights into the notion of ‘native speaker’,
the study explored the role accent played in the evaluations of ‘native speakerness’ by
“native speakers” of American English. The study collected data from 173 Ohio
University undergraduates using a structured questionnaire deployed online and semistructured focus group interviews. Based on the findings of the research, the study
claims that there are no logical/scientific bases for the concept of ‘native speaker’ and
the ‘native’/‘nonnative speaker’ dichotomy. The findings confirm the assumption that
the ‘native speaker’ concept is more a political and ideological concept than a
linguistic concept. While establishing that the ‘native’/‘nonnative speaker’ dichotomy
conceals more intense forms of division and discrimination in society, the study
emphasizes the need for more studies that aim at deconstructing the ‘native speaker’
concept in order to invalidate the said dichotomy.
Approved:
_____________________________________________________________
Peter Githinji
Associate Professor of Linguistics
4
DEDICATION
To those “rebels” in the academia
who, in the days bygone, days that are present, and the days to come,
were/are/will be committed to keeping the academia alive
by choosing dissension over conformity as their primary mode of thinking,
by being progressive as well as regressive in an academically healthy way,
by choosing to make those unheard voices, that are worthy of being heard, heard,
and by being sensitive, in a politically conscious way,
to the ground realities of human society.
5
ACKNOWLEDGMENTS
I convey my sincere gratitude to Dr Peter Githinji (Chief Supervisor and Head
of thesis committee), Dr Christopher Thompson (member of thesis committee), Dr
Ludmila Marchenkova (member of thesis committee) for their invaluable support and
guidance throughout my thesis research. I especially thank them for approving my
research proposal, which was quite different from the other studies that were bring
done in the Department of Linguistics in terms of its focus and approach, giving me
the maximum freedom possible within the broader framework of institutional
requirements to design and carry out my research the way I wanted.
I also convey my gratitude to Dr Michelle O’Malley, Dr Ludmila
Marchenkova, Dr Scott Jarvis, Christina Correnti, Annie Mitchell, and Patrick Mose
of the Department of Linguistics at Ohio University for assisting me in the data
collection stage by encouraging their students to participate in my research. I also
thank Jeff Kuhn of the Language Learning Resource Center at Ohio University for his
technical assistance in setting up the online survey of the study and Dr Greg Kessler
and Dedy Kurniawan of the Department of Linguistics for their help with the
technicalities of my data analysis.
I also express my heartfelt gratitude to all the undergraduates from Ohio and
Peradeniya who assisted me in my study by participating in the online survey and
focus group interviews.
Last but not least, I thank Ramya Vithanage, my life partner, for all the moral
support she readily gave (and continue to give) me at the most wanted moments in my
journey.
6
TABLE OF CONTENTS
Page
Abstract ......................................................................................................................... 3
Dedication ..................................................................................................................... 4
Acknowledgments ......................................................................................................... 5
List of Tables ................................................................................................................. 8
List of Figures ............................................................................................................... 9
Chapter 01: Introduction ............................................................................................. 10
1.1. Background ...................................................................................................... 10
1.2. Statement of the Problem ................................................................................. 13
1.3. Significance of the Study ................................................................................. 14
1.4. Research Questions .......................................................................................... 15
1.5. Definition of Key Terms .................................................................................. 15
1.5.1. Competence ............................................................................................... 16
1.5.2. Fluency ...................................................................................................... 17
1.5.3. Intelligibility .............................................................................................. 18
1.5.4. Pleasantness ............................................................................................... 18
1.5.5. Correctness ................................................................................................ 19
1.5.6. Accent and Accentedness .......................................................................... 19
1.6. Organization of the Thesis ............................................................................... 20
Chapter 02: Literature Review .................................................................................... 21
2.0. Introduction ...................................................................................................... 21
2.1. The Notion of ‘Native Speaker’ ....................................................................... 22
2.2. A Broader Ideological Framework for the Study of ‘Native Speaker’ ............ 24
2.3. Language Ideologies of ‘Native Speaker’ ........................................................ 27
2.3.1. ‘Native Speaker’ as a Citizen/Member of a Politically Defined Unit ....... 28
2.3.2. ‘Native Speaker’ as Having Complete Competence in Her/His Language31
2.3.3. Language as a Homogeneous and Fixed System and the ‘Native Speaker’
Category as a Homogeneous Speech/Linguistic Community ............................. 36
2.4. (Standard[ized]) Language and (Standard[ized]) ‘Native Speaker’................. 38
2.5. The ‘Native Speaker’/‘Nonnative Speaker’ of English ................................... 42
2.6. Accent/‘Non-Accent’ and ‘Native Speaker’/‘Nonnative Speaker’ .................. 46
Chapter 03: Methodology............................................................................................ 51
3.0. Introduction ...................................................................................................... 51
7
3.1. Method of Data Elicitation ............................................................................... 51
3.1.1. Questionnaire Surveys ............................................................................... 52
3.1.2. Semi-Structured Focus Group Interviews ................................................. 53
3.2. Participants ....................................................................................................... 54
3.2.1. Participants – Pilot Study .......................................................................... 54
3.2.2. Participants – Online Survey ..................................................................... 54
3.2.3. Participants – Focus Group Interviews ..................................................... 55
3.3. Instruments ....................................................................................................... 55
3.3.1. Speech Samples ......................................................................................... 55
3.3.2. Items on the Questionnaire ........................................................................ 58
3.4. Procedure.......................................................................................................... 59
3.5. Method of Data Analysis ................................................................................. 60
3.6. Limitations of the Methodology....................................................................... 61
Chapter 04: Results and Discussion ............................................................................ 63
4.0. Introduction ...................................................................................................... 63
4.1. Research Question 1 ......................................................................................... 63
4.2. Research Question 2 ......................................................................................... 70
4.3. Research Question 3 ......................................................................................... 76
4.4. Research Question 4 ......................................................................................... 81
Chapter 05: Conclusion ............................................................................................... 96
References ................................................................................................................. 103
Appendix A: Structured Questionnaire ..................................................................... 113
Appendix B: Ohio University Consent Form (Online Survey) ................................. 117
Appendix C: Ohio University Consent Form (FGI).................................................. 120
8
LIST OF TABLES
Page
Table 1: Attributes Defining the ‘Native Speaker of English’ ....................................... 64
Table 2: The Mean Scores of Different Speech Samples for Accentedness .................. 71
Table 3: The Perceived ‘Nativeness’/‘Nonnativeness’ of the Speakers of the Speech
Samples ........................................................................................................... 75
Table 4: The Perceived Places of Origin of the Speakers of the Speech Samples ......... 80
Table 5: The Mean Scores of Different Speech Samples for Fluency, Competence,
Intelligibility, Correctness, and Pleasantness ................................................... 82
Table 6: The Perceived Quality of the respondents’ English in Relation to
Respondents’ English ....................................................................................... 88
Table 7: The Chances of the Participants Recommending the Speakers as English
Teachers to a Foreigner from a Non-English Speaking Country ..................... 91
Table 8: The Professions that the Speakers Would Fit in without their English being
an issue ............................................................................................................. 94
9
LIST OF FIGURES
Page
Figure 1: Attributes defining the ‘native speaker of English’ ........................................ 65
Figure 2: Attributes defining the ‘native speaker of English’ according to those who
have had prior formal training in linguistics .................................................... 65
Figure 3: Attributes defining the ‘native speaker of English’ according to those who
have had no prior formal training in linguistics ............................................... 66
Figure 4: The perceived accentedness of the different speech samples ......................... 72
Figure 5: The perceived ‘nativeness’/‘nonnativeness’ of the speakers of the speech
samples ............................................................................................................. 73
Figure 6: The fluency rates of the different speech samples .......................................... 83
Figure 7: The competence rates of the different speech samples ................................... 83
Figure 8: The intelligibility rates of the different speech samples ................................. 84
Figure 9: The correctness rates of the different speech samples .................................... 84
Figure 10: The pleasantness rates of the different speech samples ................................ 85
Figure 11: The perceived quality of the speakers’ English in relation to respondents’
English ............................................................................................................. 87
Figure 12: The chances of the participants recommending the speakers as English
teachers to a foreigner from a non-English speaking country ......................... 90
Figure 13: The professions that the speakers would fit in without their English being
an issue ............................................................................................................. 93
10
CHAPTER 01: INTRODUCTION
1.1. Background
‘Native speaker’ is one of the key defining concepts used in linguistics.
Literature in both theoretical linguistics (Carnie, 2007; Chomsky 1965) and applied
linguistics (Davies, 1991a; Davies, 2003; Gass & Selinker, 2008; Han, 2004) shows
that the concept of ‘native speaker’ is central to many of the explanations in those
fields. Despite this position of central importance that the concept occupies in
linguistics, a clear understanding or definition of the concept is virtually absent in the
field (Davies, 1991a; Davies, 2003; Han, 2004; Lee, 2005; Train, 2009). Given the
ironic contradiction represented by the centrality of the concept in linguistics and the
absence of a definition of the concept in the field, many studies in linguistics appear
to base their understandings of complex linguistic realities on the mainstream,
commonsensical, non-technical understandings of the notion, which appear to project
it as a concept defining an almost birth-ascribed status (See Chapter 02 for a detailed
discussion of ‘native speakerness’ as a birth ascribed status).
The discussions that involve the notions/terms of ‘native speaker’ and ‘native
speakerness’ (Brown, 1992; Davies, 1991b; Leung 2005) invariably recognize two
dichotomous groups of ‘linguistic beings’: ‘native speakers’ and ‘nonnative speakers’.
While ‘native speaker’ qualifies as the unmarked status, ‘nonnative speaker’ gets
projected as the marked status and the antithesis of ‘native speaker’. The affix ‘non’
in ‘nonnative speaker’ conveys the logical implication that ‘nonnative speaker’ is
about what ‘native speaker’ is not. In a context where ‘mother tongue’, ‘first
language’, ‘dominant language’, ‘home language’, and ‘competence’ are used as the
11
defining concepts of the notion of ‘native speaker’ (Davies, 2003), ‘nonnative
speaker’, by virtue of the fact that it is the antithesis of ‘native speaker’, tends to be
seen as a status that is not defined by those concepts. While the sense of presence that
the said defining concepts attach to the ‘native speaker’ position projects the ‘native
speaker’ position as the norm/centre, the sense of absence that the fact that those
concepts do not define ‘nonnative speaker’ attaches to the ‘nonnative speaker’
position defines that position as the “non-norm”/periphery. In this sense, even before
the terms are applied to different speaker groups, the ‘native’/‘nonnative speaker’
dichotomy embodies a significant power differential. Nevertheless, the sense of
‘naturalness’ that the general tendency to view the two positions as birth-ascribed
statuses attaches to the dichotomy and the sense of objectivity and scientific validity
that the centrality of the notion of ‘native speaker’ in linguistics attaches to the same
dichotomy naturalizes/conceals this power differential.
Although the concept of ‘native speaker’ is viewed and treated predominantly
as a linguistic concept in linguistics, the political and ideological overtones that the
concept appears to have acquired over time have made it more a political and
ideological concept than a linguistic concept. An observation of the discourse on the
‘native speaker’ concept would show that the concept often goes beyond merely
explaining a person’s linguistic capacities into assigning a select group of speakers of
a given language a proprietary right not only over that language but also over
everything associated with that language. (See Chapter 02 for a detailed discussion of
this position.)
In applied linguistics, especially in the sub-fields of Second Language
Acquisition (SLA) and Teaching English to Speakers of Other Languages (TESOL),
12
the terms ‘native speaker’ and ‘nonnative speaker’ could be seen being used to refer
to those who have been exposed to a particular language in their childhood and have
grown up acquiring that language and those who have learnt/started learning that
language at a later stage in life respectively. Although the primary focus of the
studies in these fields is to discuss the nature and various aspects of the difference
between the two groups in terms of their linguistic abilities, the conceptualization of
the two groups in terms of the concepts of ‘native speaker’ and ‘nonnative speaker’,
by virtue of the fact that the two terms have acquired political and ideological
overtones, takes the difference that exists between the two groups at a linguistic level
and projects it as a fundamental difference that goes beyond mere linguistic
differences. Such a projection, on the one hand, presents a distorted understanding of
the linguistic differences between the two groups; on the other hand, it creates a
context in which the observable linguistic differences could be used to consolidate
existing stereotypical notions regarding the groups. In that way, the
‘native’/‘nonnative speaker’ dichotomy could provide a strong and fundamental basis
for division in society (Pennycook, 1994; Phillipson, 1992).
Provided that the difference between ‘native speakers’ and ‘nonnative
speakers’ is one that goes beyond the linguistic differences between those who have
been exposed to a particular language in their childhood and have grown up acquiring
that language and those who have learnt/started learning that language at a later stage
in life, and provided that the ‘native’/‘nonnative speaker’ dichotomy embodies a
power differential that is strong enough to polarize a speech community along lines
that are more “extra-linguistic” than linguistic, the notion of ‘native speaker’ should
be recognized first and foremost as a politically and ideologically defined social
13
construct. Such a recognition would highlight the need for a critical understanding of
the concept. Given that the concept of ‘native speaker’ is more a social construct than
an objective reality based on empirically observable facts, it could best (if not, only)
be studied through an investigation of the social perceptions regarding the concept.
1.2. Statement of the Problem
The present study explores the social perceptions regarding the notion of
‘native speaker’. The specific aim of the study is to examine the role accent plays in
the evaluations of ‘native speakerness’ by “native speakers” of American English.
The choice of accent as the determining variable in the participants’ assessment of
‘native speakerness’ was based on the recognition that accent is the most observable
marker of distinction between different forms of fluent English speech. Based on its
findings, the study attempts to go for a broader understanding of the notion of ‘native
speaker’ that highlights the political and ideological aspects of this seemingly
linguistic notion. The study attempts to answer, in its small way, a couple of
fundamental questions such as whether there is a definable scientific basis for the
notion of ‘native speaker’ at all, whether there is a logical basis for the
‘native’/‘nonnative speaker’ dichotomy, and whether there are more intense but
covert forms of division and discrimination in society that the traditional
‘native’/‘nonnative speaker’ binary eclipses, forms that studies that aim at better
understanding society should explore.
14
1.3. Significance of the Study
The present study acknowledges the studies that have been done regarding the
concept of ‘native speaker’ with a view to arriving at a more critical and broader
understanding of the concept (Cook, 1999; Gluszek & Dovidio, 2010; Jenkins, 2000;
Jenkins, 2005; Lee, 2005; Lindemann, 2002; Rubin, 1992; Rubin & Smith, 1990).
Although their contribution towards an alternative understanding of the concept is
commendable, their reformist approach to the concept, coupled with their implied
assumption that the ‘native’/‘nonnative speaker’ dichotomy is a fundamentally
tenable framework, fails to make a case for cancelling out the dichotomy as a basis for
division. Not only do they fail to make a case for cancelling out the dichotomy as a
basis for division, the implication that these studies in linguistics that attempt to
critique the dichotomy first and foremost accept the dichotomy as a fundamentally
tenable framework also, in a way, legitimizes the dichotomy, thereby further
consolidating it as a basis for division. In this sense, the approach that these studies
take to the notion of ‘native speaker’ is not only inadequate but also limiting in a
fundamental way.
The present study is different from those studies in that it does not accept the
‘native’/‘nonnative speaker’ dichotomy as a fundamentally tenable framework. The
study believes that there are more intense bases for division and discrimination which
a strict focus on/adherence to the ‘native’/‘nonnative speaker’ dichotomy eclipses,
and does so legitimately, by distracting society’s attention from them. The study also
believes that any serious attempt to expose the discriminatory nature of this
dichotomy should aim at deconstructing the notion of ‘native speaker’ rather than
15
trying to reform the dichotomy while treating it as a fundamentally tenable and useful
dichotomy.
1.4. Research Questions
1. How do “native speakers” of American English define the notion of ‘native
speaker’?
2. Does a speaker’s perceived accentedness function as an indicator of her/his
perceived status as a ‘native’/‘nonnative speaker’?
3. Does a person’s perceived accent function as an indicator of her/his place of
origin?
4. How do the “native speakers” of American English rate fluent speech with
“normative”/“non-normative” American accents and non-American accents
associated with “nonnative” speaking contexts?
1.5. Definition of Key Terms
The study assumes that the accent with which a speaker speaks a language,
her/his competence and fluency as indicated by her/his speech, and the intelligibility,
pleasantness, and correctness of her/his speech function as indicators/measures of the
quality of her/his speech in the eyes (if not ears) of an observer who is proficient in
that language. This section defines the terms competence, fluency, intelligibility,
pleasantness, and correctness as they are used in the study. The section also gives a
brief definition of ‘accent’ and ‘accentedness’ as the concepts are used in this study.
Detailed discussions of the terms ‘native speaker’ and ‘accent’ are given in Chapter
02.
16
1.5.1. Competence
Chomsky (1965) broadly defines competence as “the speaker-hearer’s
knowledge of his language” (p. 4), and he differentiates it from linguistic
performance, which he defines as “the actual use of language in concrete situations”
(p. 4). This notion of competence appears to refer to one’s intrinsic knowledge of the
system of grammar of her/his language on which her/his linguistic performance is
based. These definitions carry the implication that competence cannot be observed
directly. According to Chomsky’s notion of linguistic theory (1965), a person’s
linguistic performance is an accurate reflection of her/his linguistic competence only
in the case of “an ideal speaker-listener, in a completely homogeneous speech
community, who knows its language perfectly and is unaffected by such
grammatically irrelevant conditions as memory limitations, distractions, shifts of
attention and interest, and errors (random or characteristic) in applying his knowledge
of the language in actual performance” (p. 3). Given that the possibility of a speaker
of any language meeting all the requirements stipulated in this definition of the ‘ideal
speaker-listener’ is extremely low, if not nil, linguistic competence, as defined in
theoretical linguistics, remains to be a theoretical position or condition whose
existence cannot be directly observed in empirically verifiable ways. The present
study, which primarily engages with the social dimensions of the ‘native speaker’
concept, goes beyond this theoretical position and emphasizes the need to understand
linguistic competence in relation to linguistic performance. The present study defines
competence as one’s knowledge of a language as manifested through her/his
17
performance. Here, the study finds Dore’s (1986) conceptualization of competence to
be useful:
The analytical criteria for evaluating linguistic competence have been: (1)
‘well-formedness’ of a sentence’s grammatical structures; (2) the
meaningfulness of its semantic content in terms of notions like ambiguity, and
synonymy presupposition; and (3) the truth of the propositions
(presuppositions and so on) underlying otherwise well-formed and meaningful
sentences relative to the circumstances to which they correspond. (p. 4)
The present study treats “well-formedness” and meaningfulness as important bases for
an evaluation of one’s linguistic (and sociolinguistic) competence. It does not,
however, recognize the third criterion (truth) as providing a useful base for the
evaluation.
1.5.2. Fluency
Fluency is seen as one component of oral proficiency (Kormos and Denes,
2004, p. 147). The notion has variously been conceptualized as the ability to talk at
length without hesitations (Fillmore, 1979); the ability to express one’s message in a
coherent, reasoned, and ‘semantically densed’ manner (Fillmore, 1979); The ability to
be creative and imaginative in one’s language use (Fillmore, 1979); “the
communicative acceptability of the speech act” (Sajavaara, 1987, p. 62), “the native
speaker’s ability to produce fluent stretches of discourse” (Pawley & Syder, 1983, p.
191); “an impression on the listener’s part that the psycholinguistic processes of
speech production are functioning easily and efficiently” (Lennon, 1990, p. 391); the
ability to execute simultaneously the activities of planning and uttering of a speech act
18
(Rehbein, 1987, p. 104); and automatic speech production that does not require much
attention or effort (Schmidt, 1992, p. 358). The present study uses the definition
proposed by Lennon (2000), which defines fluency as “the rapid, smooth, accurate,
lucid, and efficient translation of thought or communicative intention into language
under the temporal constraints of on-line processing” (p. 26).
1.5.3. Intelligibility
Intelligibility has been defined as “the recognisability of a speech stimulus”
(Hawley, 1977, p. 2) and “being understood by a listener at a given time in a given
situation” (Kenworthy, 1987, p. 13). The present study also defines intelligibility in
terms of the recognisability and ‘understandability’ of speech by a listener. Focussing
on intelligibility assessment, Kenworthy states, “the easiest way to assess the
intelligibility of particular speakers is simply to ask someone to listen to them speak
and say how difficult or easy they are to understand. Such impressionistic or
‘subjective’ assessments are both accurate and dependable” (1987, p. 20). The
present study uses the same method in assessing speech intelligibility.
1.5.4. Pleasantness
The notion of pleasantness can be seen being used as a criterion for assessing
folk perceptions of different linguistic varieties in the fields of folk linguistics and
perceptual dialectology (Fridland, 2008; Fridland, Bartlett, & Kreuz, 2005;
Knickerbocker & Altariba, 2011; Kuiper, 2005; Niedzielski & Preston, 2000; Preston,
1989, 1993, 1996). Although pleasantness is emerging as a widely used criterion in
the assessment of linguistic perceptions, a clear definition as to what pleasantness
19
means is not to be found in the field. This situation shows that the term pleasantness
functions mainly as a self-explanatory concept. The understanding of the notion of
pleasantness in the present study is based on the definition proposed by Eadie and
Doyle (2002) according to which pleasantness is “a dimension that relates to how
‘pleasant’ you find the speaker’s voice as a listener” (p. 3017).
1.5.5. Correctness
Correctness is another self-explanatory concept that is being used in the fields
of folk linguistics and perceptual dialectology, mainly as a measurement of folk
perceptions of different linguistic varieties (Niedzielski & Preston, 2000; Preston,
1989, 1993, 1996). Myhill (2004), who investigates the notion of correctness,
recognizes three types of linguistic correctness (pp. 391-392): textual correctness (the
form of correctness based upon the written form of the language); prestige-based
correctness (the form of correctness based upon the usage of the speakers of the
language who represent the elite and whose behaviour in general constitutes the
model for society); and prescriptive correctness (the form of correctness determined
by some recognized body of authorities, either an official government institution or a
set of people who in one way or another have been informally but generally
recognized as such). The sense in which the notion of correctness is used in the
present study is similar to what Myhill defines as prestige-based correctness.
1.5.6. Accent and Accentedness
The study uses Lippi-Green’s (1997) definition of ‘accent’, which defines the
notion as “loose bundles of prosodic and segmental features distributed over
20
geographic and/or social space” (p. 42). Going by this definition, the study defines
‘accentedness’ as the extent to which a person’s speech carries prosodic and
segmental features that are markedly associated with and representative of specific
geographical and/or social spaces.
1.6. Organization of the Thesis
The thesis consists of five chapters. Chapter 01 (the current chapter) gives a
basic overview of the study. Chapter 02 presents a review of literature pertaining to
the notions of ‘native speaker’ and accent. The chapter also presents the theoretical
framework in terms of which the study reads the ‘native speaker’ concept. While
Chapter 03 discusses the methods of data collection and analysis that the study used,
Chapter 04 presents an analysis of the data and a discussion of the findings of the
study. Chapter 05 presents the conclusions of the study.
21
CHAPTER 02: LITERATURE REVIEW
2.0. Introduction
This chapter presents a discussion of some of the key texts and studies that
deal primarily with the notion of ‘native speaker’ and also with the notion of accent.
As far as the notion of ‘native speaker’ is concerned, the literature review, in keeping
with the fundamental assumptions and structure of the study outlined in Chapter 01,
focuses on the notion as an organic whole, rather than looking at it as a domain
composed of ‘native speaker’ and ‘nonnative speaker’ as the thesis and antithesis of a
dichotomy and discussing them separately.
Section 2.1 discusses some of the mainstream understandings of ‘native
speaker’ that have had a significant influence on the formal understandings of many
linguistic realities within the discipline of linguistics and also an account of a few
critical approaches to the notion of ‘native speaker’. While Section 2.2 presents a
broader ideological framework for the study of the ‘native speaker’ concept, Section
2.3 discusses three ideologies (emerging from the broader ideological framework) that
are specific to the ‘native speaker’ concept. Section 2.4 discusses certain issues
related to the more specific concepts of ‘standard language’ and ‘standard native
speaker’. Section 2.5 focuses on the notion of the ‘native speaker’ of English, while
discussing why a study of the ‘native speaker’ concept in relation to the English
language would provide better insights into the discriminatory nature of the notion.
The final section (2.6) focuses on the notion of accent in relation to the ‘native
speaker’ concept.
22
2.1. The Notion of ‘Native Speaker’
The notion of ‘native speaker’ is one of the fundamental concepts in
linguistics, especially in applied linguistics (Davies, 1991a; Davies, 2003; Gass &
Selinker, 2008). This notion could be seen being used as a defining concept, or, at
least, being commonly referred to in linguistic explanations in many sub-disciplines
of linguistics. According to Davies (1991a), “Applied linguistics makes constant
appeal to the concept of the native speaker. This appeal is necessary because of the
need applied linguistics has for models, norms and goals” (p. 1). Although the notion
operates as a defining concept in linguistics in general and applied linguistics in
particular, a proper definition of the notion is not to be found (Davies, 1991a; Davies,
2003; Han, 2004; Lee, 2005; Train, 2009). A close attention to the notion would
show that it is an “elusive enigma” (Lee, 2005) that is “neither simple nor settled”
(Train, 2009, p. 47) and it always “slips away” (Davies, 1991a, p. 1). The concept is
also “rich in ambiguity” (Davies, 1991a, p. 2). In a context where a proper, technical
definition of the notion of ‘native speaker’ is absent, researchers in the field of
linguistics seem to base their understanding of complex linguistic realities that
involve the notion in question on the mainstream, commonsensical, and non-technical
understandings of the notion.
The conventional understanding of the notion of ‘native speaker’ projects the
notion more as a self-explanatory concept (Lee, 2005) than as a concept that requires
a clear-cut definition. Chomsky (as quoted in Paikeday, 1985) seems to reinforce the
idea that the notion in question is a self-explanatory one when he says, “everyone is a
native speaker of the particular L-s that that person has ‘grown’ in his/her mind/brain.
23
In the real world, that is all there is to say” (p. 58). This statement seems to convey
the idea that the meaning of the concept of ‘native speaker’ is so obvious that the
question ‘who is a native speaker’ is a non-question. Ferguson (1983) displays a
similar attitude towards the absence of a clear definition of the notion when he says,
“In fact the whole mystique of native speaker and mother tongue should preferably be
quietly dropped from the linguist’s set of professional myths” (as quoted in Davies,
2003, p. 2).
The conventional understanding of the notion in question describes a ‘native
speaker’ as someone who speaks a given language as her/his first/native language
(Bloomfield, 1933; Doerr, 2009a) and someone who has an innate and complete
competence in a language (Davies, 2003; Pennycook, 1994; Takato, 2009). The
conventional understanding of the notion also recognizes one’s place of birth to be a
determining factor in deciding if s/he is a ‘native speaker’ of a given language or not
(Davies, 1991a).
The critical approaches to the notion of ‘native speaker’, which appear to have
been influenced by post-structuralist thinking, go beyond this conventional
understanding of the notion. The following statement by Train (2009) indicates a
significant turning point in attitudes towards this notion:
Despite nativist claims to the contrary, the nativeness of language turns out to
be more genealogical, in the postmodern sense, than genetic. Nativeness is
constructed through complex, historically contingent practices and ideologies
embedded in shifting social, linguistic, cultural, and effective contexts. (pp.
47-48)
24
This statement rejects the conventional understanding of the notion of ‘native
speaker’ as a biologically defined position and views the notion first and foremost as a
construct. Kramsch (1997), who looks at the notion from a similar point of view, says
“Native speakership ... is more than privilege of birth or even education. It is
acceptance by the group that created the distinction between native and nonnative
speakers” (p. 363). Kubota (2009) highlights the role that “non-”/“extra-linguistic”
factors play in defining the notion of ‘native speaker’ when she says, “the superiority
of the native speaker is not based purely on the linguistic attributes of individual
speakers” (p. 236). According to Takato (2009), “one’s ‘native speaker’ status can
never be fixed, innate, and complete. It can only be fleeting and fluid, or else
multiple” (p. 98).
The notion of ‘native speaker’ is bound by a socio-cultural context, and one
way to understand this notion is to approach it as a political and ideological concept.
The political and ideological nature of the ‘native speaker’ concept requires any
serious study of the concept to go beyond the conventional apolitical understandings
of the concept and examine the political/ideological factors/forces that produce,
maintain, and reproduce the concept.
2.2. A Broader Ideological Framework for the Study of ‘Native Speaker’
The success of a political/ideological understanding of the notion of ‘native
speaker’, or any other linguistic notion for that matter, depends on a critical
evaluation of the idea of linguistic differentiation. Irvine and Gal (2000) make an
important contribution to the field of linguistics research by exploring the idea of
linguistic differentiation. They conceptualize the ideological aspects of linguistic
25
differentiation as “the ideas with which participants and observers frame their
understanding of linguistic varieties and map those understandings onto people,
events, and activities that are significant to them” (p. 402). They call such ideas
‘ideologies’ because “they are suffused with the political and moral issues pervading
the particular sociolinguistic field and are subject to the interests of their bearer’s
social position” (p. 402).
This conceptualization that Irvine and Gal (2000) present is important for two
reasons. First, it challenges the common misconception that the researcher’s
assessment of the realities under study is objective. It shows that not only
commonsensical understandings but also linguistic understandings, which, most of the
time, are treated as scientific and technical understandings, are governed and
conditioned by ideology. Second, it points to the possibility of existing
understandings of specific concepts like ‘native speaker’ that are closely related to the
social aspect of linguistic differentiation being influenced and conditioned by
ideology.
Irvine and Gal (2000) identify three interconnected semiotic processes, namely
‘iconization’, ‘fractal recursivity’, and ‘erasure’, and they recognize these processes
as the means by which people construct ideological representations of linguistic
differences between/among groups of people. These semiotic processes “concern the
way people conceive of links between linguistic forms and social phenomena” (Irvine
& Gal, 2000, p. 403). According to Doerr (2009a), “Language ideologies ...
naturalize ... [the] representations [constructed by these semiotic processes] as
something universal or timeless, objective and neutral, and beyond political
contestation” (p. 18).
26
Iconization refers to a process in which “Linguistic features that index social
groups or activities appear to be iconic representations of them, as if a linguistic
feature somehow depicted or displayed a social group’s inherent nature or essence”
(Irvine & Gal, 2000, p. 403). The process leads to an ideological representation that
binds a linguistic image and a social image together on the basis of certain qualities
that are supposedly shared by both images. The ideological representation that binds
the linguistic image with the social image naturalizes this linkage, thereby projecting
the linkage as an inherent bond (Irvine & Gal, 2000, p. 403).
Fractal recursivity involves “the projection of an opposition salient at some
level of relationship, onto some other level” (Irvine & Gal, 2000, p. 403). The authors
further state:
[I]ntragroup oppositions might be projected outward onto intergroup, or vice
versa. Thus the dichotomizing and partitioning process that was involved in
some understood opposition (...) recurs at other levels, creating either
subcategories on each side of a contrast or supercategories that include both
sides but oppose them to something else. (p. 403)
The form of dichotomization and partitioning that fractal recursivity brings about
points to the possibility of certain existing dichotomies/divisions that play a central
role in organizing and hierachizing society being more a matter of mere projection
than a matter of actual distinctions. It also points to the possibility of certain
dichotomies that are perceived to be ‘real’ and exist in a particular space being a mere
projection of a division that exists in an entirely different space.
Erasure refers to a process in which “ideology ... renders some persons or
activities (...) invisible” (Irvine & Gal, 2000, p. 404). In the case of erasure, language
27
ideology erases those cases that are inconsistent with/fail to fit in a given ideological
scheme by making them go unnoticed, by transforming them to match the scheme,
and/or by explaining them away (Doerr, 2009a; Irvine & Gal, 2000). Talking further
about erasure, Irvine and Gal (2000) observe:
Erasure in ideological representation does not, however, necessarily mean
actual eradication of the awkward element, whose very existence may be
unobserved or unattended to. It is probably only when the “problematic”
element is seen as fitting some alternative, threatening picture that the semiotic
process involved in erasure might translate into some kind of practical action
to remove the threat, if circumstances permit. (p. 404)
This account shows that those cases that are inconsistent with the existing order begin
to appear as ‘problematic’ cases mostly when there is a possibility of those cases
being accommodated into an alternative order that could pose a threat to the existing
order. The account depicts erasure as a means by which the existing order repels the
alternative order in the event of challenge and attempts to maintain its superiority.
Not only do the three interconnected semiotic processes that Irvine and Gal
(2000) discuss point to the ideological force behind the creation and maintenance of
the ‘us versus them’ dichotomy, they also highlight the complexity of the power
dynamics that characterize ‘us’ and ‘them’ as separate categories.
2.3. Language Ideologies of ‘Native Speaker’
Woolard (1998) defines ‘language ideology’ as “[r]epresentations, whether
explicit or implicit, that construe the intersection of language and human beings in a
social world” (p. 3). Pennycook (1994, p. 176) presents three language ideologies
28
that are directly related to the production, maintenance, and reproduction of the
conventional notion of ‘native speaker’:
1. The idea of there being a close correspondence between holding the
citizenship of a nation-state and being a native speaker of the national
language of that nation-state
2. The idea that being a native speaker automatically bestows one with a high
level of competence in all domains of one’s first language.
3. The idea that language is a homogeneous and fixed system with a
homogeneous speech community
These three ‘native speaker’ ideologies indicate the extent to which the concept could
provide a basis for division in society. Phillipson (1992) uses the word linguicism to
refer to such “ideologies, structures, and practices which are used to legitimate,
effectuate, and reproduce an unequal division of power and resources (both material
and immaterial) between groups which are defined on the basis of language” (p. 47).
As Doerr (2009a, p. 18) observes, these three language ideologies correspond to the
three interconnected semiotic processes that Irvine and Gal (2000) present.
2.3.1. ‘Native Speaker’ as a Citizen/Member of a Politically Defined Unit
Train (2009), who takes a critical anthro-philological approach to the notion of
‘native speaker’, discusses the notion from an etymological point of view. He says,
“As the Latin etymon natives (from the verb ‘to be born’) suggests, the nativeness of
language as something one is born with is closely tied to notions of naturalness and
authenticity (...) as well as notions of belonging to a tribe, species, nation, etc. (natio)”
(p. 47). This statement captures the close connection that is generally perceived to
29
exist among the notions of ‘the nativeness of language’, ‘belonging to a tribe, species,
nation, etc’, ‘naturalness’, and ‘authenticity’. Focusing on the word ‘native’,
Williams (1976) says that the term implies “inmate, natural, or of a place in which
one is born (...) [with a] positive social and political sense, as in native land, native
country” (p. 180). Referring to the conventional notion of ‘native speaker’, Davies
(1991a) says, “The native speaker is ... one who can lay claim to being a speaker of a
language by virtue of place or country of birth” (p. ix).
Such understandings indicate the central role that one’s place of birth and
social membership (including variants such as ‘citizenship’) play in determining
her/his status as a ‘native speaker’ or otherwise. They imply that if one is to be
recognized as a ‘native speaker’ of a given language s/he needs to have been born in
and/or be a citizen of a particular country/geographical location. This birth-ascribed
nature of the ‘native speaker’ status indicates an instance of iconization in the sense
that the language spoken by/associated with the community into which one is born
becomes an iconic representation of that person’s identity, and one of the main
notions in which this iconic representation is contained is the concept of ‘native
speaker’.
Certain scholars (Davies, 1991a; Lee, 2005) in the field of linguistics
problematize this perceived link between one’s country/place of birth/citizenship and
her/his ‘native’/‘nonnative speaker’ status. However, those problematizations are
based more on a critique of related linguistic realities than on a critique of the
ideological aspect of the link. Lee’s (2005) critique of the link could be shown as a
case in point:
30
[T]his [the idea of an individual becoming a native speaker of a language by
virtue of country/place of birth] is inadequate in determining whether an
individual is a native speaker of a language, or not, due to the fact that
individuals can be resettled to other places in childhood, as in the case of
children who immigrate or are adopted in early childhood. Additionally,
being born in a place does not guarantee that the person will be a native
speaker of the native area because the language that the individual speaks at
home may not coincide with the language of the native area; and children who
are adopted in early childhood may not develop in the same linguistic
environment of his or her birthplace. (pp. 154-155)
This critique draws attention to certain scenarios in which the conventional
understanding of the perceived link between one’s birthplace and her/his position as a
‘native’/‘nonnative speaker’ becomes problematic. However, the fact that the critique
is limited to an analysis of certain context-specific practical concerns which the link
raises in certain specific contexts and that it does not go into an analysis of the
ideological aspects of the link with a view to problematizing the fundamentals of the
notion of ‘native speaker’ indicates the researcher’s tendency to accept the notion at
least at a theoretical level, an acceptance that invariably creates a space in which the
notion of ‘native speaker’ and its associated dichotomy could be viewed as tenable,
even useful, concepts.
A similar form of acceptance of this link could be seen even in the critiques of
the notion of ‘native speaker’ by scholars like Kachru (1985) and Kandiah (1998a;
1998b) who appear to call for a radical redefinition of the notion as part of the case
they make for ‘new varieties of English’/‘new Englishes’. The ‘three-circle’ model
31
proposed by Kachru (1985) and the ‘three-class’ model proposed by Kandiah (1998a),
which recognize and attempt to legitimize certain forms of English that had been
considered not up to par with “prestigious” varieties of English, indicate a significant
move towards a democratization of the English language. Although these models
appear to be groundbreaking in that they point to the need for a reconceptualization of
key sociolinguistic concepts, the fact that their categorization of different varieties of
English is based on geographical locations/territories marked by political boundaries
indicates the extent to which even radical, politically conscious scholars (who mostly
identify themselves with the periphery in the discourse on the English language) in
the field of linguistics are influenced by the naturalized iconic representation of the
link in question.
2.3.2. ‘Native Speaker’ as Having Complete Competence in Her/His Language
The following extract from Davies (2003) shows the way the conventional
understandings of the notion of ‘native speaker’, which appear to provide the basis for
the understandings of the notion in linguistics, portray the ‘native speakers’ of a given
language as having complete competence in that language:
It is surely a common sense idea, referring to people who have a special
control over a language, insider knowledge about ‘their’ language. They are
the models we appeal to for the ‘truth’ about the language, they know what the
language is (‘Yes, you can say that’) and what the language isn’t (‘No, that’s
not English, Japanese, Swahili, ...’). They are the stake-holders of the
language, they control its maintenance and shape its direction. (p. 1)
32
Pennycook (1994) expresses a similar idea when he says that a ‘native speaker’ is an
“idealized person with a complete and possibly innate competence in the language”
(p. 175). Takato (2009) adds to this idea when he says, “The commonsense notion of
‘native speaker’ assumes natural and complete competence in an individual’s first
language or ‘mother tongue’” (p. 83). Carnie’s (2007) statement that one should rely
on “the knowledge of a native speaker consultant for languages that we don’t speak
natively” (pp. 12-13) indicates the extent to which this commonsense notion has
provided a basis for understandings of linguistic realities in theoretical linguistics.
Davies (2003) describes the nature of this competence or ‘insider knowledge’
when he discusses the four forms of knowledge that a ‘native speaker’ is said to
possess of her/his language. The first form of knowledge is ‘metalinguistic
knowledge’ which means the “knowledge about the language” (p. 88), and this refers
to the speaker’s “manipulative ability ... to put together sounds, intonation, stressing,
rhythm, sentences, discourses, registers, styles, perhaps within a very limited range”
(p. 89). The second form of knowledge, which is termed ‘discriminating knowledge’,
refers to the speaker’s ability “to recognize what counts or what does not count as
being part of the language” (p. 89). ‘Communicational knowledge’, which is the third
form of knowledge, refers to the speaker’s knowledge as to “how to seek
appropriateness and how to recognize it, how to match background knowledge and
context in such a way that messages are understood and understandable” (p. 91). The
fourth form of knowledge, namely ‘skills knowledge’, “is about the level of skill ...
the speaker or user in language bring[s] to the event” (p. 92). In a context where the
‘native’/‘nonnative speaker’ dichotomy is established, the claim that ‘native speakers’
possess these forms of knowledge implies that those who do not conform to the
33
‘native speaker’ model are lacking in these forms of knowledge regardless of how
proficient they are in the language.
While drawing attention towards the birth-ascribed nature of the ‘native
speaker’ status, these understandings depict the ‘native speaker’ of a given language
as the perfect authority over that language. The ‘native speakers’ of a language are
seen as a special group of people who possess the ‘truth’ about that language, and this
possession of the ‘truth’ assigns them a special sense of ownership of the language.
This sense of ownership appears to give them the right not only to maintain the
language, but also to police the boundaries of the language and its speech community.
The idealization of ‘native speaker’ that this conventional/linguistic
understanding of the notion brings about carries the implication that a person who
does not conform to the ‘native speaker’ model is away from being an ideal person.
The analogy of a healthy person in medicine that Davies (1991a, p. x) uses to explain
the notion of ‘native speaker’ indicates the height of this idealization. At one level,
the parallel between the concepts of ‘healthy person’ in medicine and ‘native speaker’
in linguistics implies that the ‘native speaker’ position is the best position for a
speaker of a particular language to be in. At another level, it implies that the
position(s) other than the ‘native speaker’ position is/are pathological conditions, and
that those in such position(s) should aspire to reach the “sound” state of ‘native
speaker’. At the same time, both positions are about “a lack of malfunction” (Davies,
1991a, p. x), an assumption that depicts the positions other than these “ideal”
positions in their respective areas as sites of malfunction. This analogy shows the
privileged position that the commonsensical understanding attributes to the ‘native
speakers’ of languages.
34
Bourdieu (1977) problematizes the very notion of linguistic competence from
a sociological point of view. He conceptualizes the notion of linguistic competence in
terms of ‘symbolic capital’, “which is inseparable from the speaker’s position in the
social structure” (p. 646). He further states:
To move from linguistic competence to linguistic capital means refusing the
abstraction inherent in the concept of competence, i.e. the autonomization of
the capacity for specifically linguistic production. By competence, linguistics
implicitly means a specifically linguistic competence in the sense of a capacity
for infinite generation of grammatically regular discourse. In reality, this
competence can be autonomized neither de facto nor de jure, neither
genetically nor structurally – neither in the social conditions of its constitution
nor in the social conditions of its operation – with respect to another
competence, the capacity to produce sentences judiciously and appropriately.
(p. 646)
Bourdieu’s discussion of the notion of ‘linguistic competence’ shows that the
understanding of the notion that is available in linguistics is both limited and limiting.
His discussion problematizes the conceptualization of competence and
performance/production as two separate linguistic realities. Bourdieu’s claim that
competence is not autonomous of performance/production raises certain fundamental
questions about the accuracy not only of the dominant understanding of the notion of
linguistic competence, but also of the understandings of other linguistic realities that
are closely linked to that notion. In a context where the available understanding of
linguistic competence is incomplete and therefore problematic, the claim that a ‘native
35
speaker’ of a particular language has complete competence in that language is
problematic by default.
Doerr (2009a) sees the idea of the ‘native speakers’ of a particular language
having complete competence in that language as an ideology emerging from the
semiotic process of fractal recursivity:
In the case of ideological “native”/“non-native speaker” relationships,
hierarchical social relationships between the speakers of a standard language
and the speakers of a non-standard language within a “linguistic community”
are projected onto the relationships interpreted to be linguistically based
between “native” and “non-native speakers”. (p. 19)
According to this perspective, the ‘native’/‘nonnative speaker’ dichotomy, as we
know it, is a representation based more on a projection that aims at achieving a
political goal than on an actual distinction. This projection eclipses internal
hierachization and distinctions within the ‘native speaker’ category and diverts
attention to a distinction that is external to that group. Such a diversion invariably
results in strengthening a given system, which could be a country, a nation-state, or an
ethnic group, as a political unit in two ways. First, it strengthens the system by
allowing the centre to minimize, if not cancel out, the condition of possibility for any
internal resistance to the centre from disadvantaged quarters within the system. In
other words, despite the fact that the non-standard speakers of the language of a given
country occupy a secondary, if not lesser, position in relation to the standard speakers
of that language within the same country, the projected sense of inclusion that this
diversion shows towards the non-standard speakers of that language (within the same
geographic/political unit) strengthens the hegemony of the centre/standard speakers,
36
thereby hegemonically controlling/limiting the space for any possible resistance from
the non-standard speakers. In this sense, this projected sense of inclusion contained in
the notion of ‘native speaker’ functions as a unifying/centripetal force. The second
way in which this diversion strengthens the system is by projecting the space (both
physical and ideological) external to the system as a potential site of threat/resistance.
The act of locating the difference outside of the system, an act that constructs the
‘other’ not inside the system but outside of it, creates legitimate grounds on which the
centre could mobilize the system against the constructed ‘other’. As far as the
‘native’/‘nonnative speaker’ dichotomy is concerned, the act of blurring the
standard/non-standard speaker distinction that exists within the system and
highlighting the ‘native’/‘nonnative speaker’ distinction, a distinction that unites the
system on the one hand and marks the system as different from the external space on
the other hand gives the centre of the system (the ‘native speaker’ discourse) an
opportunity to organize and mobilize the entire system against the
constructed/perceived ‘other’ (‘nonnative speakers’).
2.3.3. Language as a Homogeneous and Fixed System and the ‘Native Speaker’
Category as a Homogeneous Speech/Linguistic Community
An objectification of certain social and linguistic realities and recognition of
certain key definable categories are crucial for the kind of systematic/scientific study
of language that linguistics as a discipline advocates (Davies, 1991a; also see Sapir,
1929). This approach explains the numerous attempts that are being made within
linguistics to depict language in general and individual languages in particular, which
is/are in a state of flux/constant change (Aitchison, 2001, p. 3), as fixed and
37
homogeneous entities. It also explains the tendency to recognize the speakers of a
given language who are located within/associated with a politically defined
unit/territory as forming a homogeneous speech community. The recognition of
language as a fixed and homogeneous entity and its speakers located
within/associated with a politically defined unit/territory as members of a
homogeneous speech/linguistic community allows “a rigid and clear distinction
between being a native speaker and not being so” (Pennycook, 1994, p. 176).
Linguistic homogeneity could be seen as an ideological process that embodies
continuous dialectical tensions between language-related forces of unification and
disunification (Bakhtin, 1981; Takato, 2009). The forces of unification, which
Bakhtin (1981) calls centripetal forces, are based on the notion of “unitary language,
represented by the political and institutional forces of centralization” (Takato, 2009, p.
84), while the forces of disunification, which Bakhtin (1981) calls centrifugal forces,
are based on the notion of “decentralization often represented by people at the
margins of a society” (Takato, 2009, p. 84). According to Bakhtin (1981), the
interaction between the centripetal and centrifugal forces results in “a contradictionridden, tension-filled unity of two embattled tendencies in the life of language” (p.
272). The notion of ‘native speaker’ that emerges from such a field of dialectical
tensions cannot but be ideologically charged.
As Doerr (2009a) observes, this ideology could be seen as emerging from the
semiotic process of erasure. The projection of language as a fixed and homogeneous
entity and its speakers as members of a homogeneous speech/linguistic community
overlooks and erases “the diversity and dynamic nature of language as well as diverse
linguistic practices of people who are considered as forming a ‘speech community’”
38
(Doerr, 2009a, p. 19). This homogenization, which is based on an inclusion of the
“normative” cases and those “non-normative” cases that can be/are willing to be reformed/“reformed” to fit in the normative model and an exclusion of those “nonnormative” cases that fail/refuse to be re-formed/“reformed”, results in creating a
representation that is not consistent with the ground realities that the representation is
said to mirror. Given the processes of selection and reformulation that are involved in
the broader process of homogenization, one could argue that what gets projected as
the language is an artificial form of a given linguistic reality and what gets projected
as its speech/linguistic community is an idealized community.
This idealized community is similar to the notion of ‘nation’ as discussed by
Anderson (1991). Anderson sees nation as an imagined community in the sense that
“regardless of the actual inequality and exploitation that may prevail in each, the
nation is always conceived as a deep, horizontal comradeship” (p. 224). Like ‘a deep,
horizontal comradeship’ conceals the inequalities and exploitation that prevail in a
nation, the notion of ‘native speaker’ conceals the inequalities and exploitation that
prevail in the idealized homogeneous linguistic community.
2.4. (Standard[ized]) Language and (Standard[ized]) ‘Native Speaker’
The homogenization of a language and its speech/linguistic community, which
is contingent upon the erasure of their diversity, involves standardization. Train
(2009) conceptualizes the notion of standardization in the following manner:
Standardization can be reanalyzed as a complex web of practices, policies, and
ideologies conferring language-ness on certain language practices and granting
privileged native-speakership to those speakers associated with the standard
39
language. What is considered the (standard) language comes to be nativized
as the putative native language of the educated members of society. It
becomes universalized and essentialized as the hegemonic ‘unitary language’
(Bakhtin 1981; Crowley 2003; Gramsci 1975) of the larger national and/or
international community of speakers. (p. 49)
This conceptualization, which presents a broader understanding of language, shows
how the “chosen” variety of a given language gets projected as that language in its
entirety. Along with the reduction of a given language to its standard(ized) form, the
speech/linguistic community of that language also gets reduced from the collective
speakership of all the possible variants of that language to the speakership of its
standard(ized) form. This standardization, which gives rise to the notion of “standard
native speaker” (Train, 2009), results in elevating the standard(ized) linguistic variety
and its speakership, which, according to Train, is made of the educated members of
society, to a higher position of hegemonic power.
According to Milroy and Milroy (1991), standard language is “an idea in the
mind rather than a reality – a set of abstract norms to which actual usage may conform
to a greater or lesser extent” (pp. 22-23). This claim problematizes the position of the
standard variety of a given language even as a variant of that language. If, as Milroy
and Milroy claims, it is the case that standard language is a set of abstract norms, the
usage that conforms to those norms gets projected as the ‘normative usage’ of a given
language. It logically follows that those who are marked by this ‘normative usage’
become the ‘normative speakers’ of that language. This notion of ‘normative
speaker’ is very much in line with Train’s (2009) notion of ‘standard native speaker’.
40
In his discussion of the standardization process of English, Doerr (2009b)
points to two domains in which language standardization takes place. One domain
includes those who speak a given language as their first language and those who
speak it as their second language. Bloomfield’s statement “The first language a
human being learns to speak is his native language; he is a native speaker of this
language” (1933, p. 43) could be seen as reflecting the dominant ideology that
governs the standardization process in this domain. This standardization process
always privileges the first language speakers of a particular language over the second
language speakers of that language, and this privileging has been institutionalized in
and by academic fields like Second Language Acquisition (Firth & Wagner, 1997).
The other domain of standardization involves only those who speak English as
their first language, and this form of standardization results in an internal
hierachization within that group. The standardization that takes place in this domain
points to the internal heterogeneity of the ‘native speaker’ category. A critical
understanding of the standardization process that characterizes this domain provides a
couple of important insights into the political and ideological aspects of the notion of
‘native speaker’. The notion of ‘standard language’ as discussed by Davies (2003)
captures the form of standardization that takes place in this domain:
[T]his induced homogeneity may well have a real effect on the language
systems that are being used, making them more like one another: and it is in
this sense from the sociolinguistic point of view that I can most appropriately
speak of a native speaker as someone who regards the standard language as
his/her mother tongue. (p. 64)
41
This understanding of the concept of ‘standard language’, which appears to be rather
apologetic of the entire process of standardization, equates the notion of ‘native
speaker’ with the notion of ‘standard native speaker’. While this statement indicates a
lack of sensitivity to the political and ideological complexities/tensions involved in
the process of standardization, it points to a dominant way in which the notion of
‘native speaker’ is conceptualized.
Each of these two standardization processes, when examined in isolation,
divides the linguistic community that it focuses on into two categories and projects
each as a homogeneous category. This categorization could be seen as strengthening
the ‘native’/‘nonnative speaker’ dichotomy. The idea of each of the two categories
that the standardization processes recognize being mutually exclusive, homogeneous
entities is problematic to begin with, as they overlook the inherent heterogeneity
within each category. Even in a context where the clear-cut demarcation between the
two categories and the idea of those categories being homogeneous entities are taken
for granted, an analysis of the two standardization processes in relation to each other
would point to the too simplistic and misleading nature of the dominant
‘native’/‘nonnative speaker’ dichotomy.
The first standardization process (the one that involves both the first language
speakers and second language speakers of a given language) brands those who speak
a given language as their second language as ‘nonnative speakers’ of that language
and grants the appellation of ‘native speaker’ to those who speak that language as
their first language. The second standardization process subdivides those who speak
the given language as their first language into two categories: the speakers of the
standard(ized) variety and the speakers of the non-standard(ized) variety/varieties of
42
the language. This standardization process elevates the speakers of the standard(ized)
variety to a position where they are recognized as the ‘true native speakers’ of the
language (Davies, 2003). This abstraction of the first language speakers of the
standard(ized) variety leaves the rest of the first language speakers (those who speak
varieties of the same language other than the standard[ized] variety) in an undefined
position. They cannot be categorized as ‘nonnative speakers’ as they are first
language speakers of the given language. At the same time, they do not qualify as
‘native speakers’ (in the sense the term is understood in this context) as they do not
speak the standard(ized) variety of that language. The undefined space that these
speakers occupy shows that the notion of ‘native speaker’ and the ‘native
speaker’/‘nonnative speaker’ dichotomy are far more complex than the existing
understandings of them.
2.5. The ‘Native Speaker’/‘Nonnative Speaker’ of English
Fanon (1967) lays the foundation for a broader understanding of the notion of
language when he says, “To speak means to be in a position to use a certain syntax, to
grasp the morphology of this or that language, but it means above all to assume a
culture, to support the weight of a civilization” (pp. 17-18). This claim shows that
language is much more than a mere tool for communication. Fanon’s claim that the
act of speaking is an act of ‘assuming a culture’ and ‘supporting the weight of a
civilization’ entails the idea that language is a, if not the, carrier of culture and
civilization. Ndebele (1987) makes his argument along the same line when he says,
“The problems of society will also be the problems of the predominant language of
that society. It is the carrier of its perceptions, its attitudes, and its goals, for through
43
it, the speakers absorb entrenched attitudes” (p. 25). This claim points to language as
the medium in which perceptions and attitudes exist and get transferred from one
person to another. The claim entails that the different languages embody the cultures
(which include their own versions of civilization) that they are associated with. The
fact that perceptions and attitudes are entrenched in language makes language a
political and ideological reality.
Shapiro’s (1989) discussion on language purism provides a couple of
important insights into the political and ideological aspect of language. Shapiro
states, “The locations it [language] creates for kinds of person/speakers partake of the
more general economy of place and status within a society, and its grammatical,
rhetorical, and narrative structures deploy responsibilities and authoritative forms of
control” (p. 28). This claim points to the way language functions as an organizing
force, a force that determines the nature of individual subject positions in the broader
social fabric. Shapiro’s argument, “the Other is located most fundamentally in
language, the medium for representing selves and other” (p. 28) again conceptualizes
language as an organizing force. According to his argument, not only is the
distinction between self and other created by language, the distinction also exists in
language. This argument also points to the fact that language is a politically and
ideologically defined reality.
While the idea that language is a political and ideological reality is true in the
case of every human language (given the fact that every human language is used in a
social context that involves power differentials), certain languages involve more
political and ideological complexities/tensions compared to other languages. English
could be seen as a language characterized by numerous political and ideological
44
complexities/tensions. The present study recognizes two factors that give rise to most
of the complexities/tensions that characterize the English language as we know it
today: (a) the close relationship between English and (post)colonialism and (b) the
close relationship between English and globalization.
Exploring the connection between English and colonialism, Pennycook (1998)
argues, “the long term conjunction between English and colonial discourses has
produced a range of linguistic-discursive connections between English and
colonialism” (p. 4). If it is the case that the perceptions, values, and attitudes of a
particular society are entrenched in its dominant language, it is only too logical to
assume that English embodies the perceptions, values, and attitudes of the stakeholder
groups involved in numerous contestations within the discourse of (British)
colonialism. Pennycook’s claim that “English is a product of colonialism” (1998, p.
9) indicates the role that colonialism has played in shaping the English language as we
know it today. The emergent notion of ‘new Englishes’/new varieties of English
(Kachru, 1985; Kandiah, 1998a; Kandiah, 1998b) and the claim that ‘new
Englishes’/new varieties of English (as opposed to standard[ized] varieties of English)
provide better insights into the way language in general and English in particular
works on the ground (Parakrama 1995) point to the gravity of the changes that
colonialism has introduced to the English language.
British colonialism played a central role in spreading English around the globe
(McKay & Bokhorst-Heng, 2008; Pennycook, 2003) and introducing it to contexts
that were drastically different from those in England (Kandiah, 1998a). This global
spread of English created a context in which English could eventually emerge as an
international language. Globalization, which followed (British) colonialism
45
historically/chronologically, even ideologically, confirmed and reinforced the
emergence of English as an, if not the, international language of the world.
Commenting on the impact of globalization on the spread of English at a global level,
Kandiah argues, “The result was not just that the English language reached right
across the world, but also that its pre-eminent position was very firmly secured”
(1998a, p. 2).
While elevating the English language to the position of an international
language, globalization also resulted in redefining the discourse of English by
introducing significant changes to the language and its surrounding discourse.
Commenting on the changes that the English language is undergoing in the context of
globalization, Pennycook (2003) argues that the present discourse of the language is
characterized by a “fluid mixture of cultural heritage ... and popular culture ..., of
change and tradition, of border crossing and ethnic affiliation, of global appropriation
and local contextualization” (p. 10). He also argues that the effect of globalization on
the use of English is “neither homogenization nor heterogenization” (p. 10).
On the one hand, colonialism and globalization have empowered the English
language by liberating it from the confines of England, which is seen as its
“‘birthplace’ and original homeland” (Kandiah, 1998a, p. 1). On the other hand, they
have redefined the language by taking it to contexts where it underwent drastic
changes. The various complexities/tensions created by these two developments have
raised questions regarding issues like what the ‘proper’ form of English is and who
the ‘owners’/‘proper users’ of the language are. The growing body of scholarly
research (Kachru, 1985; Kachru, 1986; Kandiah, 1998a; Kandiah, 1998b; McKay &
Bokhorst-Heng, 2008; Parakrama 1995) in the field recognizes the need to explore
46
these complexities and concerns. In a context where English has acquired strong
political and ideological overtones due to the numerous complexities/tensions created
by colonialism and globalization, an analysis of apparently neutral linguistic realities
like ‘native speaker’ in relation to English (an analysis that aims at a political
understanding of those realities) could provide better and detailed insights into the
political and ideological nature of those linguistic realities.
2.6. Accent/‘Non-Accent’ and ‘Native Speaker’/‘Nonnative Speaker’
The accent with which one speaks a language is probably the most noticeable
feature of that person’s speech. Not only is accent the most noticeable feature of
one’s speech, it is also a marker of her/his identity as a social being (Edwards, 1999;
Giles & Johnson, 1987; Gluszek & Dovidio, 2010; Lippi-Green, 1997; Mugglestone,
1995). Like the notion of ‘native speaker’, accent too appears to be a taken-forgranted reality or a self-explanatory notion in linguistics. “In as far as linguists are
concerned, accent can only be a fuzzy term” (Lippi-Green, 1997, pp. 41-42).
Although accent is a recurrent notion in sociolinguistic discussions, a clear definition
of this notion is not to be found in the field.
The study by Lippi-Green (1997) is important in that it tries to present a formal
definition of the term. She defines accent(s) as “loose bundles of prosodic and
segmental features distributed over geographic and/or social space” (p. 42). While
prosodic and segmental features, which are exclusively linguistic realities, are the
focus of the definition, the notion of geographical and social space, which is
technically an “extra-linguistic” notion, is central to this definition. This “extralinguistic” notion of geographical and social space not only emphasizes the need to go
47
beyond strictly linguistic terms to understand the notion, it also indicates the
possibility of the notion being a politically and ideologically defined one. The phrase
“loose bundles” indicates the absence of clear-cut boundaries, which makes the task
of recognizing and defining different accents a difficult and complicated one.
Lippi-Green (1997) recognizes two broader categories of accent: L1 accents and
L2 accents. Defining the first category, Lippi-Green says, “L1 accent is, then, the
native variety of US English spoken: every native speaker of US English has an L1
accent, no matter how unmarked the person’s language may seem to be” (p. 43). This
definition is important for two reasons. First, it shows, contrary to the general
understanding regarding accent, that every native speaker of a given language speaks
the language with an accent. This challenges the dominant misconception that the
speech of only certain speakers/groups of speakers is marked by an accent. Second,
the fact that Lippi-Green’s general discussion on the notion of accent establishes a
connection between ‘US English’ and L1 accent bestows a sense of normativity on
‘US English’ as opposed to other varieties of English (including those “prestigious”
varieties located outside of the US). Although the claim (emphasized by LippiGreen) that every native speaker of a given language speaks with an accent provides a
basis for a radical understanding of the notion of accent, the fact that the discussion
establishes a clear distinction between L1 accents and L2 accents and bestows a sense
of normativity on ‘US English’ indicates the extent to which the discussion is
influenced by certain mainstream assumptions in linguistics.
A close look at the notion of accent and ‘accentedness’ as understood and used
in/by society would show that the social meaning of accent is markedly different from
the way linguistics defines the term. The social meaning of accent defines it as a
48
feature that characterizes only certain types of speech. The term could be seen often
being used to describe the speech of politically weak elements of a speech community
(Calvet, 1998; Doerr, 2009b; Phillipson, 1992), and this exclusive use of the term to
describe the speech of the politically weak gives the impression that the speech of the
politically strong does not contain an accent. Fisherman (1971) attempts a synthesis
between the linguistic and social definitions of accent when he argues that accents are
classified by the degree to which they are considered ‘standard’ or ‘nonstandard’
within a particular speech community. According to him, the standard accent is the
one that is generally associated with status and power, whereas a nonstandard accent
is one associated with a lower level of socioeconomic success. These arguments
show that the questions ‘who speaks with an accent’ and ‘what is accented speech’ are
more political questions than linguistic questions.
Studies in the fields of sociolinguistics and social psychology have examined
social perceptions of accented speech. The study by Ryan, Hewstone, and Giles
(1984) shows that ‘standard’ accented speakers are treated highly on traits related to
intelligence and social status, whereas ‘nonstandard’ accented speakers are evaluated
less favourably along the same dimensions, even by listeners who themselves speak
with a ‘nonstandard’ accent. The study by Riches and Foddy (1989) shows that
‘standard-accented’ speakers, compared to ‘nonstandard-accented’ speakers, tend to
be rated positively on a rating dimension regarding performance ability. Dixon and
Cocks (2002) studied the influence of an English regional accent, the Birmingham or
‘Brummie’ accent, on listeners’ attributions of guilt towards a criminal suspect, and
the study showed that the suspect was rated as significantly more guilty when he
spoke with the ‘Brummie’ accent rather than a ‘standard’ accent. Lev-Ari and Keysar
49
(2010) investigated the impact of accented speech on the believability of the message,
and the study showed that accented speech was rated as less truthful than nonaccented/accent-neutral native speech. The study by Rubin and Smith (1990), which
looked at the impact of accent, ethnicity, and lecture topic on native speaker
undergraduates’ perceptions of nonnative English speaking teaching assistants,
showed a negative correlation between the impressions of the instructor’s teaching
ability and her perceived accentedness. The study by Riches and Foddy (1989)
showed that accented speech operated as a status cue which triggered beliefs about
ethnic status. These findings reiterate that a person’s accent plays a significant role in
defining that person’s identity.
Several studies (Derwing 2003; Derwing & Rossiter; 2002; Jenkins, 2000;
Lindemann, 2002; Lippi-Green, 1994; Munro & Derwing, 1995; Rubin 1992) have
investigated the connection between accent and communication. Jenkins (2000),
Derwing and Rossiter (2002), Munro and Derwing (1995), and Rubin (1992) show
that ‘nonstandard-accented’ speakers are conscious of the possibility of their accent
creating communication problems. According to the findings of Lindemann (2002)
and Lippi-Green (1994), the communication problems have more to do with
‘standard-accented’ speakers investing less effort in understanding ‘nonstandardaccented’ speakers than to do with anything inherent in ‘nonstandard’ accented
speech.
These studies show how accent functions as an indicator not only of a speaker’s
language skills, but also of certain “extra-linguistic” aspects of the speaker, such as
intelligence, credibility, social status, ethnic status, performance abilities, and even
morality. The role of accent as an indicator of these linguistic and “extra-linguistic”
50
aspects of a speaker clearly functions as a basis for division and discrimination. The
above studies play an important role in highlighting the possibility of accent
functioning as a basis for division and discrimination. However, the fact that a
majority of these studies fail to go beyond the framework of the conventional
understanding of the ‘native’/‘nonnative speaker’ dichotomy reduces the validity and
importance of their inquiry.
The present study is of the view that an exploration of notions like accent as
bases for division and discrimination, within a broader framework that accepts the
‘native’/‘nonnative speaker’ dichotomy as a fundamentally tenable dichotomy, fails to
provide thorough insights into the political and ideological character of those notions.
The present study is different from the previous studies in the field in that it attempts
to explore the notion of accent, with a special focus on its role as a basis for division
and discrimination in society, within a framework that goes beyond the traditional
‘native’/‘nonnative speaker’ dichotomy. Such an approach would emphasize the need
for a broader conceptualization of accent.
51
CHAPTER 03: METHODOLOGY
3.0. Introduction
The aim of this chapter is to describe the methodology used for collecting and
analysing data. The section 3.1 presents a theoretical discussion of the two methods
of data elicitation that were used in the study. While the section 3.2 gives an account
of the different participant groups of the study, the section 3.3 describes the
instruments used in the study. The section 3.4 describes the procedure involved in
collecting data, and the section 3.5 describes the method and process of analysing
data. The final section (Section 3.6) discusses the identified limitations of the study in
relation to its methodology.
3.1. Method of Data Elicitation
The claims made by the present study are based on both quantitative and
qualitative data collected from its participants. While quantitative data were collected
using a structured questionnaire, the method of semi-structured focus group
interviews was used to collect qualitative data. The methodology of data collection of
the study takes the ‘QUAN + qual’ pattern (Dornyei, 2007, p. 173) in that the
quantitative data collected using the structured questionnaire forms the primary data
pool of the study, while the qualitative data collected using semi-structured focus
group interviews are used as a secondary data pool that is mainly used to complement
the quantitative data.
52
3.1.1. Questionnaire Surveys
Brown (2001) defines questionnaires as “any written instruments that present
respondents with a series of questions or statements to which they are to react either
by writing out their answers or selecting from among existing answers” (p. 6).
According to him, “[q]uestionnaires are particularly efficient for gathering data on a
large scale basis” (p. 6). Questionnaires are generally used to collect quantitative
data; however, questionnaires may also contain some open-ended questions that
generate qualitative data (Dornyei, 2007, p. 101).
Dornyei (2007, p. 102) recognizes three types of data that a questionnaire can
yield. They are as follows:
1. Factual data – questions which are used to find out certain facts about the
respondents, such as demographics characteristics …, residential location,
marital and socio-economic status, level of education, occupation, … etc.
2. Behavioural data – questions which are used to find out what the respondents
are doing or have done in the past, focusing on actions, life-styles, habits, and
personal history.
3. Attitudinal data – questions which are used to find out what people think,
covering attitudes, opinions, beliefs, interests, and values.
Dornyei (2007) points out that the designing of items for a questionnaire is not
a hundred per cent scientific activity and that it requires a certain amount of creativity
and lots of commonsense on the part of the designer (p. 108). He emphasizes the
need for a rigorous piloting of the questionnaire in order to improve the quality of the
questionnaire. According to him, the designer should (1) aim for short and simple
items; (2) use simple and natural language; (3) avoid ambiguous or loaded words and
53
sentences; (4) avoid negative constructions; (5) avoid double-barrelled questions; (6)
avoid items that are likely to be answered the same way by everybody; and (7)
include both positively and negatively worded items (pp. 108-109).
Dornyei (2007) talks about two main types of questions that are included in
questionnaire surveys: multi-item scales and open-ended questions. The different
types of multi-scale items that he talks about are the Likert scales, the semantic
differential scale, numerical rating scales, true-false items, multiple-choice items, and
rank order items. He talks about four open-ended question types, and they are
specific open questions, clarification questions, sentence completion tasks, and shortanswer questions.
3.1.2. Semi-Structured Focus Group Interviews
Focus group interviews “involve a group format whereby an interviewer
records the responses of a small group” (Dornyei, 2007, p. 144). Discussing Focus
group interviews as a data collection method in qualitative research, Dornyei (2007)
observes:
The focus group format is based on the collective experience of group
brainstorming, that is, participants thinking together, inspiring and challenging
each other and reacting to the emerging issues and points. This within-group
interaction can yield high-quality data as it can create a synergistic
environment that results in a deep and insightful discussion. The format also
allows for various degrees of structure, depending on the extent to which the
researcher relies on an interview guide/protocol rather than giving the
participants freedom to discuss some broad topics. Just like with one-to-one
54
interviews, the semi-structured type of focus group is the most common
format because it includes both open- and closed-ended questions posed by the
researcher. (p. 144)
3.2. Participants
3.2.1. Participants – Pilot Study
The pilot study involved 5 American graduate students (3 female and 2 male)
in the Linguistics MA programme at Ohio University. All 5 participants represented
the White American ethnic identity.
3.2.2. Participants – Online Survey
The online survey involved 163 Ohio University undergraduates (114 female
and 49 male) who considered themselves ‘native speakers’ of English. These
participants were drawn from a variety of majors representing both social
sciences/humanities and natural sciences. Of these participants, 85 had had some
prior formal training in linguistics, while the other 78 had had no such training. This
group consisted of 136 (83%) White Americans, 10 (6%) African Americans, 9
(5.5%) European Americans1, 5 (3%) Asian Americans, and 3 (2%) Hispanic/Latino
Americans.
1
The present study uses the term ‘European Americans’ to refer to recent immigrants (or descendents
of relatively recent immigrants) from Europe.
55
3.2.3. Participants – Focus Group Interviews
The focus group interviews involved 10 Ohio University undergraduates (4
female and 6 male) who considered themselves ‘native speakers’ of English. These
participants were also drawn from a variety of subject areas. Of these participants, 4
had had some prior formal training in linguistics, while the other 6 had had no such
training. This participant group was made of 8 White Americans, 1 European
American, and 1 Asian American. The data from focus group interviews were used
mainly to complement the claims made on the basis of the data from the online
survey.
3.3. Instruments
The study used a structured questionnaire survey (Appendix A) deployed
online and semi-structured focus group interviews to collect data. In the case of both
the structured questionnaire and focus group interviews, the participants were
presented with six half-a-minute speech samples (audio only), and at the end of each
sample, the participants were expected to respond to a set of questions based on that
speech sample.
3.3.1. Speech Samples
Of the six speech samples that the study used, three represented the speech of
three American nationals, and the other three represented the speech of three nonAmericans living in the United States. The speakers of all six samples were male, and
all of them represented fluent and grammatically correct English speech. One of the
56
three American speech samples represented what is generally viewed as the
“standard” mainstream American pronunciation. Of the other two American speech
samples, one was marked for the Appalachian accent, while the other one was marked
for an African American accent. The three non-American speech samples represented
the speech of a Chinese speaker of English, an Indian speaker of English, and a
Kenyan speaker of English. In the case of these three speakers, English was their
second language, and they had learnt English through formal instruction.
The decision to use three speech samples representing each of the two
categories (Americans and non-Americans) was based on the need to present the
participants with an equal number of samples representing the two categories.
Although a greater number of samples would have been preferable, it was decided to
limit the number of samples to six in order to keep the survey short.
The “standard” American speech sample and the three non-American speech
samples were taken from four graduate students at Ohio University. As far as the
selection of the “standard” American speech sample was concerned, the researcher
made a list of the male graduate students in the Master of Arts (First Year) in
Linguistics programme and conducted a small informal research to find out whose
speech could be considered the closest to the “standard” American pronunciation. The
choice of the majority of the Americans (who considered themselves “native
speakers” of American English) whom the researcher consulted was chosen for the
“standard” American speech sample. In selecting speakers for the non-American
samples, the researcher used his discretion and chose, from a pool of non-American
students at Ohio University, three graduate students who the researcher thought to be
57
speaking English with accents that were generally associated with the countries they
were hailing from (China, India, and Kenya).
The four speech samples (the “standard” American speech sample and the
three non-American speech samples) were recorded at a computer lab at Ohio
University. The speakers were asked to talk about a topic of their choice for about a
minute, and their speech was recorded. From the recorded one-minute samples, halfa-minute sections were extracted for the purposes of the study.
As far as the other two American speech samples are concerned, the
researcher failed to find two speakers whose speech embodied what could be
considered typical Appalachian and African American accents. Therefore, two
speech samples marked for these two accents were taken from Youtube. The
authenticity of the two speech samples was confirmed by consulting the chief
supervisor of the present research study.
In the case of both the structured questionnaire and focus group interviews, the
first speech sample that the participants were presented with was the “standard”
mainstream American speech sample. The decision to make this sample the first in
the order of the samples was based on the assumption that the respondents in the
target group would (1) be familiar with the pronunciation found in the sample and (2)
probably recognize it as the “normative” pronunciation of English. Such a
recognition would give them a point of reference with which they would be able to
compare the other speech samples.
The order in which the speech samples appeared in both the structured
questionnaire and focus group interviews is as follows: “standard” American speech,
“Chinese” speech, “Indian” speech, “Appalachian” speech, “African American”
58
speech, and “Kenyan” speech. Henceforth, these speech samples will be referred to
as STA, CHN, IND, APP, AFA, and KEN respectively.
3.3.2. Items on the Questionnaire
The structured questionnaire used questions that aimed at generating factual
and attitudinal types of data. While the ‘Information about the Respondent’ section
(in Appendix A) contained both factual and attitudinal questions, all the questions
about the speech samples aimed at generating attitudinal data. Except for one
question in the ‘Information about the Respondent’ section, which required the
respondent to give her/his definition of ‘native speaker’, and two other short-answer
questions (one asking the participant’s area of study/major and the other asking the
participant to guess the area of the United States where the speaker of each speech
sample was from in the case of those samples that s/he recognized as representing a
American English variety), all the items on the questionnaire were closedended/multi-item questions.
The closed-ended items on the questionnaire were multiple-choice questions
and questions that required the participant to respond by marking a point on a Likert
scale. Multiple-choice questions, which required the participants to respond by
selecting one or more appropriate answers from a given list, allowed the researcher to
have some sort of a control over the responses that the participants gave. Although
multiple-choice questions could be seen as limiting in the sense that the options are
already defined for the respondent, in a context where the options reflect all the
possible options for a given question, multiple-choice questions would give the
researcher a way of gathering a mass of information in a quantifiable form.
59
The Likert scale consists of “a characteristic statement and respondents are
asked to indicate the extent to which they ‘agree’ or ‘disagree’ with it by marking ...
one of the responses ranging from ‘strongly agree’ to ‘strongly disagree’” (Dornyei,
2007, p. 105). The range of options that the Likert scale gave the respondents enabled
the researcher to gain a nuanced understanding of the realities in question. This
method was extremely effective when collecting attitudinal data.
3.4. Procedure
The process of collecting data was started following the receipt of IRB
approval for the study in June 2011. The first stage of data collection involved
collecting data from 163 Ohio University undergraduates, using the structured
questionnaire deployed online. Before participating in the study, the participants had
to give their consent to the study by creating an electronic signature by clicking a
button. This electronic signature would allow the participants to start answering the
questionnaire.
Participants for the online survey were recruited mostly from LING 270,
LING 280, and LING 480 and classes at the Ohio University Linguistics Department
in Winter 2012/2013. With the permission of the Instructors and Teaching
Supervisors of these classes, the students were requested to participate in the online
survey.2 In order to make sure that the responses of those who had not had a formal
training in linguistics were minimally influenced by the contents of those courses,
data collection was limited to the first two weeks of class. The participants were
2
To encourage participation, the LING 270 students were offered extra-credit for participating in the
study. Given that the extra-credit that the LING 270 students got was less than 1 percent of the overall
marks possible for the course and also that an alternative extra-credit option was offered for those who
did not wish to participate in the study, it is highly unlikely that this extra-credit option would have
affected the validity of the data in any serious way.
60
encouraged to take the survey on a university computer in order to make sure that
they did not encounter technical problems when accessing the online survey. (The
Consent Form that was used to obtain the consent of the participants who participated
in the online survey is given in Appendix B.)
Four focus group interviews were conducted as part of the data collection
process. (The Consent Form that was used to obtain the consent of the
undergraduates who participated in the focus group interviews is given in Appendix
C.) One interview involved 4 students, while the other three involved 2 each. Of the
total of 10 participants, 6 had had some prior formal training in linguistics, while the
other 4 had not received any such training. Participant recruitments were done using
the researcher’s personal contacts. The interviews were held at a University computer
lab. Prior to the interview, the participants signed the consent form expressing their
willingness to participate in the interview. Then they filled the personal information
forms of the questionnaire (paper copies). After that, the six speech samples were
played, and at the end of each sample, the researcher initiated a discussion using the
questions given in the structured questionnaire. While the interviews were based on
the basic framework laid down in the structured questionnaire, steps were taken to
explore the realities in question in a deeper manner and enhance the data by applying
various probes that used certain important points that the participants made to initiate
further discussions. The interviews were video recorded.
3.5. Method of Data Analysis
The data pool of the study mainly consisted of quantitative data collected
using the structured questionnaire, and it also contained some amount of qualitative
61
data collected using the focus group interviews. As far as the quantitative data are
concerned, the study used descriptive statistics generated using SPSS to recognize
patterns in its data pool. As far as the qualitative data are concerned, the study used
the method of tape analysis (Dornyei, 2007) and transcribed those sections of the
focus group interviews that were directly relevant to the research questions of the
study.
3.6. Limitations of the Methodology
The study has a couple of limitations in terms of methodology. One of the
limitations concerns the choice of accent as the determining variable. The choice of
accent as the primary determining variable in the participants’ assessment of the
different speech samples was based on the recognition that accent is the most
observable marker of distinction between different forms of fluent English speech.
While the study acknowledges the possibility of factors other than accent, such as
word choice, intonation, and stress patterns, playing a role in the participants’
assessment of different forms of speech, it does not explore the extent to which those
factors shape the participants’ judgments. Ideally speaking, the study should have
used speech samples that differ from each other only in terms of accent. However,
given that obtaining speech (natural speech) samples that are controlled for all the
variables other than accent is almost practically impossible, the study was constrained
to use samples that differed from each other in more than one respect. Although the
samples embodied differences at multiple levels, the fact that accent-based differences
emerged as the most observable marker of distinction between the samples selected
62
for the research provided a basis for the researcher to regard accent as the primary
determining factor in the participants’ assessment of the samples.
The speech samples differed from each other significantly at the level of
content, and these content differences among the speech samples could possibly have
had an impact on the participants’ assessment of the samples. Although this impact of
content differences on the participants’ assessments could have been minimized by
using samples where the speakers read out the same text, the possible sense of
monotony generated by the same text being repeated over and over again could have
discouraged the participants from actively engaging with the survey. Therefore, it
was decided to use samples with different contents. It was assumed that the accentbased differences among the samples would overweigh the content-based differences
among them.
It is possible that the participants’ familiarity with/knowledge about the speech
styles found in any or all of the samples could result in them evaluating the samples
with preconceived notions. One could argue that evaluations based on such
preconceived notions could question the validity of the study. However, given that
one of the primary aims of the study was to explore the political and ideological
nature of the factors that govern the participants’ evaluation of the given speech
samples, it was expected that any relationship between their familiarity
with/knowledge about the speech styles and the way they evaluated the samples
would, in fact, provide better insights into the discourse in question.
63
CHAPTER 04: RESULTS AND DISCUSSION
4.0. Introduction
The aim of this chapter is to present and discuss the results of the study in
relation to its research questions. Sections 4.1, 4.2, 4.3, and 4.4 are dedicated to the
research questions 1, 2, 3, and 4 respectively. Since the claims of the study are
primarily based on the data collected using the structured questionnaire deployed
online (online survey), the terms ‘participant/s’ and ‘respondent/s’ used in this chapter
primarily refer to those respondents who participated in the online survey. The
respondents who participated in the focus group interviews (FGI) will be referred to
as ‘FGI participants’ (or a similar term).
4.1. Research Question 1
The first research question aimed at exploring how the “native speakers” of
American English understood the notion of ‘native speaker’. The findings in relation
to the first research question are primarily based on the opinions of the participants
elicited by the open-ended question “How would you define a native speaker of
English?” on the questionnaire (See Appendix A). The initial analysis of the
responses pointed to a pattern in the responses with regard to the attributes that the
respondents thought defined a ‘native speaker’ of English. Taking this pattern into
consideration, the responses were categorized into 6 categories: (1) First Language
(speaking English as one’s first language); (2) Linguistic Performance (being able to
perform ‘well’ in English); (3) Geographic Location/Country (one being a citizen of
an ‘English speaking’ country or hailing from an ‘English speaking’ area in the
64
world); (4) Knowledge of the Language (one being knowledgeable of the English
language); (5) Speech with No Accent (speaking English with “no accent”); and (6)
Uncategorized (attributes that did not come under any of these categories). Of the
total responses, 15 indicated an overlap between 2 categories, and each of those
responses was counted for both the categories that they were deemed to come under.
The study was interested in finding out if there was a difference in the way
those with some form of prior formal training in linguistics and those without such
training defined the notion. Table 1 presents the category-based distribution of the
responses first for the two groups (those with and those without a prior formal training
in linguistics) separately and then for the overall respondent community. Figure 1
presents a visual representation of the category-based distribution of the responses for
all the respondents as a single group, while Figures 2 and 3 show the distribution of
the responses for those who have had a prior formal training in linguistics and those
who have not had such training respectively.
Table 1
Attributes Defining the ‘Native Speaker of English’
Total
With
Without
Linguistics
Linguistics
Training
Training
First Language
45
39
84
Linguistic Performance
17
18
35
Geographic Location/Country
11
11
22
Knowledge of the Language
08
07
15
Speech with No Accent
04
02
06
Uncategorized
06
07
13
Attribute
65
3%
First Language
7%
Linguistic Performance
9%
48%
Geographic Location/Country
13%
Knowledge of the Language
Speech with No Accent
20%
Uncategorized
Figure 1. Attributes defining the ‘native speaker of English’.
First Language
7%
4%
Linguistic Performance
9%
49%
Geographic Location/Country
12%
Knowledge of the Language
Speech with No Accent
19%
Uncategorized
Figure 2. Attributes defining the ‘native speaker of English’ according to those who
have had prior formal training in linguistics.
66
2%
First Language
8%
Linguistic Performance
8%
47%
Geographic Location/Country
13%
Knowledge of the Language
Speech with No Accent
22%
Uncategorized
Figure 3. Attributes defining the ‘native speaker of English’ according to those who
have had no prior formal training in linguistics.
Although one would expect to see a difference between the definitions of the
notion of ‘native speaker’ by those who have had some form of formal training in
linguistics and those without such training, the findings show that there is no
significant difference between the definitions from the two groups. The study views
this absence of a difference as indicative of the tendency within linguistics to base
understandings of linguistic realities that concern ‘native speakerness’ on the
mainstream, commonsense understandings of the ‘native speaker’ concept. It is also,
in a way, reflective of the relative indifference with which questions regarding the
nature of ‘native speaker’ are treated by certain dominant traditions within linguistics
(as discussed in the section 2.1).
Almost half of the participants (48%) emphasized the need for a person to
speak a language as her/his ‘first language’ for her/him to be a ‘native speaker’ of that
language. The responses categorized under the ‘First Language’ category used terms
67
like ‘first language’, ‘L1’, ‘primary language’, ‘dominant language’, ‘home
language’, and ‘native language’ to describe a speaker’s relationship with English in
the case of a ‘native speaker’ of English. They also implied the idea of a person
having to be born into an English-speaking family and raised in an English-speaking
environment for her/him to become a ‘native speaker’ of English.
The 20% of the opinions that were categorized under the heading ‘Linguistic
Performance’ primarily viewed a person’s performance in English as the primary
factor that determined her/his ‘native’/‘nonnative speaker’ status. They mostly
viewed fluency, intelligibility, and correctness of speech as the defining features of
the linguistic performance of a ‘native speaker’. Most of the respondents stated that
the ability to communicate thoughts clearly in a clear and intelligible manner was an
important aspect of this linguistic performance. While some of the opinions explicitly
stated that the linguistic performance of a ‘native speaker’ is about the ability to
perform in “proper English”, the same idea could be seen being implied by many of
the other opinions (which had not explicitly stated that) in this category. The opinions
that ‘native speakers’ “can use English without thinking about the process” and “They
are able to flow in and out of sentences” appear to point to the idea of automaticity in
language processing and production.
A 9% of the opinions viewed one’s knowledge of the English language as
what determined her/his position as a ‘native’/‘nonnative speaker’. The knowledge of
the structure of the English language, “the minute differences” and “the nuances of
the language”, and “the rules for speaking in English” were seen as composing what
they referred to as the ‘knowledge’ of the English language. According to one
respondent, one’s ability to “answer most questions, if not all, about the English
68
language” is a quality of a ‘native speaker’ of English. According to another, “They
[‘native speakers’] know the rules of the language, but they do not know why they
know the rules”.
Of the total responses, 22 (13%) viewed one’s geographic location or
citizenship in a country (English-speaking) as what determined a person’s
‘native’/‘nonnative speaker’ status. Of these 22 responses, 12 explicitly stated that
‘native speakers’ of English were those who were born and raised in the United States
and/or those who spoke ‘Standard American English’. Interestingly, 8 of these 12
responses were from those participants who had had some prior formal training in
linguistics. One respondent defined a ‘native speaker’ of English as a person who had
a “neutral dialect, similar to what is found in the midland US”, a dialect similar to
“what is displayed on a national news channel, for example CNN and FOX”. The
response, “They can have one or more accents and can be of many different races and
ethnicities since the US is very diverse in nature” could be seen incorporating the
notions of ‘race’ and ‘ethnicity’ into the definition of ‘native speaker’; however, it
still bound the ‘native speaker’ of English to ‘the US’.
A 3% (6 responses) of the total responses stated that ‘native speakers’ of
English spoke the language without an accent. Of the 6 responses in this category, 4
were given by those participants who had had some prior formal training in
linguistics. The responses that did not fit into any of these categories were listed
under the heading ‘Uncategorized’. This category included responses/definitions such
as “someone direct and could be perceived as rude if not accustomed to the culture”,
someone who is “grammatically careless and easily understood”, “someone who can
use slang and formal speech when needed”, “someone who speaks English
69
significantly slower than other speakers, particularly of different languages”, and
someone “having a decent grasp of vocabulary, with little focus on ‘correct’
grammar”. One respondent defined a ‘native speaker’ of English is “one I can
understand”.
Of the 10 respondents who participated in FGIs, 6 participants viewed ‘native
speakers’ of English as those for whom English had been the “primary language” or
“first language” “from birth”, while 4 defined a ‘native speaker’ as someone who had
been born and raised in the Unites States or someone who had lived in the United
States long enough to understand and speak the language fluently. Of the 4 of the
latter group, 2 mentioned the ‘Mid-West’ area in their definitions, implying that
‘native speakers’ were those who lived in that area or spoke the form of English that
was generally associated with that area.
The definitions given by 2 of the FGI participants implied that ‘native
speakers’ of English did not necessarily know all the rules that governed the language.
Supporting her claim, one participant said, “That’s why we take English classes”.
This reflects Carnie’s (2007) position that a ‘native speaker’s’ knowledge of the
language is largely subconscious (p. 13). The other participant said that ‘nonnative
speakers’ probably had a better knowledge of the rules of the language (compared to
‘native speakers’) because they, unlike ‘native speakers’, learnt the rules consciously.
To present the above discussed results in a summarized form, almost half of
the participants view speaking English as one’s ‘first language’ as the primary factor
that determines a her/his status as a ‘native’/‘nonnative speaker’. One fifth of the
participants and one tenth of the participants view one’s ability to perform ‘well’ in
English and one’s geographical location/country of citizenship as the primary
70
determining factors respectively. The fact that there was no significant distinction
between those who had had a formal training in linguistics and those who had no such
training could be seen as indicating the close connection, or the absence of a
significant distinction, between the mainstream commonsense understandings of the
concept of ‘native speaker’ and understandings of the concept that exist in linguistics.
4.2. Research Question 2
The second research question intended to find out if a speaker’s perceived
accent functioned as an indicator of her/his status as a ‘native’/‘nonnative speaker’.
The study obtained the participants’ opinion regarding the ‘accentedness’ of each
sample using a Likert scale of five values (1 to 5), 1 being ‘no accent at all’ and 5
being ‘a very strong accent’ (Item 4 in ‘The Respondent’s Evaluation of the Speech
Samples’ section in Appendix A). The participants’ opinion regarding the
‘nativeness’/‘nonnativeness’ of the speaker of each variety was obtained by asking
them to select one of the following statements in the case of each speech sample: (1)
The speaker is not a native speaker of English; (2) The speaker is not a native speaker
of English, but he has native-like competence in English; and (3) The speaker is a
native speaker of English (Item 2 in ‘The Respondent’s Evaluation of the Speech
Samples’ section in Appendix A). The data analysis stage involved calculating the
mean score of the ‘accentedness’ data for each sample and taking a count of the
responses for each of the three statements regarding the ‘nativeness’/‘nonnativeness’
of the speaker of each sample. The mean scores of the samples for ‘accentedness’ are
given in Table 2, and information regarding the perceived
71
‘nativeness’/‘nonnativeness’ of the different speakers is given in Table 3. Visual
representations of the two sets of details are given in Figures 4 and 5.
Table 2 and Figure 4 indicate a marked difference between STA and the rest
of the samples in terms of ‘accendetness’. While the score for STA is between ‘a
slightly noticeable accent’ and a ‘fairly noticeable accent’, the scores for the rest of
the samples are between ‘a quite strong accent’ and ‘a very strong accent’. Based on
these data, it could be argued that English speech with “native” but “nonnormative”
accents and English speech with non-American accents that are associated with
“nonnative” contexts fall into the same category in terms of accentedness.
Table 2
The Mean Scores of Different Speech Samples for Accentedness
Sample
Mean Score
STA
1.58
CHN
4.31
IND
4.21
APP
4.42
AFA
4.21
KEN
4.20
72
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
STA
CHN
IND
APP
AFA
KEN
Figure 4. The perceived accentedness of the different speech samples.
As far as the perceived ‘native’/‘nonnative speaker’ statuses of the speakers
are concerned (as indicated by Table 3 and Figure 5), the respondents have recognized
the speakers of STA, APP, and AFA as ‘native speakers’ and the speakers of CHN,
IND, and KEN as ‘nonnative speakers’. While a percentage of participants as high as
98.16 have recognized the speakers of STA and APA as ‘native speakers’, 91.41% of
participants have recognized the speaker of AFA as a ‘native speaker’. Although it is
interesting to see that close to 9% have said that the speaker of AFA is not a ‘native
speaker’, the high percentage that AFA has recorded for the ‘native speaker’ option
points to the strong tendency to view the speaker of AFA as a ‘native speaker’ of
English. The high percentages that CHN, IND, and KEN have recorded for the
‘nonnative speaker’ options (more than 99% in the case of CHN and about 94% in the
case of IND and KEN) point to the dominant tendency to recognize the speakers of
these samples as ‘nonnative speakers’ of English. The 60% that IND and KEN have
recorded for the ‘nonnative speaker with native-like competence’ as opposed to the
73
38% that CHN has recorded for the same option points to variation that exists within
the ‘nonnative speaker’ category.
The ‘accentedness’ data and ‘native/nonnativeness data’ combined show that
although the respondents tend to categorize “nonnormative” American accents with
non-American accents associated with “nonnative” English speaking contexts, in
relation to ‘accentedness’, they clearly draw a line between the two groups in terms of
‘native’/‘nonnativeness’. The fact that all 10 FGI participants recognized the speakers
of APP and AFA as ‘native speakers’ of English who necessarily ‘spoke the language
with accents’, while recognizing the speaker of STA as a ‘native speaker’ and the
speakers of CHN, IND, and KEN as ‘nonnative speakers’ without necessarily
referring to the ‘accentedness’ of those samples confirms this claim.
180
160
140
120
Nonnative speaker
100
80
Nonnative speaker, but
native-like cometernce
60
Native speaker
40
20
0
STA
CHN
IND
APP
AFA
KEN
Figure 5. The perceived ‘nativeness’/‘nonnativeness’ of the speakers of the speech
samples.
74
In a context where accent is considered the most observable marker of
distinction between different forms of fluent English speech, it could be argued, on
the basis of this finding, that the perceived accentedness of a speaker does not
function as an indicator of that person’s status as a ‘native’/‘nonnative speaker’.
Although the perceived accentedness does not function as an indicator of a person’s
‘native’/‘nonnative speaker’ status, the data indicate that the participants are able to
determine which accents come under the ‘native speaker’ category and which accents
come under the ‘nonnative speaker’ category. In that sense, a person’s accent could
be seen as carrying an indication of her/his ‘native’/‘nonnative speaker’ status. The
study emphasizes the need for further research in this area to gain better insights into
the nature of the connection between accent/accentedness and ‘native’/‘nonnative
speakerness’.
Total
speaker
Nonnative
competence
native-like
speaker with
Nonnative
Native speaker
163
0
1.84
3
0
98.16
%
160
No.
STA
1
162
99
62
No.
CHN
60.74
38.04
0.61
%
9
163
57
97
No.
IND
34.97
59.51
5.52
%
The Perceived ‘Nativeness’/‘Nonnativeness’ of the Speakers of the Speech Samples
Table 3
162
1
1
160
No.
APP
0.61
0.61
98.16
%
163
9
5
149
No.
AFA
5.52
3.07
91.41
%
7
162
56
99
No.
KEN
34.36
60.74
4.29
%
75
76
4.3. Research Question 3
The aim of the third research question was to find out if a person’s perceived
accent functioned as an indicator of her/his place of origin. The data for this research
question were obtained by asking the respondents to select a region (largely at a
continental level), from a list of given regions, that they would associate each speaker
with (Item 9 in ‘The Respondent’s Evaluation of the Speech Samples’ section in
Appendix A). Table 4 presents the data collected using this question. In the case of
those speakers that they recognized as from ‘North America (USA)’, the participants
were asked to state which specific area in the USA that they thought the speaker was
from (Item 10 in ‘The Respondent’s Evaluation of the Speech Samples’ section in
Appendix A).
In the case of STA, APP, and AFA, the participants clearly recognized the
speakers as from ‘North America (USA)’. The high percentages of responses that this
option has recorded for these three speech samples (as shown in Table 4) indicate the
participants’ ability to distinguish between American speakers and non-American
speakers of English. This is further confirmed by the fact that the participants did not
recognize the speakers of CHN, IND, and KEN as from ‘North America (USA)’
(except for the 1.84% and 1.23% that recognized the speakers of IND and KEN
respectively as from the USA).
Of those who categorized STA under ‘North America (USA)’, 75% said that
the speaker was from ‘Midwest’, while 15% said that he was from the
‘northern’/‘northeastern’ area of the USA. Nine of the 10 FGI participants stated that
the speaker was from ‘Midwest’, while the other one said that he was from the
‘northern’/‘northeastern’ area. In the case of APP, 55% said that the speaker was
77
from the ‘Appalachian region’ (Appalachia and the bordering States), while 40% said
that he was from the ‘southern area’ of the country. In the FGIs, half of the
participants said that the speaker was from ‘Appalachia’, while the other half said that
he was from ‘South’. The study notes that the category ‘South’/‘southern area’ is a
broad category that is marked for internal diversity, and, in the case of APP, it could
well be the case that what the participants meant by ‘South’/‘southern area’
overlapped with the ‘Appalachian region’.
As far as AFA is concerned, the data with regard to the specific place/s in the
USA that the speaker was associated with failed to present patterns as clear as those
found in the case of STA and APP. In the case of AFA, 30% said that the speaker
was from ‘South’, while 17% said that he was from any ‘urban’/‘inner-city’/‘big city’
area (without specifying any geographic location). Five participants (3%) said that
the speaker could be from any area where ‘African American English’/‘AAVE’ was
spoken. About 19% of the participants either said that they were not sure where the
speaker was from or refrained from answering that question. However, quite contrary
to the sense of uncertainty manifested in the survey data with regard to AFA, almost
all the FGI participants said that the speaker was from ‘South’.
As far as the perceived places of origin of the speakers are concerned, the
recognizable patterns that emerge in the data for STA and APP and the absence of a
clearly recognizable pattern in the data for AFA point to a couple of significant
perceptions regarding the three accented speech forms in question. The relatively
clear tendency to associate STA with Midwest and APP with the Appalachian region
indicates the presence of a ‘mental map’ of dialect areas (Diercks, 2002; Preston,
1986) in the minds of the participants. It also shows that the boundaries of the space
78
associated with the forms of English that the participants perceived STA and APP to
represent are geographically defined. Although one could argue that the form of
English that STA represents is no longer limited to a particular region, the general
emphasis on the historical link between that form of English and the geographical
area referred to as Midwest seems to play a significant role in defining the mental
map for that form of English.
Interestingly, the participants’ mental map for AFA appears to be markedly
different from those for STA and APP. The absence of a clearly recognizable pattern
in the data for AFA with regard to the speaker’s perceived place of origin could, in a
way, be seen as problematizing the presence of a mental map for AFA; however, the
fact that 30% of the participants and all 10 FGI participants have stated that the
speaker is from ‘South’ gives evidence for the presence of some form of mental map.
That said, the tendency of that 30% of survey participants and 10 FGI participants to
place AFA under the relatively broad banner of ‘South’ without associating it with a
more specific region, and the fact that the data from the rest of the participants fail to
show any recognizable pattern with regard to the speaker’s perceived place of origin
show that the mental map for AFA is not defined geographically to the same extent
that the mental maps for STA and APP are. While the considerable number of
responses that have associated the speaker with ‘urban’/‘inner-city’/‘big city’ areas in
general indicates the tendency to define the AFA mental map along the ‘urban versus
non-urban’ line, the comments that associate the speaker with any area where
‘African American English’/‘AAVE’ is spoken hint at the possibility of the African
American ethnic identity being a defining feature of that mental map. The study
highlights the need for further research in this regard.
79
As far as the speech samples with non-American accents are concerned,
74.85% of the participants (and 8 FGI participants) correctly recognized the speaker
of CHN as from Asia, 57.06% of the participants (and 5 FGI participants) correctly
recognized the speaker of IND as from Asia, and 41.10% of the participants (and 4
FGI participants) correctly recognized the speaker of KEN as from Africa. According
to the data, the majority of the participants have correctly recognized the place of
origin (at a continental level) of the speakers of CHN and IND. In the case of KEN,
only less than half of the participants have correctly placed the speaker in relation to
his place of origin (at a continental level). However, the fact that the percentage for
the ‘Africa’ option is higher than the percentages for each of the other options in the
list indicates the possibility of the speaker’s speech carrying some indication of his
place of origin in the view of the participants who are “native speakers” of American
English.
In a context where accent is considered the most observable marker of
distinction between/among different forms of fluent speech, the results discussed in
this section indicate that a person’s accent functions, for the most part, as an indicator
of her/his place of origin. They also point to accent as a ‘trigger’ of the mental maps
(in the sense that an accent activates the relevant mental map in the mind of the
listener) that people have regarding speech styles or accents associated with different
regions/countries.
1.23
0.61
0
2
1
0
Europe
South America
Australia
(Outside of USA)
North America
(USA)
0.61
97.55
159
1
0
0
Africa
North America
0
0
%
Asia
No.
STA
0
9
24
1
0
7
122
No.
CHN
0
5.52
14.72
0.61
0
4.29
74.85
%
IND
1
12
16
3
3
35
93
No.
The Perceived Places of Origin of the Speakers of the Speech Samples
Table 4
0.61
7.36
9.82
1.84
1.84
21.47
57.06
%
0
1
0
0
1
5
156
No.
APP
0
0
0.61
3.07
95.71
0
0.61
%
2
0
2
3
5
7
144
No.
AFA
0
1.23
1.84
3.07
4.29
88.34
1.23
%
1
15
10
3
2
67
64
No.
KEN
0.61
9.20
6.13
1.84
1.23
41.10
39.26
%
80
81
4.4. Research Question 4
The fourth research question aimed at finding out how “native speakers” of
American English rated fluent speech with “normative”/“nonnormative” US accents
and non-American accents associated with “nonnative” speaking contexts. In
addition to exploring how the participants assessed the ‘accentedness’ of the samples
and the ‘native’/‘nonnative speaker’ status and place of origin of the speakers, the
study also looked at how they assessed the samples in terms of the ‘fluency’,
‘competence’, ‘intelligibility’, ‘correctness’, and ‘pleasantness’ of the
speaker/speaker’s English (Items 3, 1, 5, 6, and 7 respectively in ‘The Respondent’s
Evaluation of the Speech Samples’ section in Appendix A). In the case of
‘competence’, the participants assessed the samples by selecting one of the five given
options (‘bad’, ‘poor’, ‘fair’, ‘good’, and ‘excellent’), while in the case of
‘intelligibility’, ‘correctness’, and ‘pleasantness’, they assessed the samples on a fivevalue scale, 1 being ‘not intelligible at all’, ‘not correct at all’, and ‘very unpleasant’
respectively, and 5 being ‘totally intelligible’, ‘totally correct’, and ‘very pleasant’
respectively. As for ‘fluency’, the participants responded to the questions “Would
you consider this speaker a fluent speaker of English?” with either ‘yes’ or ‘no’. In
the analysis stage, the mean score of each sample for fluency was calculated as out of
5 in order to make those scores comparable with the scores for ‘competence’,
‘intelligibility’, ‘correctness’, and ‘pleasantness’. Table 5 presents the mean scores of
each sample in terms of those features, while Figures 6-10 present visual
representations of those scores.
The data indicate that the participants have placed STA above all the other
samples in terms of ‘fluency’, ‘competence’, ‘intelligibility’, ‘correctness’, and
82
‘pleasantness’. Provided that these features are indicators of the quality of a given
speech sample, the high mean scores for these features, which are between the values
4 and 5, indicate the general impression among the participants that the form of
English that STA represents (relatively “accent neutral”, “standard” American
Midwestern English) is the ‘best’ form of English. The similar responses that the FGI
participants gave confirm this claim.
Table 5
The Mean Scores of Different Speech Samples for Fluency, Competence,
Intelligibility, Correctness, and Pleasantness
STA
CHN
IND
APP
AFA
KEN
Fluency
5.00
2.08
4.20
4.70
4.45
4.00
Competence
4.61
3.23
3.90
3.40
3.03
3.90
Intelligibility
4.34
3.55
3.83
3.36
2.71
4.03
Correctness
4.39
3.29
4.06
3.12
2.68
4.15
Pleasantness
4.28
3.01
3.47
3.03
2.60
3.65
83
6
5
4
3
2
1
0
STA
CHN
IND
APP
AFA
KEN
AFA
KEN
Figure 6. The fluency rates of the different speech samples.
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
STA
CHN
IND
APP
Figure 7. The competence rates of the different speech samples.
84
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
STA
CHN
IND
APP
AFA
KEN
Figure 8. The intelligibility rates of the different speech samples.
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
STA
CHN
IND
APP
AFA
Figure 9. The correctness rates of the different speech samples.
KEN
85
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
STA
CHN
IND
APP
AFA
KEN
Figure 10. The pleasantness rates of the different speech samples.
Although APP and AFA are in the same category as STA in the sense that the
speakers of these samples are Americans whom the participants recognized as “native
speakers” of English, they have clearly been rated below STA in the areas of fluency,
competence, intelligibility, correctness, and pleasantness. APP and AFA have
recorded high scores in the area of fluency (4.70 and 4.45 respectively), which place
them below STA (with a fluency score of 5.00) but clearly above IND, KEN, and
CHN (with fluency scores of 4.20, 4.00, and 2.08 respectively). Although APP and
AFA have scored relatively high scores in the area of fluency, they have failed to
maintain the same success rate in the other areas. AFA has recorded the lowest rates
in all these four areas, and APP has recorded the second lowest in intelligibility and
correctness. Although APP’s scores for competence and pleasantness rank APP
slightly above CHN, APP has been ranked clearly below IND and KEN in those
areas.
86
The participants’ responses to the question “How do you assess your English
in relation to the speaker’s English?” (Item 8 in ‘The Respondent’s Evaluation of the
Speech Samples’ section in Appendix A) in the case of each sample provide further
insights into the way the participants ranked the samples. According to the data
presented in Table 6 and Figure 11, the participants have shown a clear tendency to
identify their English with the form of English in the STA sample (with 91.41%
selecting the option ‘His English is more or less similar to my English’). The fact that
APP and AFA have recorded only 25.15% and 15.95% respectively for the option
‘His English is more or less similar to my English’, which are markedly lower than
the percentages IND and KEN have recorded for the same option (40.49 and 46.62)
indicates the participants’ tendency not to identify themselves with the speakers of
APP and AFA. At the same time, the fact than APP and AFA have recorded
percentages as high as 74.23 and 83.44 respectively for the option ‘My English [a.k.a.
the participant’s] is better than his English [a.k.a. the speaker’s English]’, which are
much higher than 57.67% and 47.85% that IND and KEN have recorded for the same
option respectively, indicates the extent to which the participants distance themselves
from the speakers of APP and AFA and their tendency to view the forms of English
represented by APP and AFA as being inferior not only to their own English, but also
to the forms of English represented by IND and KEN, which they recognized as
‘nonnative speaker’ forms of English.
The data indicates a clear tendency on the part of the participants to distance
themselves from the speaker of CHN and view the form of English represented by
CHN as inferior to their English. However, as far as the other non-American samples
(IND and KEN) are concerned, the participants have shown a tendency to view the
87
forms of English represented by those samples in a relatively favourable light. The
percentages of 5.52 and 1.84 that KEN and IND have recorded respectively for the
option ‘His English [a.k.a. the speaker’s English] is better than mine’, especially
compared to the 1.23% that STA has recorded for the same option, points to a slight
tendency to accept the possibility of the forms of English represented by KEN and
IND, forms that were predominantly judged to have been produced by ‘nonnative’
English speakers, being superior to the participants’ English.
180
160
140
120
My English is better than his
English
100
His English is more or less
similar to my English
80
His English is better than
mine
60
40
20
0
STA
CHN
IND
APP
AFA
KEN
Figure 11. The perceived quality of the speakers’ English in relation to Respondents’
English.
Total
English
Better than his
My English is
English
similar to my
more or less
His English is
better than mine
His English is
2
163
12
149
No.
STA
7.36
91.41
1.23
%
1
163
149
13
No.
CHN
91.41
7.98
0.61
%
3
163
93
66
No.
IND
57.67
40.49
1.84
%
The Perceived Quality of the Speakers’ English in Relation to Respondents’ English
Table 6
1
163
121
41
No.
APP
74.23
25.15
0.61
%
1
163
136
26
No.
AFA
83.44
15.95
0.61
%
9
163
78
76
No.
KEN
47.85
46.62
5.52
%
88
89
The participants’ responses to the question “Assuming that this speaker is a
teacher of English, would you recommend him as a teacher to a foreigner coming from
a non-English speaking country?” (Item 11 in ‘The Respondent’s Evaluation of the
Speech Samples’ section in Appendix A) also provide further insights into the way the
participants ranked the different samples. The data (Table 7 and Figure 12) indicate
that a percentage as high as 82.21% have responded in the affirmative in the case of the
STA speaker. As far as APP and AFA are concerned, a large majority of the
participants (90.80% and 93.87% respectively) have said that they would not
recommend the speakers as English teachers to a foreigner. The better scores that
KEN, IND, and CHN have recorded for this question indicate the participants’
preference for “nonnative speakers” over “native speakers” with “nonnormative”
accents as teachers of English.
In the FGIs, 9 of the 10 participants said that they would recommend the STA
speaker as an English teacher. In the case of CHN, IND, and KEN, the participants
had a mixed response. About half of the participants responded in the affirmative, the
rest responded in the negative. In the case of APP and AFA, all the participants
responded in the negative without the slightest hesitation. When the question was
asked from the participants regarding CHN in one FGI, 2 participants said that they
would recommend the speaker as an English teacher, but they would prefer a ‘native
speaker’ to that speaker. However, when the same question was asked from them
regarding the speakers of APP and AFA (speakers whom the participants had judged to
be ‘native speakers’ of English) they clearly said that they would not recommend those
speakers as English teachers. These apparently contradictory responses indicate their
reluctance to recognize those “native speakers” of English who speak the language
90
with “nonnormative” accents as an authority on the language, despite their position as
“native speakers” of the language.
180
160
140
120
100
No
80
Yes
60
40
20
0
STA
CHN
IND
APP
AFA
KEN
Figure 12. The chances of the participants recommending the speakers as English
teachers to a foreigner from a non-English speaking country.
82.21
16.56
134
27
161
No
Total
%
Yes
No
STA
37
162
125
No
CHN
76.69
22.70
%
81
82
163
No
IND
49.69
50.31
%
15
163
148
No
APP
90.80
9.20
%
8
161
153
No
AFA
93.87
4.91
%
73
88
161
No
KEN
The Chances of the Participants Recommending the Speakers as English Teachers to a Foreigner from a Non-English Speaking Country
Table 7
44.79
53.99
%
91
The study also tried to see if the participants would associate the forms of English
represented by the different speech samples with different professions. The question
asking the participants to select three professions (from a list of given professions) which
they thought the speaker would fit in without his English being an issue (Item 12 in ‘The
Respondent’s Evaluation of the Speech Samples’ section in Appendix A) was used to
obtain data in that regard. As the data presented in Table 8 and Figure 13 show, the
respondents have indicated certain patterns in their responses. The most striking pattern
could be seen with regard to the option ‘Janitor’. Of 163 participants, 140 and 138 have
selected ‘Janitor’ for APP and AFA respectively. While 52 have selected ‘Janitor’ for
CHN, only 16, 26, and 16 have selected that option for STA, IND, and KEN respectively.
While all FGI participants chose this option for APP and AFA, some of them
supplemented their choice with statements like “I hate to do this, but this is the bitter
truth”. Interestingly, only 2 of the 10 participants selected this option for the nonAmerican samples, and no one chose it for STA.
An entirely different pattern could be observed in relation to the option of
‘Doctor’. While 39, 55, 66, and 74 participants selected ‘Doctor’ for STA, CHN, IND,
and KEN respectively, only 3 and 5 have selected that option for APP and AFA
respectively. Roughly similar patterns could be recognized for the professions of
‘Banker’ and ‘Teacher’. An interesting pattern could be seen in relation to the option of
‘Radio/TV Broadcaster’. While 71 participants have selected this option for STA, 27 and
20 have selected the option for APP and AFA respectively. Only 11, 6, and 5 have
selected the same option for KEN, IND, and CHN respectively. In a context where APP
and AFA had predominantly been seen as less favourable compared not only to STA but
93
also to the three non-American samples, the preference that the participants have shown
for APP and AFA over the non-American samples in relation to the option of ‘Radio/TV
Broadcaster’ is interesting. As 3 FGI participants stated, this preference for American
forms of English speech over non-American forms of English speech could probably be
based on the participants’ desire to hear ‘familiar’ voices on the radio and television.
Two participants said that they would like to hear voices like those presented in the APP
and AFA samples more as a secondary voice rather than as a primary voice.
450
400
Banker
350
Doctor
300
Janitor
250
IT professional
200
Lab assistant
150
Receptionist
100
Radio/TV broadcaster
Teacher
50
0
STA
CHN
IND
APP
AFA
KEN
Figure 13. The professions that the speakers would fit in without their English being an
issue.
94
Table 8
The Professions that the Speakers Would Fit in without Their English Being an Issue
STA
Teacher
Radio/TV
CHN
IND
APP
AFA
KEN
104
45
74
26
16
98
71
5
6
27
20
11
51
23
50
27
41
18
47
106
83
36
38
84
46
79
65
17
39
71
16
52
26
140
134
16
39
55
66
3
5
74
48
31
39
23
13
35
Broadcaster
Receptionist
Lab Assistant
IT Professional
Janitor
Doctor
Banker
The results of the study presented in this section indicate a clear preference for
STA on the part of the participants. The results also point to their assumption that the
form of English that STA represents (relatively “accent neutral”, “standard” American
Midwestern English) is the ‘best’ form of English. As far as the rest of the samples are
concerned, the results indicate a marked preference for the non-American samples
associated with “nonnative” English-speaking contexts over the American “native
speaker” samples that are marked for “nonnormative” accents.
These results indicate the participants’ underlying assumption that a (‘proper’)
‘native speaker’ of English is not just any ‘native speaker’, but a ‘native speaker’ who
95
speaks a particular form of English in a particular way. Although they recognize the
speakers from the United States who speak English with “nonnormative” accents as
‘native speakers’ of English, those speakers, in their view, are inferior to the those
speakers who speak the language with accents that are close to standard American
English speech. This situation points not only to the inherent heterogeneity of the ‘native
speaker’ category, but also to the clear accent-based stratification within that category.
The fact that the “nonnormative” accents have been rated lower than the non-American
accents in many respects points to the intensity of this stratification.
96
CHAPTER 05: CONCLUSION
The study does not project its findings to be representative of the opinions and
perceptions of all “native speakers” of American English regarding the notion of ‘native
speakerness’, nor does it consider itself to be exhaustive of all possible dimensions of the
notion in question. The study, however, considers itself to have explored the notion in a
substantial manner and pointed to certain dominant ways in which the notion is perceived
in the United States. Given the facts that ‘native speakerness’ is more a social construct
than an objective reality based on empirically observable facts and that the best (if not,
only) way in which such a social construct could be explored is through an investigation
of the social perceptions of the construct, the study considers its findings and the claims
that are made on the basis of those findings to be valid and contributing towards a
broader understanding of the notion of ‘native speaker’.
The participants’ tendency to recognize the speakers with American accents as
‘native speakers’ and the speakers with non-American accents as ‘nonnative speakers’ on
the one hand and view those who speak English with “nonnormative” accents as inferior
to the one who speaks the language with a “normative” accent on the other hand points to
the inherent complexities/tensions that the notion of ‘native speaker’ embodies. It also
points to the two simultaneous standardization processes that Doerr (2009b) describes.
While the tendency to label American speakers and non-American speakers as ‘native
speakers’ and ‘nonnative speakers’ respectively indicates the first standardization process
(standardization on the basis of the L1 and non-L1 distinction), the tendency to rank
97
American accents indicates the second standardization process (standardization based on
an internal hierarchy within the ‘native speaker’ category). The fact that the ‘native
speaker’ status of those who spoke English with “nonnormative” American accents had
to be supplemented with the additional information that ‘they spoke the language with an
accent’ when the speaker with the “normative” American accent and the speakers with
the non-American accents associated with “nonnative” English-speaking contexts could
be directly recognized as a ‘native speaker’ and ‘nonnative speakers’ respectively without
having to associate their statuses with the additional piece of information regarding their
accentedness indicates the undefined “special” position that the simultaneous operation of
the two standardization processes have assigned American speakers with “nonnormative”
accents. The fact that the ‘native’/‘nonnative speaker’ dichotomy fails to accommodate
this “special” position renders the “native speakers” who speak the language with
“nonnormative” accents ‘spaceless’ in the ‘native speaker’ discourse.
The inferior position that the “nonnormative” American accents and their
speakers have earned not only compared to the “normative” American accent and its
speaker, but also compared to the non-American accents associated with “nonnative”
English-speaking contexts and their speakers challenges the sense of homogeneity that
the label ‘native speaker’ attaches to/imposes on the unity that it refers to. The vastness
of the disparity between “native speakers’” perception of “normative” and
“nonnormative” accents of American English and their clear tendency to rate
“nonnormative” American accents lower than non-American accents associated with
“nonnative” English-speaking contexts point to the absence of a logical basis for the
98
notion of ‘native speaker’ and the ‘native’/‘nonnative speaker’ dichotomy. This observed
absence of a logical basis for the notion strengthens two significant arguments; the first
argument is specific to linguistics as a discipline and the second argument largely
concerns the “extra-linguistic” realm.
The fact that linguistics, which is generally understood as the scientific study of
language, has failed to provide an empirically verifiable definition for the notion in
question creates a space in which the observed absence of a logical basis for the notion
gets translated into an absence of a scientific basis for the same notion. In the light of this
absence of a logical/scientific basis for the notion, the evident uncritical acceptance of the
notion as a taken-for-granted reality in linguistics makes one question the validity of
linguistics as a scientific discipline.
At an “extra-linguistic” level (‘extra-linguistic’ in the sense of going beyond the
strict confines of linguistics), this observed absence of a logical, empirically verifiable
basis for the ‘native speaker’ concept exposes the politically and ideologically defined
nature of the concept. The findings of the study support Irvine and Gal’s (2000) theory
of the three ‘interconnected semiotic processes’ and Pennycook’s (1994) claim regarding
three ‘native speaker ideologies’. The “native speaker” respondents’ ability to recognize
mainly the geographical spaces and also the social spaces associated with the different
accents, and their tendency to recognize all American samples as ‘native speaker’
samples indicate an instance of ‘iconization’ (Irvine & Gal, 2000, p. 403). They also
support Pennycook’s (1994) claim regarding there being “a close correspondence
99
between holding the citizenship of a nation-state and being a native speaker of the
national language of that nation-state” (p. 176).
Despite the vast disparities that the participants recognized between the sample
with a “normative” accent and the two samples with “nonnormative” accents, the
tendency to categorize the three American samples as ‘native speaker’ samples, thereby
separating them from ‘nonnative speaker’ samples, also indicates an instance of ‘fractal
recursivity’ (Irvine & Gal, 2000, p. 403). It indicates a situation where the ‘native
speaker’ concept is used to conceal the vast differences that exist between the American
sample with a “normative” accent and the other two American samples (those with
“nonnormative” accents) by ‘abstracting’ the difference from the ‘native speaker’
category and projecting it onto a different level, a level where the difference between
‘native speakers’ and ‘nonnative speakers’ gets highlighted. To conceptualize this
process in a different way, the excessive focus placed on the ‘native’/‘nonnative speaker’
dichotomy (by both society and the academia) distracts an observer’s attention from the
vast disparities that exist among different groups within the ‘native speaker’ category.
Pennycook’s second ideology (“being a native speaker automatically bestows one with a
high level of competence in all domains of one’s first language” [1994, p. 176]), which
appears to emerge from the semiotic process of ‘fractal recursivity’, indicates one
argument that is widely used to unite different American English speaker groups into one
group and distance them from non-American English speaker groups. The findings of the
present study have drawn attention towards the politically and ideologically defined
100
nature of the ‘native speaker’ concept by showing that there is no logical/scientific basis
for this categorization/unification.
The tendency to use the term ‘native speaker’ to refer primarily to the speaker
with a “normative” American accent while setting him apart from the speakers with
“nonnormative” American accents by recognizing them not just as ‘native speakers’, but
as ‘native speakers who spoke the language with an accent’ indicates an instance of
‘erasure’ (Irvine & Gal, 2000, p. 404). This additional information that the participants
used to define the speakership of the speakers of “nonnormative” American accents is
usually overlooked and considered less important, if not unimportant, in attempts to
understand a speech community in terms of the ‘native’/‘nonnative speaker’ dichotomy.
Those speakers could be seen being absorbed into the broader ‘native speaker’ speech
community, which is usually centred around and dominated by the “normative”
American English speakers with “normative” accents. This absorption erases or dilutes
the identity of the “nonnormative” speakers. Pennycook’s (1997) third ideology, which
conceptualizes language as “a homogeneous and a fixed system with a homogeneous
speech community” (p. 176), runs parallel to this semiotic process of ‘erasure’. The
homogenization that the notion of ‘native speaker’ carried out within the ‘native speaker’
category involves an obvious privileging of the ‘standard speaker’ and ‘standard
language’ against the ‘nonstandard speakers’ and ‘nonstandard forms of the language’.
The findings of the study indicate that the forms of division that exist within the
‘native speaker’ category, especially those involving ‘standard’/‘nonstandard speakers’,
are more intense than those that exist between the two categories (‘native speakers’ and
101
‘nonnative speakers’). However, due to the fact that the primary focus is on the
‘native’/‘nonnative speaker’ dichotomy, these intense forms of division within the ‘native
speaker’ category tend to pass unnoticed. This situation points to the need for more
critical approaches (like Kramsch, 1997; Kubota, 2009; Pennycook, 1994; Takato, 2009;
Train, 2009) to the study of the concept, especially those approaches that recognize the
concept as a construct. The present study believes that attempts to re-form/“reform” the
‘native speaker’ concept or the ‘native’/‘nonnative speaker’ dichotomy, while treating
them as fundamentally tenable realities, fail to be effective as they fail to acknowledge
and address issues related to division and discrimination at an intra-group level in a
substantial manner. Therefore, the study emphasizes the need for more studies that
recognize the notion of ‘native speaker’ first and foremost as a basis for division and
discrimination and aim at deconstructing the notion. Such a deconstruction would
automatically invalidate the ‘native’/‘nonnative speaker’ dichotomy.
The findings of the present study point to the need to liberate linguistics from the
clutches of absolute positivism and establish it as a more socially grounded discipline that
recognizes the power relations that define society. Academic studies have a vital role to
play in bringing about this transformation. One way in which academic studies can
contribute to this transformation is by exposing the value-laden nature of the taken-forgranted concepts in the field. This kind of awareness-raising is important for the
practitioners in the field to revisit certain dominant understandings that define their line
of thinking. Academic studies can also contribute to this transformation by attempting to
reconceptualize the fundamental concepts in the field in such a way that the concepts
102
become accurate representations of the realities that they are meant to represent. The
present study considers itself to have made its contribution in its small way towards this
transformation.
103
REFERENCES
Aitchison, J. (2001). Language change: Progress or decay? (3rd ed.). Cambridge:
Cambridge University Press.
Anderson, B. (1991). Imagined communities: Reflections on the origin and spread of
nationalism (Rev. ed.). London: Verso.
Bakhtin, M. (1981). The dialogic imagination: Four essays. Austin: University of Texas
Press.
Bloomfield, L. (1933). Language. New York: Holt Rinehart Winston.
Bourdieu, P. (1977). The economics of linguistic exchanges. Social Sciences Information,
16(6), 645-668.
Brown, J. D. (2001). Using surveys in language programs. Cambridge: Cambridge
University Press.
Brown, K. (1992). American college student attitudes towards non-native instructors.
Multilingua, 11(3), 249-265.
Calvet, L. (1998). Language wars and linguistic politics. Oxford: Oxford University
Press.
Carnie, A. (2007). Syntax: A generative introduction. Malden, MA: Blackwell
Publishing.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, Massachusetts: The
MIT Press.
Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL
Quarterly, 33(2), 183-209.
104
Davies, A. (1991a). The native speaker in applied linguistics. Edinburgh: Edinburgh
University Press.
Davies, A. (1991b). The notion of native speaker. Journal of Intercultural Studies, 12(2),
35-45.
Davies, A. (2003). The native speaker: Myth and reality. Clevedon: Multilingual Matters.
Derwing, T. M. (2003). What do ESL students say about their accents? Canadian
Modern Language Review, 59, 547-566.
Derwing, T. M., & Rossiter, M. J. (2002). ESL learners’ perceptions of their
pronunciation needs and strategies. System, 30, 155-166.
Diercks, W. (2002). Mental maps: Linguistic-geographic concepts. In D. Long & D. R.
Preston (Eds.), Handbook of perceptual dialectology: Volume 2 (pp. 51-70).
Philadelphia: John Benjamins North America.
Dixon, J. N., & Cocks, B. M. R. (2002). Accent of guilt? Effects of regional accent, race,
and criminal type on attributions of guilt. Journal of Language and Social
Psychology. 21(2), 162-168.
Doerr, N. M. (2009a). Investigating ‘native speaker effects’: Toward a new model of
analyzing ‘native speaker’ ideologies. In N. M. Doerr (Ed.), The native speaker
concept: Ethnographic investigations of native speaker effects (pp. 15-46). Berlin:
Mouton de Gruyter.
Doerr, N. M. (2009b). Uncovering another ‘native speaker myth’: Juxtaposing
standardization processes in first and second languages of English-as-a-SecondLanguage learners. In N. M. Doerr (Ed.), The native speaker concept:
105
Ethnographic investigations of native speaker effects (pp. 185-208). Berlin: Mouton
de Gruyter.
Dore, J. (1986). The development of communicative competence. In R. L. Schiefelbusch
(Ed.), Language competence: Assessment and intervention (pp. 3-60). San Diego,
California: College-Hill Press.
Dornyei, Z. (2007). Research Methods in Applied Linguistics: Quantitative, Qualitative,
and Mixed Methodologies. Oxford: Oxford University Press.
Eadie, T. L., & Doyle, P. C. (2002). Direct magnitude estimation and interval scaling of
pleasantness and severity in dysphonic and normal speakers. The Journal of the
Acoustical Society of America, 112(6), 3014-3021.
Edwards, J. (1999). Refining our understanding of language attitudes. Journal of
Language and Social Psychology, 18, 101-110.
Fanon, F. (1967). Black skin, white masks. (C. L. Markmann, Trans.). London : Pluto
Press. (Original work published 1952).
Fillmore, C. J. (1979). On fluency. In D Kempler & W. S Y. Wang (Eds. ), Individual
differences in language ability and language behaviour (pp. 85-102). New York:
Academic Press.
Firth, A., & Wagner, J. (1997). On discourse, communication, and (some) fundamental
concepts in SLA research. The Modern Language Journal, 81(3), 285-300.
Fishman, J. (1971). Sociolinguistics: A brief introduction. Boston: Rowley.
106
Fridland, V., Bartlett, K., & Kreuz, R. (2005). Making sense of variation: Pleasantness
and education ratings of southern vowel variants. American Speech, 80(4), 366387.
Fridland, V. (2008). Regional differences in perceiving vowel tokens on Southernness,
education, and pleasantness ratings. Language Variation and Change, 20, 67-83.
Gass, S. M., & Selinker, L. (2008). Second language acquisition: An introductory Course
(3rd ed.). New York: Routledge.
Giles, H., & Johnson, P. (1987). Ethnolinguistic identity theory: A social psychological
approach to language maintenance. International Journal of the Sociology of
Language, 68, 69-99.
Gluszek, A., & Dovidio, J. F. (2010). Speaking with a nonnative accent: Perceptions of
bias, communication difficulties, and belonging in the United States. Journal of
Language and Social Psychology, 29(2), 224-234.
Han, Z. (2004). To be a native speaker means not to be a nonnative speaker. Second
Language Research, 20(2), 166-187.
Hawley, M. E. (1977). Introduction. In M. E. Hawley (Ed.), Speech intelligibility and
speaker recognition (pp. 1-11). Stroudsburg, Pennsylvania: Dowden, Hutchinson &
Ross, Inc.
Irvine, J. T., & Gal, S. (2000). Language ideology and linguistic differentiation. In P. V.
Kroskrity (Ed.), Regimes of language: Ideologies, polities, and identities (pp. 3583). Santa-Fe: School of American Research Press.
107
Jenkins, J. (2000). The phonology of English as an international language. Oxford, UK:
Oxford University Press.
Jenkins, J. (2005). Implementing an international approach to English pronunciation: The
role of teacher attitudes and identity. TESOL Quarterly, 39(3), 535-543.
Kachru, B. B. (1985). Standards, codification and sociolinguistic realm: The English
language in the outer circle. In R. Quirk & H. G, Widdowson (Eds.), English in the
world (pp. 11-30). Cambridge: Cambridge University Press.
Kachru, B. B. (1986). The alchemy of English. Oxford: Pergamon Press.
Kandiah, T. (1998a). Why new Englishes? In J. A. Foley, T. Kandiah, B. Zhiming, A. F.
Gupta, L. Alsagoff, H. C. Lick, L. Wee, I. S. Talib, & W. Bokhorst-Heng (Eds.),
English in new cultural contexts: Reflections from Singapore (pp. 1-40). London:
Sage Publications.
Kandiah, T. (1998b). The emergence of new Englishes. In J. A. Foley, T. Kandiah, B.
Zhiming, A. F. Gupta, L. Alsagoff, H. C. Lick, L. Wee, I. S. Talib, & W. BokhorstHeng (Eds.), English in new cultural contexts: Reflections from Singapore (pp. 73105). London: Sage Publications.
Kenworthy, J. (1987). Teaching English pronunciation. London: Longman.
Knickerbocker, H., & Altariba, J. (2011). Bilingualism and the impact of emotion: The
role of experience, memory, and sociolinguistic factors. In V. Cook & B. Bassetti
(Eds.), Language and bilingual cognition (pp. 453-478). New York: Psychology
Press.
108
Kormos, J., & Denes, M. (2004). Exploring measures and perceptions of fluency in the
speech of second language learners. System, 32, 145-164.
Kramsch, C. (1997). The privilege of the nonnative speaker. PMLA, 112(2), 359-369.
Kubota, R. (2009). Rethinking the superiority of the native speaker: Toward a relational
understanding of power. In N. M. Doerr (Ed.), The native speaker concept:
Ethnographic investigations of native speaker effects (pp. 233-247). Berlin: Mouton
de Gruyter.
Kuiper, L. (2005). Perception is reality: Parisian and provencal perceptions of regional
varieties of French. Journal of Sociolinguistics, 9(1), 28-52.
Lee, J. J. (2005). The native speaker: An achievable model? Asian EFL Journal, 7(2),
152-163. Retrieved from http://www.asian-efljournal.com/June_2005_EBook_editions.pdf
Lennon, P. (1990). Investigating fluency in EFL: A quantitative approach. Language
Learning, 40, 387-412.
Lennon, P. (2000). The lexical element in spoken second language fluency. In H.
Riggenbach (Ed.), Perspectives on fluency (pp. 25-42). Michigan: The University
of Michigan Press.
Leung, C. (2005). Convivial communication: Recontextualizing communicative
competence. International Journal of Applied Linguistics, 15(2), 119-144.
Lev-Ari, S., & Keysar, B. (2010). Why don’t we believe non-native speakers? The
influence of accent on credibility. Journal of experimental social psychology, 46,
1093-1096.
109
Lindemann, S. (2002). Listening with an attitude: A model of native speaker
comprehension of non-native speakers in the United States. Language in Society,
31, 419-441.
Lippi-Green, R. (1994). Accent, standard language ideology, and discriminatory pretext
in the courts. Language in Society, 23, 163-198.
Lippi-Green, R. (1997). English with an accent. London: Routledge.
McKay, S. L., & Bokhorst-Heng, W. D. (2008). International English in its
sociolinguistic contexts: Towards a socially sensitive EIL pedagogy. New York:
Routledge.
Mugglestone, L. (1995). ‘Talking proper’: The rise of accent as social symbol. Oxford:
Clarendon Press.
Munro, M. J., & Derwing, T. M. (1995). Processing time, accent, and comprehensibility
in the perception of native and foreign-accented speech. Language and Speech, 38,
289-306.
Myhill, J. (2004). A parameterized view of the concept of ‘correctness’. Multilingua, 23,
389-416.
Ndebele, N. (1987). The English language and social change in South Africa. The
English Academy Review, 4, 1-16.
Niedzielski, N., & Preston, D. R. (2000). Folk linguistics. Berlin: Mouton de Gruyter.
Paikeday, T. M. (1985). The native speaker is dead. Toronto & New York: Paikeday
Publishing Inc.
110
Parakrama, A. (1995). De-hegemonizing language standards: Learning from
(post)colonial Englishes about ‘English’. London: Macmillan.
Pawley, A., & Syder, F. H. (1983). Two puzzles for linguistic theory: Nativelike selection
and nativelike fluency. In J. C. Richards & R. W. Schmidt (Eds.), Language and
communication (pp. 317-331). Amsterdam: Elsevier.
Pennycook, A. (1994). The cultural politics of English as an international language.
London: Longman.
Pennycook, A. (1998). English and the discourses of colonialism. Oxon: Routledge.
Pennycook, A. (2003). Beyond homogeny and heterogeny: English as a global and
worldly language. In C. Mair (Ed.), The politics of English as a world language
(pp. 3-17). Amsterdam: Rodopi.
Phillipson, R. (1992). Linguistic imperialism. Oxford: Oxford University Press.
Preston, D. R. (1986). Five visions of America. Language in Society, 15(2), 221-240.
Preston, D. R. (1989). Perceptual dialectology: Non-linguists’ views of aerial linguistics.
Dordrecht: Holland, Foris.
Preston, D. R. (1993). Folk dialectology. In D. R. Preston (Ed.), American dialect
research (pp. 333-378). Amsterdam: Benjamins.
Preston, D. R. (1996). Where the worst English is spoken. In E. W. Schneider (Ed.),
Focus on the USA (pp. 297-360). Amsterdam: Benjamins.
Rehbein, J. (1987). On fluency in second language speech. In H. W. Dechert & M.
Raupach (Eds.), Psycholinguistic models of production (pp. 97-105). Norwood:
Ablex.
111
Riches, P., & Foddy, M. (1989). Ethnic accent as a status cue. Social Psychology
Quarterly. 52(3), 197-206.
Rubin, D. L. (1992). Nonlanguage factors affecting undergraduates’ judgements of
nonnative English-speaking teaching assistants. Research in Higher Education, 33,
511-531.
Rubin, D. L. & Smith, K. A. (1990). Effects of accent, ethnicity, and lecture topic on
undergraduates’ perceptions of nonnative English speaking teaching assistants,
International Journal of Intercultural Relations, 14, 337-353.
Ryan, E. B., Hewstone, M., & Giles, H. (1984). Language and intergroup attitudes. In J.
Eiser (Ed.), Attitudinal judgement (pp. 135-160). New York: Springer.
Sajavaara, K. (1987). Second language speech production: Factors affecting fluency. In
H. W. Dechert & M. Raupach (Eds.), Psycholinguistic models of production (pp.
45-65). Norwood: Ablex.
Sapir, E. (1929). The status of linguistics as a science. Language, 5(4), 207-214.
Schmidt, R. (1992). Psychological mechanisms underlying second language fluency.
Studies in second language acquisition, 14, 357-385.
Shapiro, M. (1989). A political approach to language purism. In B. Jernudd & M. Shapiro
(Eds.), The politics of language purism (pp. 21-29). Berlin: Mouton de Gruyter.
Takato, M. (2009). ‘Native speaker’ status on border-crossing: The Okinawan Nikkei
diaspora, national language, and heterogeneity. In N. M. Doerr (Ed.), The native
speaker concept: Ethnographic investigations of native speaker effects (pp. 83100). Berlin: Mouton de Gruyter.
112
Train, R. (2009). Toward a ‘natural’ history of the native (standard) speaker. In N. M.
Doerr (Ed.), The native speaker concept: Ethnographic investigations of native
speaker effects (pp. 47-82). Berlin: Mouton de Gruyter.
Williams, R. (1976). Keywords: A vocabulary of culture and society. New York &
London: Fontana & Croom Helm.
Woolard, K. A. (1998). Introduction: Language ideology as a field of inquiry. In B. B.
Schieffelin, K. A. Woolard, & P. V, Kroskrity (Eds.), Language ideologies:
Practice and theory (pp. 3-47). New York: Oxford University Press.
113
APPENDIX A: STRUCTURED QUESTIONNAIRE
Information about the Respondent
1. Sex (tick the correct option):
___ Male
___Female
2. What is your major? _______________________________________________
3. Select the statement(s) that is/are true about you:
I have taken LING 270 offered by the Linguistics Department of OU.
I have taken LING 351 offered by the Linguistics Department of OU.
I have taken at least one course in Linguistics (LING) other than LING 270
and LING 351.
I have taken at least one course in Linguistics OUTSIDE of the Linguistics
Department of OU.
I have NEVER taken a course in linguistics.
4. Where are you from?
Cincinnati Ohio
Cleveland Ohio
Columbus Ohio
Other
City: __________
State: __________
5. Are you a native speaker of English?
___Yes
___ No
6. Do you speak English with an accent?
___Yes
___ No
7. Of the following labels, tick the one that would best describe your ethnic identity:
___
___
___
___
___
African American
Asian American
European American
Hispanic/Latino American
Native American
114
___ White American
___ Non-American
8. How would you define a native speaker of English?
________________________________________________________________
The Respondent’s Evaluation of the Speech Samples
Instructions: The survey will have 6 one-minute speech samples. At the end of each
speech sample, you will be asked to answer a set of multiple-choice questions about
the sample. Please note the questions are NOT about the ideas/facts discussed in the
speech, BUT about the quality of the speech. [Please note that the speakers are all
American residents.]
CLIP 01 [The same questions were asked regarding all six clips.]
1. What can you say about the speaker’s English competence?
___
___
___
___
___
Bad
Poor
Fair
Good
Excellent
2. Based on your assessment of the speaker, select one of the following statements
about the speaker’s language competence.
___
___
___
The speaker is not a native speaker of English.
The speaker is not a native speaker, but he has native-like
competence in English.
The speaker is a native speaker of English.
3. Would you consider this speaker a fluent speaker of English?
___
___
Yes
No
115
4. Does the speaker have an accent?
1
2
No accent
at all
3
4
5
A very
strong accent
3
4
5
Totally
intelligible
5. Is the speech intelligible?
1
2
Not intelligible
at all
6. Does the speaker’s English sound correct?
1
2
Not correct
at all
3
4
5
Totally
correct
7. Does the speaker’s English sound pleasant?
1
2
Very
unpleasant
3
4
5
Very
pleasant
8. How would you assess your English in relation to the speaker’s English?
___
___
___
His English is better than my English.
His English is more or less similar to my English.
My English is better than his English.
9. Where do you think this speaker is from?
___
___
___
___
___
___
___
Asia
Africa
North America (USA)
North America (outside of USA)
Europe
South America
Australia
116
10. If the answer to the previous question is North America (USA), please specify the
region.
__________
11. Assuming that this speaker is a teacher of English, would you recommend him as a
teacher to a foreigner coming from a non-English speaking country?
___
___
Yes
No
12. Which profession do you think the speaker fit without his English being an issue?
You may choose up to 3 options.
___
___
___
___
___
___
___
___
Teacher
Radio/TV Broadcaster
Receptionist
Lab Assistant
IT Professional
Janitor
Doctor
Banker
117
APPENDIX B: OHIO UNIVERSITY CONSENT FORM (ONLINE SURVEY)
Title of Research: The Role Accent Plays in the Evaluations of Native Speaker-hood by
Native Speakers of American English
Researcher: Nandaka Maduranga Kalugampitiya
You are being asked to participate in research. For you to be able to decide whether you
want to participate in this project, you should understand what the project is about, as
well as the possible risks and benefits in order to make an informed decision. This
process is known as informed consent. This form describes the purpose, procedures,
possible benefits, and risks. It also explains how your personal information will be used
and protected. Once you have read this form and your questions about the study are
answered, you will be asked to print your name. This will allow your participation in this
study. You should receive a copy of this document to take with you.
Explanation of Study
This study is being done with a view to exploring the role accent plays in the
evaluations of native speaker-hood by native speakers of American English.
If you agree to participate, you will be asked to listen to six speech samples and
respond to a set of questions listed after each recording.
You should not participate in this study if you have had any formal training in
linguistics before.
Your participation in the study will last 30-40 minutes.
Risks and Discomforts
The risk or discomfort that you might experience is fatigue from completing the
survey.
Benefits
This study is important to science/society because it tries to recognize the role accent
plays in identity construction. The conclusion and recommendations of the study will
point to socially favorable ways in which different accents could be dealt with.
118
Individually, you may benefit in the form of getting to experience different
forms/varieties of English and observing the differences between those
forms/varieties.
Compensation
In addition, if you are a student of LING 270 (2011/12 Winter), your participation in
this study will earn you an extra credit worth of 10 points towards your LING 270
final grade.
Confidentiality and Records
Your study information will be kept confidential by recording your responses
anonymously and without possibility of identification.
Additionally, while every effort will be made to keep your study-related information
confidential, there may be circumstances where this information must be shared with:
* Federal agencies, for example the Office of Human Research Protections, whose
responsibility is to protect human subjects in research;
* Representatives of Ohio University (OU), including the Institutional Review
Board, a committee that oversees the research at OU;
* Dr Peter Githinji – the advisor of the present study
Contact Information
If you have any questions regarding this study, please contact me (740-856-1757;
[email protected]) or my advisor Dr Peter Githinji (740-274-0950;
[email protected]).
If you have any questions regarding your rights as a research participant, please
contact Jo Ellen Sherow, Director of Research Compliance, Ohio University,
(740)593-0664.
By signing below, you are agreeing that:
• you have read this consent form (or it has been read to you) and have been
given the opportunity to ask questions and have them answered
• you have been informed of potential risks and they have been explained to
your satisfaction.
• you understand Ohio University has no funds set aside for any injuries you
might receive as a result of participating in this study
• you are 18 years of age or older
• your participation in this research is completely voluntary
• you may leave the study at any time. If you decide to stop participating in the
119
study, there will be no penalty to you and you will not lose any benefits to
which you are otherwise entitled.
[ ] CHECK THIS BOX to ACCEPT terms of the study and to create an 'Electronic
Signature' confirming your voluntary participation.
120
APPENDIX C: OHIO UNIVERSITY CONSENT FORM (FGI)
Title of Research: The Role Accent Plays in the Evaluations of Native Speaker-hood by
Native Speakers of American English
Researcher: Nandaka Maduranga Kalugampitiya
You are being asked to participate in research. For you to be able to decide whether you
want to participate in this project, you should understand what the project is about, as
well as the possible risks and benefits in order to make an informed decision. This
process is known as informed consent. This form describes the purpose, procedures,
possible benefits, and risks. It also explains how your personal information will be used
and protected. Once you have read this form and your questions about the study are
answered, you will be asked to sign it. This will allow your participation in this study.
You should receive a copy of this document to take with you.
Explanation of Study
This study is being done with a view to exploring the role accent plays in the
evaluations of native speaker-hood by native speakers of American English.
If you agree to participate, you will be asked to listen to six speech samples and
respond to a set of questions listed after each recording.
You should not participate in this study if you have had any formal training in
linguistics before.
Your participation in the study will last 30-40 minutes.
Risks and Discomforts
The risk or discomfort that you might experience is fatigue from completing the
survey.
Benefits
This study is important to science/society because it tries to recognize the role accent
plays in identity construction. The conclusion and recommendations of the study will
point to socially favorable ways in which different accents could be dealt with.
Individually, you may benefit in the form of getting to experience different
forms/varieties of English and observing the differences between those
forms/varieties.
121
Confidentiality and Records
Your study information will be kept confidential by recording your responses
anonymously and without possibility of identification.
Additionally, while every effort will be made to keep your study-related information
confidential, there may be circumstances where this information must be shared with:
* Federal agencies, for example the Office of Human Research Protections, whose
responsibility is to protect human subjects in research;
* Representatives of Ohio University (OU), including the Institutional Review
Board, a committee that oversees the research at OU;
* Dr Peter Githinji – the advisor of the present study
Contact Information
If you have any questions regarding this study, please contact me (740-856-1757;
[email protected]) or my advisor Dr Peter Githinji (740-274-0950;
[email protected]).
If you have any questions regarding your rights as a research participant, please
contact Jo Ellen Sherow, Director of Research Compliance, Ohio University,
(740)593-0664.
By signing below, you are agreeing that:
• you have read this consent form (or it has been read to you) and have been
given the opportunity to ask questions and have them answered
• you have been informed of potential risks and they have been explained to
your satisfaction.
• you understand Ohio University has no funds set aside for any injuries you
might receive as a result of participating in this study
• you are 18 years of age or older
• your participation in this research is completely voluntary
• you may leave the study at any time. If you decide to stop participating in the
study, there will be no penalty to you and you will not lose any benefits to
which you are otherwise entitled.
Signature
Date
Printed Name
Version Date: [insert mm/dd/yy]
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
Thesis and Dissertation Services
!