Vol1 PDF

Roberts, Paul (2005) Spoken English as a world
language: international and intranational settings. PhD

thesis, University of Nottingham.
Access from the University of Nottingham repository:

http://eprints.nottingham.ac.uk/11110/1/423642_vol1.pdf
Copyright and reuse:
The Nottingham ePrints service makes this work by researchers of the University of
Nottingham available open access under the following conditions.
This article is made available under the University of Nottingham End User licence and may
be reused according to the conditions of the licence. For more details see:
http://eprints.nottingham.ac.uk/end_user_agreement.pdf
For more information, please contact [email protected]

Spoken English as a World Language: international and intranational settings
SPOKEN ENGLISH AS A WORLD LANGUAGE: INTERNATIONAL AND

INTRANATIONAL SETTINGS
Paul Roberts B. A.
Thesis submitted to the University of Nottingham

for the degree of Doctor of Philosophy
August 2005
NGy
G2`FR J
ITY
Abstract
This thesis sets out to characteriseEnglish as a World Language, in contrast with
English used in homogeneous,intranational settings. After a brief introduction, the
relevant literature is reviewed in two chapters: firstly the concept of an international
variety of English is challenged and, following this, there is consideration of current
thinking under the headingsof English as an International Language and English as
Lingua Franca. This preliminary part of the thesis leads to some hypotheses
concerning the way in which EWL might be characterised,with particular attention to
attitudes among different sorts of speaker.
ChapterFive introducesmethodologies(1) for finding data-providing participants and
(2) assessingtheir language-relatedattitudes relevant to the research questions. It
continues by (3) examining ways of obtaining spoken data and (4) of transcribing and
(5) analysing it. Chapter 6 presentsthe specific methodological choices for this thesis.
The following four chapters provide results. Firstly, brief results are given of tests
applied to ascertainparticipants' language-relatedattitudes. Following this, the results
of analysing and explaining the spoken data itself are given. Chapter 8 closely
compares one EWL conversation with one homogeneousone and draws tentative
conclusions about what might be found in the remaining conversations: that EWL
may be characterisedby greater convergenceamong irrespective

speakers, of whether
or not they are native speakers.Chapter 9 examinesthe whole suite of conversations
in this light and the previous results are generally confirmed: the speakers in
homogeneousconversationstend to be as divergent as they are convergent, where in
they try their bestto maintainan atmosphereof comity. Chapter

EWL conversations
10 completesthe results sectionby comparingthe performanceof six speakersin
ii
particular, who each participate in an EWL conversationand in a homogeneousone.
They are found to draw on convergencestrategiesfor their EWL conversationswhile
being more direct and divergent in their homogeneousones. Chapter 11 attempts to
the
summarise precedingchaptersand to draw someconclusions from the results.
111
Acknowledgements
I would like above all to acknowledgethe excellent, inspiring tuition I received for
this thesis from my supervisor,ProfessorMichael McCarthy.
I would also like to acknowledgethe following:
the large number of studentsand staff at the University of Hertfordshirewho
consentedto take part in the various stages of data-gathering on which this thesis
rests;
the School of Combined Studies at the University of Hertfordshire which made my
work possible by providing resources,most importantly time resources,allowing the
work to be carried out; colleagues in the School were also of invaluable help in
sounding out ideas, reading parts of the thesis and being generally supportive of my
work;
the Ethics Committee in the Humanities Faculty at the University of Hertfordshire
who allowed me to go aheadand gatherdata from human subjects;
Dr. Tim Parke, in the Humanities Faculty at the University of Hertfordshire who
made useful suggestions during the early stages of writing.
IV
CONTENTS
Chapter 1: Introduction
1. ChangingEnglish, Changing view of English 2

..........................................................
1.1 The growing number of people speakingEnglish 2
.........................................
1.2 Researchin English studiesrelevant to the changes 4
......................................
2. This researchthesis in context 7
..................................................................................
3. Overview 10
.................................................................................................................
Chapter 2: Is English as a World Language a Variety of English?
1. Introduction 15
.............................................................................................................
20
2. StandardEnglish
.....................................................................................................
21
2.1 Defining StandardEnglish
...........................................................................
2.2 Descriptive problems 25
...................................................................................
2.3 Ideological problems 29
....................................................................................
31
2.4 Conclusion
...................................................................................................
3. World StandardEnglish 32
..........................................................................................
4. International English 33
...............................................................................................
English language 33
4.1 International - as a standard ............................................
4.2 International English -a Lingua Franca 34
......................................................
4.3 A specially-constructed International English 36
.............................................
4.4 Conclusion 37
...................................................................................................
38
5. World English
.........................................................................................................
43
6. Global English
........................................................................................................
43
7. Conclusion..............................................................................................................
Chapter 3: English as a ..... Language
1. Introduction 46
.............................................................................................................
2. EFL, ESL and ENL 47
.................................................................................................
2.1 People,countries,role and function............................................................. 47
2.2 Confusion over EFL, ESL and ENL 49
............................................................
2.3 The supremacyof the Native Speaker 50
.........................................................
2.4 ENL, ESL and EFL - Conclusion 51
...............................................................
3. EIL and ELF 51
...........................................................................................................
3.1 EIL 51
...............................................................................................................
3.2 ELF 53
..............................................................................................................
3.2.1ELF is built on the LF construct 53
3.2.2 Users ELF .........................................................
54
of
.......................................................................................
321PT. FicgFnnctinn 55
V
3.2.4 ELF and the conceptof `common core' 56

.............................................
3.2.5 ELF as a languagevariety 58
...................................................................
4. EWL - English as a World Language 61
...................................................................
4.1 EWL Users 62
...................................................................................................
4.1.1 Difficulties defining native speakers 62
..................................................
4.1.2 Native speakersare a significant minority 70
..........................................
4.1.3 Four casestudies 71
.................................................................................
4.2 EWL as a function 73
.......................................................................................
4.3 EWL and the variety and `core' questions 74
...................................................
5. Conclusion 75
..............................................................................................................
Chapter 4: Hypotheses concerning EWL interactions
1. Introduction 77
.............................................................................................................
2. EWL asymmetricalconversations 78
........................................................................
..
2.1 The centrality of the native speaker 79
...........................................................
..
2.1.1 Ownership 80
..
.........................................................................................
2.1.2 Authority 82
...........................................................................................
..
2.1.3 Learner preferences,teacherpredilections 83
..
.......................................
2.2 NNS and NS in EWL interactions 85
.............................................................
..
2.2.1 Accommodation 85
................................................................................
..
2.2.2 ForeignerTalk 88
...................................................................................
..
2.2.3 Loss of confidence 89
............................................................................
..
3. EWL Symmetrical conversations 90
.........................................................................
..
3.1 Equality among speakers 91
...........................................................................
..
3.2 Accommodation 91
.........................................................................................
..
3.2.1 The basesof ELF accommodation 92
..
...................................................
3.2.2 Culture transfer 92
.................................................................................
..
3.2.3 ELF Strategies 94
...................................................................................
..
3.2.4 Community construction 95
...................................................................
..
3.2.5 A sharedform of language 97
..
................................................................
4. Symmetrical interactions including non-native speakersand native speakers:
asymmetriesamong non-native speakers.................................................. 98
..
5. Conclusion 100
............................................................................................................
Chapter 5: Gathering and analysing data
1. Introduction 103
...........................................................................................................
2. Selectingdata-producingparticipants 105
...................................................................
3. Exploring language-basedattitudes and perceptions 106
............................................
3.1 Discussions 107
................................................................................................
3.2 Listening and questionnaires 109
.....................................................................
4. Homogeneousgroups for recording conversations 110
..............................................
5. Data-gatheringand transcription methods 111
............................................................
5.1 Data-gathering 111
...........................................................................................
5.2 Transcription 114
..............................................................................................
V1
6. Analysis of the conversationdata: word counting and annotation 115

....:..................
6.1 Superordinateannotationheadings:convergenceand divergence............116
6.2 Convergence:accommodation 116
...................................................................
6.3 Convergence:speechacts 117
..........................................................................
6.4 Convergence:discoursemarkers 117
...............................................................
6.5 Convergence:other strategicphenomena 119
..................................................
6.6 Divergence 120
.................................................................................................
6.7 Conversationannotation 121
............................................................................
7. Analysis of the conversationdata: explanation 121
....................................................
7.1 Setting/Ends 122
...............................................................................................
7.2 Interpersonalexplanations 123
.........................................................................
7.2.1 Deterministic approaches 123
..................................................................
7.2.2 Constructionistapproaches 126
...............................................................
8. Three explanatoryformats 127
....................................................................................
Chapter 6: Results: preliminary discussionsand questionnaires
1. Introduction 130
...........................................................................................................
2. Discussions 130
...........................................................................................................
2.1 The native speakerconstruct 131
......................................................................
2.1.1 Inheritanceand nationality 131
................................................................
2.1.2 Proficiency 132
........................................................................................
2.2 Ownership and authority 133
............................................................................
2.3 Asymmetrical interactions 136
.........................................................................
137
2.4 Summary
....................................................................................................
138
3. Listening and questionnaires
................................................................................
3.1 Group 1 138
......................................................................................................
3.2 Group 2 138
......................................................................................................
3.3 Group 3 139
......................................................................................................
3.4 Group 4 139
......................................................................................................
3.5 Group 5 139
......................................................................................................
3.6 Group 6 140
......................................................................................................
3.7 Group 7 140
......................................................................................................
3.8 Group 8 140
......................................................................................................
3.9 Group 9 141
......................................................................................................
3.10 Group 10 141
..................................................................................................
3.11 Summary 142
..................................................................................................
4. Conclusion 142
............................................................................................................
Chapter 7: Results: A comparison between and international and an

intranational conversation
1. Introduction 145
...........................................................................................................
2. EWL Conversation6 146
............................................................................................
2.1 Summaryand further explanation 154
.............................................................
3. Conversationamong homogeneousNorwegian usersof English 157
........................
VII
3.1 Summaryand further explanation 163

.............................................................
4. Comparisonof the two conversations 164
...................................................................
4.1 Vocabulary 165
.................................................................................................
4.2 Conversationalstyle 166
...................................................................................
4.2.1 EWL Conversation6 167
........................................................................
4.2.2 HomogeneousconversationamongNorwegian speakers 167
................
5. Preliminary conclusions 169
........................................................................................
Chapter 8: Results: Overview of all conversations
1. Introduction 171
...........................................................................................................
2. Vocabulary 172
............................................................................................................
3. Discourse strategies 174
..............................................................................................
4. Conversationalstyle 175
..............................................................................................
4.1 Achievement of interactional goals 175
...........................................................
4.1.1 Laughter and humour 176
........................................................................
4.1.2 Colloquial languageand vaguelanguage 177
.........................................
4.1.3 `We', Ùs' and `You' 180
........................................................................
4.1.4 Accommodation, collaborative turns and backchannels 181
...................
4.1.5 Explicit agreementand concessions 185
.................................................
4.1.6 Maintaining comity despitetransactionalgoals 187
................................
4.2 Divergent behaviour 189
..................................................................................
4.2.1 Challenging 190
.......................................................................................
4.2.2 Demands,not suggestions 190
.................................................................
4.2.3 Closure 191
..............................................................................................192
4.2.4 Other divergent behaviour
................................................................
4.3 High involvement and high consideratenessstyles 194
...................................
4.3.1 Involvement and consideratenessstyles in EWL conversations 194
......
4.3.2 Involvement and consideratenessstyles in homogeneous
conversations................................................................................. 197
4.4 Conclusion 201
.................................................................................................
5. Explanations..........................................................................................................202
5.1 Sex/Gender 202
.................................................................................................
5.2 Cultural background 203
..................................................................................
5.3 Perceptionsof the meeting generic framework 206
....................................
5.3.1 Leaders 207
..............................................................................................
5.3.2 Other, non-leaderroles 210
.....................................................................
5.4 Interactants' perceptionsof eachother 213
......................................................
6. Conclusion 215
............................................................................................................
Chapter 9: Results: six casestudies
1. Introduction 217
...........................................................................
................................
2. Greta and Bai 219
........................................................................................................
2.1 Participation and role 219
.................................................................................
2.2 Use of words 223
..............................................................................................
vii'

...................................................................................
2.4 Greta and Bai - Conclusion 229
.......................................................................
3. Comfort and Shray 230
................................................................................................
2.1 Participation and role 230
.................................................................................
2.2 Use of words 233
..............................................................................................
...................................................................................
2.4 Comfort and Shray- Conclusion 239
...............................................................
4. Betty and Milne 240
....................................................................................................
240
2.1 Participation and role
.................................................................................
2.2 Useof words 244
..............................................................................................
247
2.3 Conversationalstyle
...................................................................................
2.4 Betty and Milne - Conclusion 250
...................................................................
5. Conclusion 250
............................................................................................................
Chapter 10: Conclusion
1. Summary 252
...............................................................................................................
254
2. Discussion
.............................................................................................................
256
2.1 Anthropological and sociological universals
.............................................
2.2 Universal manifestations 257
..................................... ....................................... 258
2.3 The native speaker question
.......................................................................
2.4 International English revisited 259
...................................................................
261
References................................................................................................................
ix
Chapter One
Introduction
Chapter 1I Introduction
Spoken English as a World. Language: international and intranational settings
1. Changing English, Changing view of English
1.1 The Growing number of people speaking English
It is a commonplace notion that language changes and a widely held view that
language spread and language change form one unified concept (Brutt-Griffler
2002:109-110), but the nature and speed of recent changes in English are, when
to
compared past changes,quite extraordinary. Whereas two hundred years ago the
number of people using English, whether written or spoken was limited to Britain
to
and a few pockets of colonial settlers,by 1950 the number had increasedto about
900 million and the figure for the year 2000 seemsto be in the region of 2 billion
(Graddol 1997). Graddol predicts that this number is set to increaseby a great deal
more, suggesting that between 2005 and 2010, `nearly a third of the world
population will all be trying to learn English at the sametime' (Graddol 2004:5).
It is also generally agreed that this exponential and continuing increase in the
number of users of English has taken two distinct directions. On the one hand,
change has meant the fragmentation of English into diverse varieties, often referred
to under the heading `World Englishes' or `New Englishes', while on the other, the
rapid change has also led to unification, under the heading `World English' or
`Global English' (Preisler 1995, Graddol 1996, Toolan 1997, Brutt-Griffier
2002:176, Bolton 2003:4,).
This thesis does not concern itself, or concerns itself only tangentially, with the
former result of language change and will not deal centrally with either `local'
standardEnglishes or `local' vernaculars(Graddol 1996). Instead,it will focus on the
latter result of change,the apparentcoming together of English users and whatever it
is thatthey share,ratherthanwhatthey do not.
Chapter 12 Introduction
The current notion of English as a World language has little to do with the
suggestionand prediction made in 1832 by JacobGrimm, that `the English language
may with all right be called a world-language; and, like the English people, appears
destined hereafterto prevail with a sway more extensive even than its presentover
all portions of the globe' (Grimm 1832,cited in Pennycook, 1994:70). In connecting
English to `the English people', Grimm was, accurately, predicting the colonial
spreadof that nation and its language.The current linguistic unification of the world
has several explanations, of which the British colonial past is but one (Graddol
2004). There are, however, still similarities with English changesbrought about by
colonialism: chief among them is that English as a World Language (as opposedto
World Englishes) is, still, the currency of an elite or those aspiring to be in an elite.
Fishman, one of the first to write about English in its new guise, commentsthat the
spreadof English may have a lot `to do with the growing dominance of the richer
countries over the poorer ones (and not merely economically or particularly
politically, but also culturally)' (Fishman et. al. 4),

1996: Toolan describes Global
English as `the languageof international high-flyers' (Toolan 1997) and McArthur,
somewhatreluctantly perhaps,concludesthat `what is sharedworldwide, in fact, is
not the languageat large but a standardvariety common to the media, business,and
is to
what one constrained call, for want of a better phrase, a Western-educated
international elite' (McArthur 1996:14). Pennycook, too, considers that those
participating in World English are, in fact, minorities who, by implication, are at the
top of their relative social hierarchies(Pennycook2002).
The important political and cultural question concerning the extent to which the use
of English is a liberating experience, allowing movement across hierarchies and the
expansion of the elite to the potential inclusion of all or, conversely, a constraining
and excluding phenomenon,serving to limit some people to being subordinate to
others, will barely be touched on here. Suffice it to say that, as part of the global,
cultural flow of persons, information, technology, finance, ideology and ideas
(Appadurai 1996), English is clearly part of the `liberation' of some while it is part
of the suppressionof others.
1.2. Research in English studies relevant to the changes
Researchand intellectual interest in the fragmentation result of recent changesin
English has been rich and diverse for many decades.Much of this interest has
followed well-establishedtraditions, treating `World Englishes' or `New Englishes'
in the sameway as English has always been studied. Bolton provides a useful chart,
plotting the different approachesand grouping them according to their intellectual
roots. He thus fits the study of `World Englishes' into Ènglish Studies', Ènglish
corpus linguistics', `Sociolinguistic approaches', `Kachruvian studies', `Pidgin and
Creole studies', Àpplied linguistics', `Lexicography', `Popularisers', `Critical
linguistics' and `Linguistic futurology' (Bolton 2003:42-43).
The other result of the changes, the unificatory one, does not fit so easily into these
traditions (although some studies have attemptedto make it do so), nor has it had a
tradition of its own. In 1994,Ammon noted, in his entry Ìnternational Languages'in
`The Encyclopedia of Language and Linguistics', that '.... there is no significant
research tradition within which a widely accepted concept (of international
languages)could have been established'(Ammon 1994: 1725).
Ammon's assertion was made more than ten years after the pioneering research
carried out by Larry Smith in the late 1970s.Smith suggestedthat English as an

International Language might be taught to native and non-native speakers alike,
Chapter14 Introduction
serving the purpose of communicating across national and cultural boundaries
(Smith, L. 1976,1978,1979,1981). But inasmuch as Smith was working within
what Bolton has called `Kachruvian Studies', it could be claimed that he was, after
all, seeking to describe and to propagateanother English in the fragmented field of
World Englishes.
Around the time that Ammon published his encyclopoediaentry, Pennycook's work
emergedas one of the first to acknowledgewhat he calls a 'bottom-up' approachto
World English, based on people using English, rather than on a description of the
English language or a variety of it (Pennycook 1994). Pennycook's work, and the
earlier study by Phillipson on linguistic imperialism (Phillipson 1992), are both
included in Bolton's table under the `Critical linguistics' heading; Pennycook's
study stands out, however, as having little to do with the establishment, whether or
not through imperialist design, of `New Englishes' and a great deal to do with the
post-modern, world-wide `discourse' of English. More recently, Brutt-Griffler has
attempted to bring together the development of World Englishes and World English
by refuting Phillipson's claims (Brutt-Griffler 2002).
Since 1994, there have been several smaller-scalepieces intent on problematising
World English in different ways. Three examples may be found in the review
Ènglish Today' and one in `NeusprachlicheMittteilungen': in the former journal,
Ahulu suggestsa re-definition of `StandardEnglish' in the context of English being
used internationally (Ahulu 1997), Toolan, referred to above, focuses on the
distinction betweenWorld Englishesand World English (Toolan 1997) and Modiano
proposes an alternative to Kachru's three circles which accounts for global, as
opposed to local English competence (Modiano 1999a); in the German publication,
Gnutzmann suggests that what he calls Global English 'refers to a set of
communicative contexts definable by extralinguistic factors' (Gnutzmann 1998: 133).
Alongside journal articles such as these, there has indeed grown up, since Ammon's
piece on International Languages, what may be thought of as a research tradition
where the focus of interest is the way in which English is used to achieve
communication across national and cultural boundaries. An early precursor to later
work may be found in a paper by Campbell, Ekniyom, Haque and Smith, whose
research focused on problems encountered in various teaching institutions where the
main thrust of the learning was towards using English internationally (Campbell et.
al. 1983). But one of the first attempts seriously to describe what happens when
English is used internationally came with Firth's 1990 use of the `Lingua Franca'
label and the proposal to use an interactional approach to its description (Firth 1990).
This was followed by a paper in 1996 where Firth attempted to establish a principle
of `Lingua Franca' English, the `let-it-pass' principle (Firth 1996). The `Lingua
Franca' label has been enthusiastically taken up by other researchers such as House,
Jenkins, Meierkord, Lesznyäk and Mauranen. House follows directly on from Firth,
questioning and re-examining the `let-it-pass' principle (House 1999 and 2002b)
while Jenkins provides one of the very few book-length studies, taking up the
`Lingua Franca' label and proposing a set of phonological `core' items which may
characterise it (Jenkins 2000b). Christiane Meierkord was the first to produce a
helpful bibliography of English used in this way and has gone on to produce several
brief studies of `Lingua Franca' small talk, the contribution of speakers' home
culture to English and the extent to which participants in `Lingua Franca'
conversations pool features from their home varieties of English (Meierkord 2000,
2002,2004). Seidlhofer has been concerning herself with building up a corpus of
`Lingua Franca' English while also contributing to the discussion of how it may be
characterised(Seidlhofer 2000,2001,2002) and Mauranen has, similarly, begun the
setting up of a `Lingua Franca' corpus (Mauranen 2003). Finally, Lesznyäk has
contributed a book-length study of topic management in Lingua Franca settings
(Lesznyäk 2004).
As well as seekingto characterise

or describeEnglishas it currentlyexistsfor the
sake of satisfying intellectual, scientific requirements,there is an underlying current
in much researchwhich hopesto identify norms, or at least patterns in contemporary
global English which could provide models for teaching purposes. Here too, the
distinction between the fragmentation tradition, with its emphasis on World
Englishes, and the more recent research tradition concerned with unification, or
World English, is crucial. Many of those working within the former field show a
concern to establish local norms for the teaching of English, which reflect local use
(Bamgbose 1998). On the other hand, researcherssuch as Seidlhofer and Jenkins are
keen to establish principles which may inform English language teaching for cross-
national, cross-cultural purposes(Jenkins 2000b, Seidlhofer 2001, Dürmiiller 2003).
This endeavourrequires norms, if indeed they can be established,which are neither
local, in the World Englishes sense, nor traditionally connected to native speakers.
1 2. This research thesis in context
This thesis is situated firmly in the areaexplored by Campbell et. at., by Firth and by
those following him. More precisely, it is to be seenin the context of the work done
by Firth, by House, by Meierkord and by Lesznyäk, in that the focus is on spoken
English acrossnational and cultural boundaries,the approach is mostly at discourse
level (rather than at phonological, lexical or grammatical level) and the ultimate
application is to teachingEnglish for global purposes.
The researchcarried out thus far in the field has, however, been founded on the
tradition of dividing English usersinto distinct groups. The division with the longest
tradition assignsusersto either the native speakeror the non-native speakergroup; a
later traditions breaks users up into three groups: users of English as a national
language, as a second language and as a foreign language. Kachru began the
tradition of re-naming these three groups as inner circle, outer circle or expanding
circle users (Kachru 1985). Others have attempted further re-naming (see, for
example, Jenkins 2000b).
Whereas these divisions have been questioned and problematised (Rampton 1990,
Medgyes 1992, Singh et. al. 1998) they have nevertheless underlain much of the
research into World English. In particular, `native speakers' and ÈSL' users of
English (or Òuter Circle' users) have been mostly excluded from the research
carried out by Firth, House, Meierkord and Jenkins and continue to be almost
excluded from the lingua franca corpora being developed by Seidlhofer and
Mauranen. This tendency towards exclusion may be a reaction to the previous
tradition whereby English spoken by anyone but so-called native speakers was
always considered against native-speaker English. Indeed, Lesznyäk's study,
concerned with characterising English as Lingua Franca, uses non-native speaker
interactions with native-speakersas its point of contrast: Lingua Franca English is
defined in opposition to English used by, or in the presence of, native speakers
(Lesznyäk 2004).
The conclusions so far achieved concerning discoursal characteristics of World
English are conclusions about the way expanding circle users achieve
communicative aims when speaking with other expanding circle users (just as the
conclusions concerning the phonology of English as an International Languageare
based on expanding circle speakers Jenkins 2000b). One of the reasonsfor this
-
limitation seemto be that expanding circle users constitute the majority of English
speakersand that, as the label suggests,will soon constitute an even greatermajority.
Another reasonhas to do with the pedagogicimplications of the research:in order to
train languagelearnersfor a global future, it has been rightly consideredthat a native
speakermodel or a `New English' model, is quite inappropriate; this has, it seems,
led to the exclusion of native speakersand expandingcircle users from the data.
This thesis follows the `post-modern' view that language users are not so easily
divided into groups (Canagarajah 2005) and that, in any case, World English
includes all usersof English, irrespectiveof how they acquired their languageand of
whether or not they constitute the majority. By using spoken English data supplied
by a wide range of users from different countries, the thesis, following Firth, House,
Meierkord and Lesznyäk, attempts to characterise the way in which participants in
international conversationsachievetheir aims at discourselevel.
In order to verify whether this characterisationhas truly to do with international
settings, a contrast is naturally necessary. Where Lesznyak chose to contrast a
conversation among non-native speakers of English with one where non-native
speakerswere interacting with native speakers,this thesis, given the abandonmentof
traditional divisions, will instead contrastconversationsin international settingswith
conversationswhere all the participants share a nationality. In most instances,these
homogeneousgroups share much more than that, also having in common another
language and culture. It is hoped, therefore, that a picture of the conversational
discourseof World English may emergewhich is both inclusive and a true reflection
of the current multilingual world of global flows (Appadurai 1996).
Having establishedthat the approachtaken to languagein this thesis is a discoursal
one, it is important to emphasisethat, within that framework, several different
analytical methods have been drawn on. In order to find differences between
conversationsin international settings and those among homogeneousspeakers,an
eclectic range of analytical tools have been used from pragmatics, conversation
analysis, discourse analysis, intercultural communication and communication
accommodationtheory.
3 Overview
The first part of this thesis concerns itself with delineating the research tradition
within which it is located.
Chapter Two is concernedwith the tradition of identifying varieties of English and
focuses on the attempted establishment of a global variety. It begins by an
examination of `StandardEnglish', long held up as the international norm to which
all global users would, or even should accommodateand goes on to examine the
concept of World Standard English, a would-be replacement for Standard English
without the national, or even nationalistic overtones of the former and then of
International English, World English and Global English. The aim of this is to
investigate the possibility of fitting the data gatheredfor this thesis, and its analysis,
into a ready-madevariety of English.
Chapter Three turns to descriptive models where Ènglish' is accompaniedby às'
and then either: àn international language' (EIL) or à lingua franca (ELF). By
Chapter1 10 Introduction
focusing on users,rather than on an idealised form of à' language,a firmer basis for
this type of researchis suggested.The Chapter is neverthelesscritical of both terms.
In the first instance, the ìnternational' label seems, still, to insist on often false
national boundariesand is predicatedon a modernist view of the world divided into
neat nationalities, eachwith its own language,

using English for communication with
other nationalities. The term Ènglish as Lingua Franca' has been used almost
exclusively to refer to èxpanding circle' users (see above) and is therefore
potentially confusing in the context of the more genuinely global aims which this
thesis setsout to achieve.
Having establisheda researchtradition within which to operate,the thesis then turns
to an examination of what might be expectedin conversationsconducted in English
across national and cultural boundaries. Chapter Four therefore deals with
perceptions speakers may have of each other, including the perception that some
users are `native speakers' while others are not, the possibility or otherwise that
speakers will accommodate to each other, the directions accommodation might take
and the likelihood of users attempting to follow particular rules. The chapter
attempts, therefore, to justify the explanatory approach to language analysis which
informs Chapters 8,9 and 10.
Following this speculatory chapter, Chapters Five and Six are concerned with
approachesto the concrete language data. Chapter Five provides an overview of
different possible ways of obtaining data-providers and of screening them for
language-basedattitudes, particularly those to do with the status of native and non-
native speakersbut also language-basedattitudes towards each other. The chapter
continues by examining ways of obtaining conversationaldata and different ways in
which the data may be analysed,bearing in mind that the aim is to take a discoursal
Chapter 1 11 Introduction
view. Chapter Six then focuses on the actual procedures used to identify data-
providing people, to screen them, to elicit conversational data from them and to
analysethe data in order to fulfil the researchaims.
Chapter 7 presentsthe results of the screeningproceduresand reports findings from
researcher-leddiscussion groups and questionnairesbased on participants listening
to each other. These results are used to help explain conversational behaviour in
subsequentchapters.
Chapters 8,9 and 10 present the data gathered and the results obtained therefrom.
Chapter 8 takes one international conversation and compares it with one
intranational conversation. This one-to-one approach allows for a turn-by-turn
analysis of each conversationand servesas an overview of the method adopted for
the data as a whole. Provisional conclusions concerning discourse strategies and
conversational style are drawn from the contrast between the two conversations
under scrutiny.
Chapter 9 takes the findings from Chapter 8 and applies them, in less detail, to the
remaining conversations.Thus discourse strategy and conversational style provide
the framework for considering how all the conversations work and how the
international conversations differ from the intranational ones. Explanations for
differences are offered in conclusion.
Chapter 10 consists of six case studies where six of the seventy plus data-providers
are examined in detail with a view to affirming the generality of the conclusions
drawn in Chapter 9 and to providing further in-depth analysis of the differences
between international and intranational conversations.
Finally, Chapter 11 summarisesthe thesis and looks briefly at upshots: some
considerationis given to possiblepedagogicapplications of the researchand
suggestionsfor possible future research
Chapter Two
Is English as a World Language a
variety of English?
Chapter2 14 Is EWL a varietyof English

1. Introduction
Chapter One considered,albeit briefly, the possibility that World English might be a
variety in its own right, to be seen alongside other varieties of English or other
Englishes.
This Chapter will attempt to capturethe current world-wide phenomenonof English
by dealing with it in these terms. If such endeavoursare successful,the comforting
prospect is raised of a describableobject of particular interest to teachersused to a
transmissionmode of teaching,who would then have something concreteto transmit.
Severalproposalshave beenmade along theselines and a range of adjective+English
formulae have been suggestedas labels to classify the way people speak and write
English acrosscultural and national boundaries.They are: StandardEnglish, World
Standard English, International English, World English and Global English. Each
one will be dealt with in order to consider its validity as a framework within which
to present the data gathered for this research. In other words, this Chapter will
attempt to answer the question posed in its title.
The naming of Englishes in this way, and the assumption that the name exists to
denote a reified entity, seems to belong to a long tradition of English Studies,
referred to by Bolton as Ànglistik', which has incorporated synchronic language
study into the diachronic tradition. (Bolton 2003:42-43). McArthur's variation on the
chronological model for language description puts `World English' at the end of a
narrative time line starting with `Pre-Old English' and passing through `Middle
English' and`ModernEnglish' (McArthur 1998,Chapter4).
Similarly, World English may be seen as an entry on what McArthur calls a
`biological' model or a `geopolitical' model of English: in the former, English
English is represented as the root from which other varieties have stemmed while in
Chapter2 15 Is EWL a varietyof English?

the latter adjective+English formulae are mapped onto national or regional
boundaries (McArthur 1998 Chapter 4). A good example of a geopolitical model is
to be found in Görlach (Görlach 1988 in Crystal 1997) where a large number of
Englishes, or varieties of English, are identified as segments of a circle (See Fig 1).
/up, p-eprn;
scuts m w. rt.
er/t . 41.1«
aut . wrie r
et[ etuw r. ney
a.
G.J. r. t; E
X T KH. 1 a. 1191,10 e.
. . Ir1 an E
Yak plain Scptti sn E wtlw
sp1 IiJin Enp11a1 F Y. isn F 1.1., E
. '.,! a- Egt
Ital. r perm
wrtnam 11E/IEV
apor; g f SW0rn
Auaf Ir[ C"nf ýw1i"n
pitutrn ' pl0pin'
eZE
us[
papu. Mte-
I1Gf o..., n
ntlrmtlpml JrfJC
Egttfn
J. r[ Gd/C
E IntE GqE I, J. n
WIF C"ý
e. nf/c
I. rOaa
se(
s ui. n
[-n uriun
Im F-n
r. a[
11"11 f
["N. n f
eunar [
/ YAR [AR f4/C SI, Mn
LIOE [rny. n f IYL. i [

fý, n [ i[ne C Ifrlkum
1119E
[Ho, Gwrroon rC
.a
Fig 1. Görlach's geopolitical model of Englishes
In order to be admitted into the fold of adjective+Englishes, potential entrants may
have to satisfy one or more of a set of criteria which are worth mentioning in order to
provide a critical context in which to judge the validity or otherwise of the
unificatory varieties of English listed above (International English, World English,
etc.).
A possible first criterion is the criterion of stability. English may not be preceded by
an adjective if the resulting reference is to an unstable entity. The authors of the
authoritative `Grammar of Contemporary English' imply as much when they state
that what they call ìnterference' varieties of English have stabilised themselves and
Chapter 2 16 Is EWL a variety of English'?

becomevarieties in their own right (Quirk et. al. 1972) and Davies defines standard
varieties of English as oneswhich `do not have instability' (Davies 1989:461).
Further criteria come in a paper by Llamzon, who proposes four features which
allow the identification of a `New', adjective+English label to be addedto the list. (1)
Ècological features' have to do with the environment in which the variety develops
and comprise phenomenasuch as code-switching and lexical borrowings from local
languages. (2) `Historical features' means, simply, that `New Englishes' have a
relatively short history of codification while (3) `Sociolinguistic features' indicates
the domains of use which include intimate ones. Finally, (4) `Cultural features'
indicates the presenceof creative writing products or a local literature (Llamzon
1983:100-104).
Platt, Weber and Ho parallel Llamzon, at least in part, by suggesting that `New
Englishes' must be ones which are used for a range of functions among those who
use them and which adopt features of their own. They add, however, that a `New'
English must also be developed through the local education system. (Platt, Weber
and Ho 1984: 2-3). The inclusion of an institutional criterion seems to be implied in
McArthur's list of `de facto' situations where English has taken root and becomes
explicit where the situation is `de jure' (McArthur 1998, Chapter 2).
Apart from suggesting criteria, Platt, Weber and Ho also identify different
approachesto defining Englishes: these are political, regional, ethnic, functional and
language description (Platt, Weber and Ho 1984:5-6). Görlach's segmentedcircle
(see Fig. 1) seemsto include Englishes defined according to the first three of these
such
approaches, as United StatesEnglish, New England English and Native Indian
English and Andrew Dalby's list of ìnner-languages', a term he takes from David
Dalby (Dalby, D. 1998), `within the entity usually defined as English' reflects a

similar focus and therefore includes Canadian English, Antipodean English and
Talkin-Black (Dalby, A. 2002: 25).
More recently than Llamzon and Platt et.al., Butler's list of criteria has endorsedthe
ecological and cultural features criteria by suggestingthat a World English can be
identified by `particular words and phraseswhich spring up usually to expresskey
featuresof the physicalandsocialenvironmentandwhich areregardedaspeculiarto
the variety' and by à literature written without apology in that variety of English'.
Butler also seems to concur with the points concerning history and
institutionalisation, asserting that a World English has its own reference works,
evidence of its own norms. She gives more weight to the notion of historical
community than her predecessorsand addsthe criterion of pronunciation which must
have a `standard and recognisable pattern (which is) handed down from one
to
generation another' (Butler 1997: 106).
The criteria of stability and institutionalisation have not necessarilybeen universally
accepted. Kachru, glossing his renowned three-circle model of English, identifies
outer circle varieties as institutionalised, but calls expanding circle Englishes
`performance' varieties, with a highly restricted functional range (Kachru 1992c).
More recently, Meierkord has distinguishedbetweenEnglisheswhich have stabilised
through development within education systemsand those which have not stabilised
at all (Meierkord 2004). The important point here is that the label `variety' seems
still to be applicable, as, indeed, is the adjective+English formula.
At the same time, the political, regional and ethnic approachesto variety definition
have been added to by what may be termed a `corporate approach'. There are, then,
attested claims to support the existence of, for example, German English (Smith
1984: 57, Viereck 1996: 23), Korean English (Meierkord 2004), China English as

well as Chinese English (Qiong, 2004), Euro-English (Warschauer2000, Modiano
2001) and also Nokia English, the logical conclusion of that company's policy to use
English only as its languageof business(McArthur 2003). The notion of a political,
regional or ethnic speech community may therefore been extended to that of a
company community.
The adjective+English given above,usedto describethe currentglobal

expressions
phenomenonwill now be dealt with in turn. A couple of generalpoints are, however,
worth making first.
The criterion of stability is unlikely to apply to any of the formulae except `Standard
English'. With current technologically-aided descriptive approaches to language
study, as opposed to older prescriptive methods, spoken language of any kind is
unlikely to be very stable. Crystal, for example, describes `the spoken British
English of Britain' as à mass of hybrid forms' (Crystal 2001: 61). Stability is even
less likely if spoken languageis being used by a much vaster number of people from
a great diversity of backgrounds.
Similarly, the institutionalisation criterion is bound to fail for all the Englishes listed
except StandardEnglish, since there is no single world-wide structure which acts to
make a particular variety of English the acceptableone, either dejure or de facto.
Finally, regional, ethnic and corporate approachesto definition are, self-evidently,
doomed to failure, unless the entire world is consideredto be a region, and, staunch
internationalists and journalistic cant apart, `the international community' is non-
existent, so any political approach to defining Ìnternational English' would be
aspirational,not factual.

Spoken.English as a World Language: international and intranational settings
2. Standard English
Until fairly recently, the phenomenonof English being used widely throughout the
world was accountedfor by referring to StandardEnglish: historically there has been
no perceived difference between the English used acrosscommunity boundariesand
'Standard English' or, more precisely, 'Standard British English' or 'Standard
American English'. It is a history which has been perpetuateduntil very recently:
Honey, for example, assertsthat 'StandardEnglish, in either its British or American
variety, has become recognised as the language of international communication'
(Honey 1991:28) and Quirk confirms that 'Germansand Japaneselearn standardBrE
or AmE' (Quirk 1991a).
In a possible rebuttal to those objecting to the diffusion of Standard English, Preisler
argues that'in its two main varieties' (meaning British and American), it should form
the basis for the functioning of English as international lingua franca 'for
communicative reasons' (Preisler 1995:342), while Lowenberg implies that Standard
English is the world-wide norm when he states that 'Standard English differs only
minimally across varieties, generally sharing a large set of common norms', being
used 'with mutual comprehensibility in international communication' (Lowenberg
2000: 211). The implicit dominance of Standard English is present even in very
recent writing. Qiong, for example, in defending China English, states that this is às
good a communicative tool as standard English' (Qiong 2004).
There are three main problems in considering Standard English to be the language
used for general communication across the globe. The first has to do with defining
what Standard English is, while the second centres on the difficulty of using a
monolithic model of English, an idealisation rather than a reality. The third problem

Spoken English as a World Language: international and Intranational settings
may be thought of as an ideological one, where the British or American associations
have to be called into question.
2.1. Defining Standard English
According to Davies, the term `Standard English' seems to refer to grammar: he
concludesthat is
standardisation a processwhereby an individual's grammar (which
Davies calls Grammar 1) is converted to a shared grammar (Grammar 2) (Davies
1991a:125). Trudgill appearsto agree,arguing that dialects show themselvesthrough
grammar, therefore implying that StandardEnglish does the same.He also maintains
that there is such a thing as StandardLexis, which is often sharedwith non-standard
varieties (while non-standard lexis does not occur in Standard English) (Trudgill
1999:117-8).
Crystal gives grammar,vocabulary and orthography as featuresof StandardEnglish,
and leaves out pronunciation (Crystal 1994). Widdowson agrees on the
orthographical question and that pronunciation is not a factor (Widdowson 1994), a
point of view with which Strevensconcurs (Strevens 1983, cited in MacKay 2002:
51-2). Smith uses field researchresults to reach the 'logical' conclusion that Standard
English can be spoken in any accent (Smith 1992), and Honey agrees,with the rider
that Standard English ìs seldom (indeed perhaps never) spoken in the broadest
forms of regional accent' (Honey 1997: 1). His referenceto broad forms of regional
accentsdoes point, however, to a 'narrow' form of 'non-regional' accent which may
perhapsbe taken to be the standardform.
For others writing on Standard English, phonological forms are most definitely
included. Partridge, for example, refers to Cooper's Grammar as having established
that `the criterion of correct Standard English rests firmly on its pronunciation'

(Partridge 1969) and Giles and Ryan identify people using 'non-standard'
pronunciation (Giles and Ryan 1982). Modiano is categorical on the matter: having
defined Standard English as a spoken standard including only forms of language
which are comprehensibleto competent speakersof the language worldwide, he
concludesthat accent is important and that StandardEnglish cannot be spoken with
anything other than a standardaccent(Modiano 1999b).
Mugglestone's work entitled `Talking Proper' would appearto be, at least currently,
the last word on the subject. She shows clearly how pronunciation went through a
standardisation process within England and how that process continues; she also
raises the question of the of

standardisation pronunciation as an ideological goal and
points to the long tradition of `polite' speechand its renaming as èducated' speech
being held up as standardsfor all (Mugglestone2003).
The existence of codified phonological systems such as 'RP' and 'GA', for long
enshrinedin dictionaries, as well as the fact that there is a long tradition of teaching
materials aimed at helping learners of English to acquire a `correct' pronunciation
also suggeststhat there is, de facto, a StandardEnglish phonology. In a well-known
English Languagetext book, the authors refer to pronunciation in terms of Standard
British English (Swan and Walter 1993:40).
Orthography is a far less contentious contender for inclusion into StandardEnglish.
Derwing, Rossiter and Ehrensbreger-Dow,for example, include punctuation in their
list of Standard English features adhered to by non-native users of English but
by
abandoned native speakers(Derwing, Rossiter Ehrensbreger-Dow2002).
The inclusion of orthography in Standard English is part of the perception that
Standard English is a written language rather than a spoken one: the 'channel'
argument, that Standard English is a reference to a written code, is a strong one in

the literature. Quirk opines that Standard English is an ùnmarked variety, associated
with written English' (Quirk 1990, cited in McKay 2002: 51-2) while Carter
suggests that `there is a close relationship between Standard English and the written
language'; he notes that `native speakers consider language ungrammatical when it
does not occur in formal written styles' (Carter 1999: 12-13) and Cheshire makes the
point that `the spoken language has not received enough attention to be able to talk
about a standard' (Cheshire 1999).
Honey seemsto disagree,maintaining that while non-standardEnglish is a spoken
language,StandardEnglish is both written and spoken(Honey 1991).
If there is some confusion regarding the identification of Standard English with the
written channel, there is considerably more when it comes to equating it with a
particular style. Preisler opines that Standard English is `the overtly prestigious
variety that speakers of non-standard English will accommodate to according to the
formality of the situation. ' (Preisler 1999:262), while Trudgill maintains the opposite,
stating that non-standard English has nothing to do with its being colloquial or
informal (Trudgill 1995, quoted in Davis 1999:70).
The problem here seems to be that, despite Preisler's matching of variety and
formality, style is really often a question of languageusers' perceptions, rather than
items inherent in grammar, lexis or phonology.
This leads to the question of whether or not pragmatics can enter the Standard
English equation.
For Nero, the narrow area of 'essayist rhetorical convention' is part of Standard
English, although beyond this particular element, Standard written English is, he
thinks, a myth (Nero 2002: 55). Meierkord implies that there are discourse features
inherent in Standard English by noticing, among participants in Lingua Franca

exchanges,that signals such as backchanneling,topic markers and phraseboundaries
are different from those normally employed in 'Standard L1 varieties' (Meierkord
2002:120).
It has already beennoted that Preisler proposesthat the English to be used acrossthe
world should be basedon British or American StandardEnglish. He appearsto want
to extendthis basisto pragmaticfactors,thus perhaps,perceivingpragmaticnorms
as part of StandardEnglish: he proposesthat 'shared cultural assumptionsshould be
reflected in the teaching of pragmatics and they should derive from the cultural
contexts of StandardEnglish' (Preisler 1999:266).
In the main, however, all the voices in the literature concur that StandardEnglish
does not include pragmatic norms. Kramsch is very clear on the matter, stating that
'while there is such a thing as standardisedor conventionalised English usage (a
linguistic concept), there can be no such thing as standardisedlanguageuse (a social
concept) (Kramsch 1998) and Davies notes that communication among users of
Standard English goes awry becauseof a lack of similar cultural assumptions;he
gives an example of politenessnorms which are violated by some users of Standard
English and concludesthat the 'reasonfor triggering (of norm-violation) is external
to the languageitself (Davies 1991a:143).
As can be seen, then, there is no strong consensusas to the meaning of Standard
English and in particular on the issues of whether or not there is a standard
pronunciation, whether or not there is a standardspokenlanguageand whether or not
pragmatic norms form part of the standardlanguage.

2.2. Descriptive problems
It would seemtherefore to be fruitless to attempt to account for the world-wide use
of English by referring to an ill-defined object; it seemsequally fruitless,
or at least inaccurateto assertthat StandardEnglish in one of its two major forms is,
in fact, either the basis for oral communication on a global scale or even the norm of
correctnessto which those involved in spoken cross-cultural communication refer.
When Quirk affirms that Germansand Japaneselearn StandardEnglish (see above),
it may be more correct to say that this is what they are taught. Whether or not they
learn it is anothermatter. Honey's assertion,in the samevein, that StandardEnglish
is recognisedas the languageof international communication may be true only at a
similarly theoretical, even hopeful level.
In the first place, it is quite clear that the majority of British and American people do
not speak in Standard English and do not refer to it as a norm of correctness. If, as
has been suggested by many, Standard English is a written code (see 2.1. above) it is
a logical non sequitur that it is used in speech: spontaneous speech only very rarely
replicates the written word (Milroy 1999:21, Trudgill 1999: 118). When it comes to
using Standard English as a criterion of correctness, even for spoken conversation,
the situation is a little more complex. Because of (largely old-fashioned) schooling
conventions, it is customary at least for some 'native speakers' to refer to learned
rules of Standard English in order to adjust their own speech or to criticise the
speech of others (Milroy 1999:25). In a piece of research conducted by Hultfors,
'native speaker' participants 'corrected' the speech of 'non-native' users by simply
making utterances more formal, underlining, perhaps, the idea that Standard English
is the ultimate, though unattained goal of conversational English (Hultfors 1987).

This perceivedgoal explains the shame,detectedby Joos and felt by Americans who
were aware that they did not use Standard English in their conversations (loos
1967:4).
Other 'native speakers'seem,however, to make a point of not referring to Standard
English in their oral production. In data gatheredfor this thesis (reported in detail in
Chapter7), severalspeakersproudlydefendtheir own variety of spokenEnglishand
ridicule the idea that they should even try to adjust their speech to conform to
StandardEnglish norms. This example comes from a twenty-two year-old, adamant
that his languagenorms are as good as those stereotypically associatedwith Oxford:
John: The way of speaking it. I am a Hemelite so this is like a Hemel, a
Hemel tone, Hemel phrases. I do 'ain't' and 'innit, you know what I mean?
But if someonefrom Oxford would comedown and say well that's not English
but to me that's English. But that Oxford person has come in and learned the
words inside out for him to tell me it's not English.. you can't really put a
specificationon it, because

it is..
(Appendix K. D3: 105-110)
As Crystal points out: `the spoken British English of Britain is already a mass of
hybrid forms (Crystal 2001: 60-61).
English used acrosscultural boundariesseemsto follow the samepatterns as English
among British or American speakers. Many, perhaps most, 'non-native' users of
English turn away from the norms of StandardEnglish as a way of regulating their
conversations. In a 1981 survey, conducted by W. D. Shaw, between a third and a
half of Singaporeans and Indians thought they should learn their own variety of

English (and not a British or American one) (Shaw 1981). The idea that they needed
schooling in Singaporeanor Indian English seemsto show that they perceived these
as the varieties supplying norms of correctness.
Nearly twenty years later, Kramsch notes that the native-speaker norm is being
erodedbecauseof (among other things) the variety of different Englishes acrossthe
world and the inadequacy

of NS Englishto expresssocialnormsof other societies
(Kramsch 1998), a point also made by Kachru, who arguesstrongly against the idea
that those failing to adhereto 'native speaker'norms are learnersof English who will
never acquire full competence(Kachru 1994b,quoted by Silva 2000:162).
Warschauerconcludesthat users of 'New Englishes' in the 'Outer Circle' emphasise
their own version of English' while even those in the Expanding Circle seem to be
rejecting native speaker like models' (Warschauer2000:515) and Lowenberg also
concludesthat the norms acceptedby highly educatednative speakersare no longer
valid and that uses in countries where English has achieved official status have led
to modifications which are de facto norms (Lowenberg 2000).
There is clearly a great deal of variability in languageuse, making the concept of
StandardEnglish difficult to sustain when it comes to oral interaction of any kind
(Ahulu 1997:19). This seems particularly true of cross-cultural oral interaction.
Knapp observes that "It is not self-evident that interactants in L(ingua) F(ranca)
situations apply the norms of a particular English speaking community' and Beneke
notes that "E(nglish as a) L(ingua) F(ranca) is characterisedby manifestations of
uncertainties concerning the applicable norms at all levels of the language system
including socio-pragmatic norms" (Knapp 1987, Beneke 1991, cited by Lesznyäk,
2002: 163-4).

Finally, it has been suggestedthat any users of Standard English tend to diverge
from their standards in order to make themselves understood internationally
(Warschauer2000:515).
If Standard English does not account for the way English is used, or is not the
criterion against which users attempt to adjust or correct their speech,perhaps the
answer is to redefine StandardEnglish or even to construct a new 'spoken' Standard
English.
Carter seemsto think this is possible, claiming that '..... there are forms of spoken
English which are perfectly standard and which are indeed grammatically correct
(Carter 1999: 158), while Milroy rejects the idea out of hand (Milroy 1999:24). It is
neverthelesstrue that patternsnot normally codified in SE grammar books, do occur
regularly (not to say standardly) in speech.Cheshire,for example, identifies items in
what she calls 'Standard Spoken Grammar', items such as left dislocation (Cheshire
1999).
But where patternshave been identified, they have remained on the level of patterns,
and seemto be much more culturally specific than the standardsfor written language.
Further, patterns among 'native speakers'have received considerably more attention
than those occurring among 'non-native speakers', making their existence less
relevant to global communication where the latter far outnumber the former: locating
models of correctness within 'native-speaking countries' is incompatible with the
'global future' of the language(Graddol 1997:10).
In other words, where they exist, thesepatternscannot account for the way English is
used for oral communicationacrossa range of different cultures by people in
multilingual contexts.

2.3. Ideological problems
The point has already been made that `StandardEnglish' is generally consideredto
mean StandardBritish or StandardAmerican English. Alongside this convention, the
term 'native speaker'usually refers to citizens of Britain or the United Statesand to
those migrating from these countries, or having descendedfrom these migrants.
Consequently,'StandardEnglish' is very frequently taken to mean the languageused
by 'native speakers',irrespective of the fact that both terms are idealisations, rather
than empirically identifiable entities (Seeabove,Milroy 1999:18 ).
Davies explicitly uses one to define the other, maintaining that 'you are a native
speakerif you speakthe standardlanguage'because'the process of standardising is
an operational definition of the native speaker' (Davies' italics, Davies 1991a: 63),
and Braine places the adjectives 'standardised'

and 'native' together before the phrase
'norms of the English language'while describing a tension between those norms and
the ability of the 'non-native' language user to test their limits (Braine 1999).
While they have produced 'native speakers',some of whom have travelled the world,
directly or indirectly influencing the learning of English, British and American
institutions have also harnessedthe standardisingtendenciesof English within their
state boundaries to produce the grammar books and dictionaries to which most
learners and users of English have traditionally referred. More recently, some
compilers of linguistic corpora, influential in the world of English Language
Teaching, have relied on 'native' speakersof 'Standard' English to provide all the
examplesin their work (for example Chafe et. al., 1991:69).
Such a state of affairs is bound to attract opposition of a political or ideological
nature, especially in a world where the majority of users of English are not 'native

speakers' and do not think of themselves (or are not thought of) as using 'Standard
English'.
One prong in the attack, developedby Phillipson, suggeststhat StandardEnglish has
been imposed on the world by the centralising and imperialist forces of Britain and
the United States.Phillipson traces standardisationin its institutional form back to
the Spanishmonarchy at the close of the fifteenth century and shows how a national
language such as Standard English comes to hold sway over local languagesby
political force. (Phillipson 1992: Chapter 2). The national political theory is also
developed by Blank, who underlines the connectednessof standard language and
nationhood (Blank 1996: Chapter 1). Phillipson proceeds to explain how Standard
English has come to be imposed on large areasof the world as the languageof the
61ite,relegating other languagesand other varieties of English to categoriessuch as
'vernacular', 'lingua franca' or 'international language' (Phillipson 1992: Chapter 3).
Brutt-Griffler has, however, attemptedto refute Phillipson's main claim, maintaining
that English has becomethe major world languagenot becauseit has been imposed,
but rather as the fruit of resistanceto imperialism (Brutt-Griffler 2002).
A researcherin the twenty-first century cannot analyse spoken English data against
an imperialist model of Standard English, nor can he or she equate English as the
language of resistanceand liberation with British or American standards.To do so
would be to consider a great deal of English as 'defective', rather than as the
naturally-evolved expressionof a speechcommunity. Thus Ahulu, for example,calls
for a redefinition of StandardEnglish which might accommodatedevelopmentsoften
described as 'New Englishes' and which would acknowledge 'liberation linguistics'
(Ahulu 1997).

Ahulu's is not the only 'post-colonial' voice to rail against the predominance of
standardised,native norms. Kachru argues that the British and the Americans
seemedto 'have lost the exclusive prerogative to control (the) standardisation(of
English)' (Kachru 1985, quoted in Pennycook 1994:10), while Singh et. al. point up
the flawed reasoningunderlying the adoption of British English as a yardstick (Singh
et. al. 1998) and Mufwene underlinesan almost moral case for abandoningthe
norms of 'some little islander with no experience of the world' when trying to
describean international communicationinstrument (Mufwene, 1998).
To continue to refer to Standard English when attempting to describe how
communication takes place across community boundaries in the world is therefore
politically incorrect to the point of insensitivity. It is also likely to be factually
incorrect, as Singh et al suggest(seeabove).
2.4. Conclusion
StandardEnglish, as was mentioned in the introduction to this Chapter, satisfies the
institutionalisation criterion for acceptance as a World English. It also clearly
satisfies the creativity features criteria. Where Standard English is a less likely
candidate for a place among World Englishes is in the ecological features area:
lexical borrowings are discounted in Standard English (Trudgill 1999: cit) or are
only counted once they have been assimilated and codified. Following Butler's
criteria, it is debatablewhether or not StandardEnglish has its own pronunciation.
Even without reference to Llamzon, Platt, Weber and Ho or Butler, it would seem
clear that accounting for cross-cultural, international or global oral communication
by simply referring to Standard English is both inaccurate and politically
unsustainable: for a term in widespread use in much serious literature, there is little

agreementon what it meansor encompasses,

it servesas neither a real pattern nor an
ideal model for the vast majority of oral interactions in English, whether among
'native speakers' or across community boundaries and its use fails to take into
accountthe justified political sensibilities of those who feel left out of its scope.
3. World Standard English
WhereasStandardEnglish exists, despiteresisting careful definition, World Standard
English is an idea which has no concrete manifestation. No language data exists
which has been dealt with under the `World StandardEnglish' heading.
For Quirk, a 'standard of standards'should emerge in the same way that national
standardshave arisen as the result of languageusers accommodatingto each other
(Quirk 1985: 4-5, cited by Davies, 1989: 458) and Crystal envisagesa future World
StandardSpokenEnglish, available to a very large number of people who may use a
dialect of English, or another language completely, in their daily life. World
Standard Spoken English would be used by everyone when in contact with people
from different speechcommunities (Crystal 1997: 13).
The problem with this vision is that it supposesthat a standard variety emerges
simply as the result of speakers communicating with each other and somehow
agreeing on shared norms after a due period of accommodation. This process
undoubtedly takes place ad hoc when speakersfrom different speechcommunities
engagein interactions (Hope 2000). It is doubtful, however, whether a new standard
be
can constructedwithout somekind of 'top-down' intervention.
National standard languages have been codified, propagated and taken up as an
index of nationhood, as one way among many for people to express their national
cultural identity (Fishman 1969 and 1972). As Crystal seems to imply, there is no

question of World Standard English being used for this purpose. Without the
unlikely institution of an international commission to oversee and codify itl, and
without the equally unlikely emergenceof a 'world citizen' identity, there is little
possibility of World StandardEnglish coming out of its primitive state.
World StandardEnglish fails all of the criteria for a `New English' and, given its
does
generalunlikeliness, not providea plausibleframeworkwithin which to present
the data in this researchthesis. In short, spoken English used internationally and
cross-culturally cannotbe presentedby referenceto World StandardEnglish.
4. International English
While StandardEnglish defies definition and seemspolitically incorrect and World
Standard English seemsto be a chimera, the term Ìnternational English' presents
different difficulties: it has been used to refer to three distinct phenomena:a world-
wide standardacrolect, a more demotic basilect and a possible artificial language.
4.1. International English - as a standard language
Whereas the coiners and users of the term World Standard(Spoken) English seem
fairly sure that it does not as yet refer to anything concrete, some users of the term
International English are confident of the reality of the language. The Nuffield
Enquiry Team, called upon, in the year 2000, to report on the state of language
teaching and learning in schools in the United Kingdom, turned its attention to the
way English is used and taught. Perhapsin an attempt to shock British schools into
1 The Society for Pure English was set up in 1913 in responseto the rapid spreadof English over the
world in order to assistthe developmentof the language, guided by acknowledgedprinciples of
tradition and taste' and with an emphasison inheritance,beauty, integrity and adequatenessto express
modern ideas. (SPE 1919).This, and similar organisationsdo not appearto have had any noticeable
effect on English used over the world.

teaching pupils a form of English close to StandardEnglish, the Team declaredthat
'UK English continuesto become effectively a dialect of International English' and
that 'native-speakers'of English will one day lose the capacity to use English
internationally (Nuffield LanguagesEnquiry 2000:15). The word 'dialect' may have
been chosen for its shock value, yet the Team seemsto be sure of the existenceof
International English, presumably an acrolect above localised 'UK English'. The
Enquiry report doesnot attempt to clarify what International English may be, or how
it differs from other standardEnglishes.
In an early paper, Davies seems to make explicit the underlying assumption of the
Nuffield team's statement:that International English may be any one of the standard
varieties (Davies 461).

1989: Two years later he is not so sure: in his 1991 book, he
hesitatesbetween having International English as the automatic corollary of being a
user of an èxisting Ll standardEnglish' or as the result of being a secondlanguage
user (Davies 1991a: 68).
Medgyes takes the same view of International English as Crystal (see section 3.
above) does of 'World Standard',

that it does not yet exist but that it undoubtedly will
do so in the future (Medgyes 1999:185).
4.2. International English -a Lingua Franca
The view that International English may be a standard,or at least that there may one
day be a standardInternational English, is a fairly recent view compared to the one
that it is of relatively lower, basilectal status.
In 1978, Lester drew on language contact theory to explain the phenomenon of
International English which, he maintained, could be separatedinto varieties such as
'Greek English' and 'Japanese English': these are creolised forms, or contact

languages.International English itself 'is a contact language made up of contact
languages'(Lester 1978:13, quotedin Brutt-Griffler 2002:13).
There appearsto be a certain lack of logic in the two statements:it is not clear how
contact languages can be varieties of, and at the same time constituents of
International English. At all events,the point seemsto be that International English
is a hotch-potch of pidgins and creoles. Lester seems to be describing a different
phenomenon from the International English claimed by the Nuffield Languages
Enquiry twenty odd yearslater (seeabove,4.1).
Samarin takes a similar view to Lester's, but differs in that he sees International
English as originating not in creolised or contact languages,but in the 'standard
languagesof politically and economically dominant nations'. He is neverthelesssure
that International English is a Lingua Franca,'a functional tool, lacking the elegance
and sophisticationof a standard (Samarin

language' 1987:
372).
Much more recently, Graddol seems to imply that International English, far from
being a standard variety, is a reduced or simplified English and goes on to wonder
whether International English has not èmerged as a natural consequence of the
increasing use of English as a medium of communication between speakers of
different languages?' (Graddol 1996).
However simplified or lowly compared to standard languages, lingua francas are
nevertheless 'full languages' according to Würm (Wärm 1971:1017), or at least
language varieties according to Hall (Hall 1966). Lingua Franca English is, for
Seidlhofer, àn additionally acquired system that serves as a means of
communication between speakersof different first languages, or as a language by
means of which the members of different speech communities can communicate
with eachother but which is not the native languageof either -a languagewhich has
Chapter2 35 Is EWIGa varietyof English?

no native speakers' (Seidlhofer 2001:146). It ought therefore to be possible to study
International English (or `Lingua Franca English'), and to describe it in accurate
detail, in the sameway that other varieties of spoken English have been described.
Indeed, Seidlhofer is engagedin just such a project: the Vienna Oxford International
Corpus of English (Seidlhofer 2005).
If International English in this senseis to be considered a variety in its own right,
then it must be a performancevariety and not an institutionalised one since it cannot
fulfil the stability criterion. Its failure to do so leads Gnutzmann and House, for
example, to conclude that it is not a variety at all (Gnutzmann 1998:133, House
2002a).
The second problem is that International English - Lingua Franca has no native
Lesznyäk
speakers. is quite clear on the point: Lingua Franca is 'a contact language
between speakerswho do not sharea mother tongue'...... 'Furthermorethis language
is per definition mother tongue to none of the participants' (Lesznyäk 2002:166).
The focus of the Vienna Oxford corpus is on communication among speakers'from a
wide range of first languagebackgroundswhose primary and secondary education
(and socialization) did not take place in English' (Seidlhofer 2000).
It is precisely this limitation which this thesis has set out to overcome (see Chapter
One).
4.3 A specially constructed International English
While the Nuffield team make claims for an International English as a standard
language and Seidlhofer takes the contrary view, that it is a Lingua Franca, others
have expressed the idea that International English might somehow be invented. The
desire for an internationally serviceable, and therefore simplified English, led to no

fewer than sevenproposals,made between 1889 and 1990, for a 'reduced English'
(Graddol 1996).
The most recentproposal,expressedmore as a fantasy than as a concretesuggestion,
was made by Johnson,who imagines a language that would reflect what he called
'International English culture', consisting of 'what is experienced throughout the
modem,technologically
advancedworld and taughtin 'modem'curricula'(Johnson
1990: 305).
This artificial variety of English would contain only a few items specific to the
cultures traditionally associatedwith English, would regularise 's' for all plurals and
'-ed' for all past tensesand would use 'un' as the only negative prefix. It would be
functional in style, 'inelegant,impersonal,soulless,boring and unstimulating beyond
its literal content' (ibid, 309-310). The resulting International English would at least
be easierto learn.
The pedagogic value of a specially-constructed language is emphasised by
Seidlhofer in her analysis and application of Ogden's Basic English (Seidlhofer
2002). Given that English is largely a taught system for a majority of users,
Seidlhofer proposes matching the prescriptivism of taught systems to her English as
Lingua Franca data. This might be a fruitful area of analysis, but for the limitations
implicit in studying Lingua Franca English (see above).
4.4. Conclusion
It seems,then, that there is someconfusion as to what International English is
but, of the three approaches,the one reported second here, equating International
English with Lingua FrancaEnglish, is clearly the most robust.

Taking this approach,International English does not, of course, satisfy the stability
and institutionalisation criteria, nor does it show any cultural features.There may be
a case for considering it to be a performance variety, and therefore akin to
Expanding Circle Englishes, in which case, only the sociolinguistic or functional
criteria seem to apply. Furthermore, using Platt, Weber and Ho's approachesto
naming Englishes, Ìnternational English' doesnot seemto fit easily into any of their
suggestedcategories,unless staunchinternationalists consider it important to use the
term as a political marker.
At all events, large and growing numbers of people seem to be excluded from
International English approachedas Lingua FrancaEnglish: theseare the people who,
however much they may participate in international interactions, perhaps code-
switching from a different, local variety, are disqualified on the grounds that their
primary and secondaryeducation and socialization have taken place in English. This
in effect means that whatever emergesfrom the spoken conversational data in this
thesis cannot properly be considered to form part of International English.
5. World English
Leaving aside the dreamy hypothesis of World StandardEnglish, the two remaining
Englishes so far dealt with are based on exclusion. On the one hand, Standard
English excludes other norm-providing Englishes from legitimacy in world-wide
communication and on the other International English - Lingua Franca English,
perhaps by reaction, excludes L1 Englishes. Neither English can therefore provide
parametersfor the current researchwhich is based on data which is inclusive and
representativeof English used acrossa wide range of community boundaries.

Brutt-Griffler's use of the term 'World English' seemsto go a long way towards an
all-inclusive variety. Sheproposesa 'unified theory' of World English, to accountfor
the coming together of different varieties of English, those associatedwith native
speakers,as well as so-called 'New Englishes' (Brutt-Griffler 2002:109). Her World
English theory does not accord any privileged status to so-called native speakers;
neither does it exclude them. It appears,on the contrary, to give equal status to all
users, therefore realising the wish expressedby D'Souza, in 1998, that `The term
World English will only make sensewhen all branchesof the tree make an equal
....
contribution to any description of English' (D'Souza 1988).
Brutt-Griffler is able to do this by effectively showing how the so-called 'spread'of
English has little to do with a centre-outwardsmovement, often assumedby other
writers on the subject: Brutt-Griffler writes in particular criticism of Phillipson
(Brutt-Griffler 2002:26). Instead,the existenceof different varieties is due either to
the migration of English-speaking people, providing the so-called 'native speaker'
varieties, or to what she calls 'macroacquisition': the learning or acquisition of
English by people across the world, often in defiance of British imperialist edicts.
Thus World English is as much about 'the linguistic counter-penetrationof the new
varieties found in Africa and Asia on mother tongue varieties' as it is about the
reverse process (op. cit. page 178). This multi-centred account of the formation of
different varieties of English is also to be found in Pennycook: 'multilingual
communities appropriate forms of English that impinge from the globalising
periphery. The metropolis does not spread its English to the periphery, the
tropicopolitans (Aravamudan 1999)rearticulate English' (Pennycook 2002).
In this way, the linguicist' discourse (Phillipson 1992, Chapter 3) of the centre-
periphery dichotomy is avoided: there cannot be any such thing as a high level

'standard'English or a correspondinglow-level 'contact language'or 'lingua franca',
becausethere cannotbe a hierarchy of varieties when eachvariety, native speakeror
otherwise, has developed in a similar way, a point made by Ahulu: 'the same
productive processesare being manifested in the use of English everywhere despite
the geographical spread of the English language today' - both NS and NNS are
bendingthe language,creatingnew rules,in the sameprocess(Ahulu 1997:19) and
by Kandiah:
'transformations of linguistic and discursive structures (to create 'New
Englishes') were, of course, necessary but were not ìnterference'

,
`simplification', èrror' `deviation' `deficiency' ìnadequate or failed
acquisition or replication' ìnterlanguage' etc - even if you forgive all these
things under the blanket of calling the new language the expression of a speech
community. Linguistic change does not come about because of these things,
any more than Shakespeare's English arose out of deviation etc from Middle
English (Kandiah 1998: 99).
Brutt-Griffler is able to reach her conclusion largely becauseof her starting point or
'unit of analysis in linguistics'. Sherejects the 'ideal native speakerin a homogeneous
speech community', (into whose head grammar set up on standard linguistic
is
evidence stuffed - Yngve 40)
1981: and takes instead the speechcommunity itself
as the to
place study language change and development. The 'fixed code' theory of
language is therefore rejected in favour of a process approach whereby varieties of
English evolve within the communities that use them.
Chapter2 40 Is EWL a variety of English ?

The theory seems to founder, however, at the last threshold. Having equitably
accountedfor the existenceof different varieties of English without any hierarchical
distinctions among them, Brutt-Griffler concludes by implying the existence of a
'world English speech community' into which speakers converge, despite the
divergence of their different varieties. This would, naturally, provide a 'world'
variety of English which could be studied and used for teaching purposes.
If World English, as a convergenceof languagevarieties, dependson the existence
of a World English speech community, that community seems either very elusive or
to be a part of the rejected discourse of centre and periphery. Gnutzmann, for
example, seems to consider the term 'speech community' too difficult to apply in
multilingual situations (Gnutzmann 1998: 133), while a whole spate of authors
readily conflate the idea of a 'global community' with the discourse of the 'centre',
the 'metropolis' rather than the 'tropicopolis'. Toolan describes a convergent Global
English, as opposed to divergent varieties, as 'the language of international high-
flyers' (Toolan 1997). Johnson, it will be remembered, insisted on International
English belonging to a technologically advanced world and taught in 'modern'
curricula' (Johnson 1990:305). Quirk and Honey, arguing together for Standard
English, emphasise the fact that 'local' varieties are not good enough for those who
'want access to the world of technology of the industrialised west' (Quirk 1991a) or
who need to process 'all forms of information required in a modernised society'
(Honey 1991). McArthur identifies the users of World English as ' what one is
constrained to call, for want of a better phrase, a Western-educated international
elite', and brings the argument firmly back into the Standard English camp, by
opining that 'what is shared worldwide, in fact, is not the language at large but a
standard variety' (McArthur 1996: 14). Bolton suggests that World English `generally

refers to the idealised norms of an internationally propagated and internationally
intelligible variety of the language,increasingly associatedwith American print and
electronic media' (Bolton 2003:4) and, finally, Brutt-Griffler herself asserts that
World English is tied to the business,technological and scientific community and
cultural and intellectual life (Brutt-Griffler 2002: 11Off).
The coming together of different varieties of English to form a World English, the
index of an identifiable World SpeechCommunity which is exclusive, rather than
inclusive is, like Standard English, both implausible and politically incorrect (see
Pennycook 2002: 'The 'World Englishes' paradigm is, therefore and paradoxically,
exclusionist, not inclusionist'). Brutt-Griffler seems to be attempting to use a
traditional paradigm to account for something which it will not contain. As
Rajagopalan suggests,World English `defies our time-honoured view of language
which is structured around the unargued assumption that every natural language is
typically spoken by a community of native speakers' (Rajagopalan 2004: 112).
It seemsrather more logical to think of World SpeechCommunities in the plural,
rather than to think of a unique World Community, and to think of them as 'fleeting'
rather than fixed, in the sameway that Wagner and Firth consider English as Lingua
Franca talk to be fleeting in nature (Wagner and Firth, cited in House 1999:75), or
'always constituted anew in the ongoing talk', House's characterisation of the
community of Lingua FrancaEnglish speakers (House 1999: 84 and 2002b:259).
In conclusion, World English, as proposedby Brutt-Griffler, may only be thought of
as a variety in sociolinguistic terms. Unlike International English, it is inclusive of
different nationalities, although exclusive in other ways, calling to mind
Pennycook's critique of Kachru's circles which fail to capture the social hierarchies
within them (Pennycook 2002). The trickiness of the World Community theory does
2
Chapter 42 Is EWLa varietyof English?
not preclude data such as that likely to emergefrom this researchfrom being part of
World English: by extendingthe Lingua Francatag to everyone,conversationaldata
collected in international settings might be said to be typical of World English-
Lingua Franca English and, therefore, of the ad hoc construction of world
communities,rather than the reflection of a single, ready-madeentity.
6. Global English
This term has been used loosely by many specialists and non-specialists to refer
simply to the phenomenonunder considerationand not to a real or potential variety
of English. For Phillipson, for example, `Global English', with its links to economic
is
globalisation, another way of referring to U. S. and U. K. norms and, as such, is
thoroughly undesirable(Phillipson 2003).
Gnutzmann, although preferring the term `Global English' to competing labels,
clearly considersit to be the samething as Lingua Franca English, dealt with above
under Ìnternational English' (Gnutzmann 1998:133) and Nair-Venugopal seemsto
share his view, suggestingthat `the notion of a Global English may not in fact be
tenable, for variation is at the very heart of the view of English as a global language'
(Nair-Venugopal 2003).
7. Conclusion
The idea that conversational data collected in international settings may be
considered as a constituent of a variety of English seems to be a tenuous one. It
cannot be thought of as forming part of StandardEnglish since not all speakerswill
either have learned it (even if all have been taught it) or will recogniseit as the norm

to which they should be accommodating.It certainly cannot be thought of as World
StandardEnglish, since such an entity does not exist (and is unlikely ever to do so)
and it cannot really be included into International English, since some of the
speakerswill have been educated and socialised in English. World English is a
possible contender as a label for the data, with the rider that it is a performance
variety, unstable and not institutionalised, and that it therefore refers to the
expressionnot of an existent community but of an infinite, ever-changingnumber of
communities under construction.

Chapter Three
English as a Language
.............
Chapter 3 45 English as a language

.....
1. Introduction
The adjective+Englishparadigm may not be the most appropriateone when it comes
to accounting for instancesof spokenEnglish being used acrossnational and cultural
boundaries. As was seen in the previous Chapter, in order to account for normal
international use of English using any adjective+Englishmodel, either some kind of
exclusion is required, as in the casesof StandardEnglish and International English,
or the concept of variety has to be stretchedto a point where it ceasesto have much
of a meaning: World English is only viable as a variety if it is a performance variety,
and if it is acceptedthat its speechcommunity is forever under construction.
This thesis is centred on conversational data and seeks to help understand how
English is used in international settings, compared to intranational ones. While
participants in international conversations may indeed be constructing a World
Community and, therefore, a World English, they are doing so, ad hoc, using what
they already have, which is not StandardEnglish and not International English. It
therefore seemslogical to investigatecurrent views which reflect languageuse, rather
than languagevarieties.
This Chapter will therefore focus on formulations which use the Ènglish as '
.....
label. It will start with a brief overview of historical instancesof Ènglish as.... ' and
then move on to examine the two prominent Ènglish as... ' labels which seek to
characterisethe way English is currently used across the world. It will conclude by
proposing and defining a fresh Ènglish as.... ' formula to be applied in the remaining
chaptersof the thesis.

.....
2. EFL, ESL and ENL
English as a Foreign Language(EFL) and English as a SecondLanguage(ESL) have
been in common currency, at least in Britain and among British expatriates,for at
least fifty years. ENL (English as a National Language) seemsto be a more recent
coinage,basedon the model of the other two acronyms.
2.1. People, countries, role and function
A rigorously historical view would insist on using these labels to refer primarily to
people learning English and only at a secondarystageto anything else. They might,
with even more historical rigour, be combined with the verb `teach' since the
acronyms were put into circulation so that teacherscould explain to lay people what
exactly their profession involved and how it was distinguished from teaching English
in the traditional senseof making schoolchildren more literate and more sensitive to
literature. Both EFL and ESL are acronyms,then, which were invented for the benefit
of native speakersteaching English either to foreigners or to people who already
spoke another language and needed English in order to prosper in an English
speaking environment. For the majority of teachersin the world, `teaching English'
was quite good enough: where there are no native-speaking children to be taught
English in the making-literate and sensitising-to-literature sense,there is no need to
invent a different term.
The important point is, then, that EFL and ESL are terms originally oriented towards
teachersand learners.The È' in each casedoes not seem,historically, to have been
in question: traditionally, it representedStandard English, encoded in prescriptive
grammar books and canonical literature. It is not different in any way from the È' in
ENL - Englishasa Native Language;only the learnersandusersaredifferent.
Chapter 3 47 English as a..... language

Over time, however, the terms have becomeused to describedifferent communities,
settingsor countries:a specific community or country where most usershave learned
English as a foreign language becomes an ÈFL country'; another setting,
community or country, where usersof English have learned the languagein order to
prosper in a home-basedEnglish-speaking environment acquires the label ÈSL'
country.(Nayar1997,Trudgill andHanna1994:121,McArthur 1996:15).
Alongside this newer application of the EFL and ESL acronyms,there is the growing
acceptanceof them as referring to the role and function of English. Where English is
a `foreign' language, its role has to do with communicating beyond normal
community boundaries,i. e. with foreigners and in particular with foreigners who are
native speakersof English. Where it is a `secondlanguage', on the contrary, its role
may be internal: English may serve the purpose of communicating with co-nationals
in particular settings,typically formal, official or institutional.
Whether having mostly to do with its learnersand users or with its role and function,
there is still no fundamental claim to be made about the È' itself which, when
combined with SL or FL, is unlikely to be anything but the Standard variety.
Chevillet, for example, points out that òf course' an EFL teacher would not
recommend to students to acquire a Nigerian or Indian accent (Chevillet 1993).
While native speakerpractitioners may continue to consider that they teach EFL or
ESL, few learners will say that they are learning either, preferring to declare, more
simply, that they are learning English. Non-native-speakerpractitioners are unlikely
to say they are teaching anything other than plain Ènglish'.

2.2 Confusion over EFL, ESL and ENL
While the three acronyms may have provided convenient shorthand for native
speakerteachers,there is now considerableconfusion in their use and application.
McArthur sums up the situation: "We live in a time when the classic divisions
describing users of English are becoming ever harder to maintain. We all know the
three categories native-users, second-languageusers, and foreign language users.
Once they were fairly clear: the first were born to English, the second had it thrust
upon them in colonial times, and the third was everybody else who knew any English.
Now, however, they have very fuzzy edges"(McArthur 2003: 57).
The fuzziness is particularly evident where the acronyms are applied to countries
rather than to language users and language learners. In South Africa, for example,
there are many people who are considered to be speakers of ENL while for the
majority, English has traditionally been thought of as a SecondLanguage.Within that
majority, there are nations which are monolingual and where English is taught for use
as a foreign language(Brutt-Griffler 2002:146).
In India there are many families bringing up their children to be English speakers,
creating new generationsof speakersof ENL who are not immigrants or descendants
of immigrants and who are living in what has traditionally been classified as an ESL
country (Graddol 2004).
In Europe, Lesznyäk cites Ammon and Witte to confirm that Hungary, traditionally
considered an EFL is
country, witnessing the nativisation of English, in much the
sameway that Gnutzmann suggeststhat the Netherlands and Scandinaviancountries
"are more adequately described as being on their way to gaining ESL-status."
(Gnutzmann 1998:133).

.....
Graddol recognisesthat, given this confusedand fluid situation, we neednew
conceptsin order to recognisethe current diversity of new learnersand new cultures
(Graddol 2004).
2.3. The supremacy of the Native Speaker
As well as their inadequacyto deal with the current situation of different English
users,the three acronymsthinly veil the notion that ENL is superior to the other two.
Stemming from Chomsky's basis for linguistic theory in the "ideal speaker-listener,
in a completely homogeneousspeech-community" (Chomsky 1965:3), the idea is
born, that a real native speaker, monolingual member of a native-speaking
community, is the basis for all judgements about what is and is not good, correct
English.
In what Graddol describesas a Victorian approachto languageand languagevarieties,
monolinguals are `normal', and, therefore, more worthy than anyone of supplying
norms (Graddol 2004). `Second' seemsto lead to second-best

and `foreign' naturally
means òther'. Even Kachru's circles, intended to provide a fresh paradigm,
dislodging the native speakerfrom his pedestal,have the traditional ENL countries at
the centre and the EFL oneson the periphery.
The argumentsagainstusing theseacronymsare, therefore, similar to those rehearsed
in Chapter 2 above regarding the abandonmentof StandardEnglish as a framework
within which to place this piece of research.

Spoken English as a World Language: International and intranational settings
2.4 ENL, ESL, EFL - Conclusion
Given the intention in this researchto focus on language use, rather than on the
identification of a languagevariety or severallanguagevarieties, a label emphasising
languageuse, such as Ènglish as a......... ' seemsto be appropriate.The traditionally
available Ènglish as... ' labels are, however, not appropriateat all, reflecting as they
do values and practiceswhich have disappeared,are disappearingor at the very least
ought to disappear.
3. EIL and ELF

Two more recent Ènglish as.... ' labels are EIL - English as an International
Language - and ELF - English as Lingua Franca. The two are often used
interchangeably but will be dealt with separately here, at least in the first instance, in
order to see if there are any differences between them. As with ENL, ESL and EFL,
there is a primary emphasis on users and learners. Whereas, as has been seen, the
native speaker was supreme in the old order, he or she remains a shadowy, even
haunting presencein the new one. EIL and ELF also refer to the function and role of
English, although the latter does so more strongly than the former, and, at variance
with the older tradition, there seemsto be the implication that the È' may now be
different from the English taught as secondor foreign language.
3.1. EIL
The haunting presenceof the native speakeris there to distinguish EIL, which has
been thought of as all-inclusive, from ELF, where there is a tendency to focus on
non-nativespeakers.

.....
One of the earliest instancesof the term EIL comesin the work of Larry Smith, who
might well be thought of as having invented it. Smith has no hesitation in including
both native speakersand non-native speakersas potential studentsin his proposal for
the teaching of EIL (Smith 1978)
Campbell et. al, are also all-inclusive, suggestingthat EIL meanswhat happenswhen
`speakers
of morethanonecountryor cultureinteract'(Campbellet. al. 1983:36-37)
and Knapp considersthat EIL is 'very broad in its scope and only vaguely defined -
virtually any use of English among speakersfrom different nations is so labelled'
(Knapp 1987:1026). Similarly, Firth does not seem to differentiate between native
and non-native users when he statesthat English as an international lingua franca is
for `different nationality groups' (Firth 1990:270)
Modiano, in a slightly less clear way, suggeststhat EIL is a `general term' for
English which functions well in cross-cultural communication, emphasisingall users
without distinction. His two circles (as opposedto Kachru's three) do not distinguish
between native speakersand non-native, only between EIL-competent speakersand
those who can use English only at a more local level (Modiano 1999a:25).
Jenkins, in her key work on phonology, uses the EIL label, but unlike Smith and
Modiano, restricts her field to people learning English who are unlikely, or
unmotivated to become fluent users. She calls these people `NBES' - non-bilingual
English speakers(Jenkins 2000b:10). Jenkins quite properly and justifiably rejects
the term `non-native speakers'to describe these people but the spectre of the native
speaker is neverthelesscalled up by her deliberate exclusion of L1 (and fluent L2)
users. Jenkins' research subjects are all people whom Kachru would have placed in
his Expanding Circle, who do not share another language and who use English only
to cross `linguacultural boundaries'. Jenkins' use of the term EIL is, therefore, rather

limited whereasfor the previous scholarsit could quite properly be used to describe
interactions between Australians and Americans, Ghanaiansand Nigerians, English
and Welsh.
In order to addressthis limitation, McKay suggeststwo sorts of EIL, `local EIL'
where English may be connectedto local cultures, and `global EIL' where it may not
(McKay 2002: 12). Gnutzmannmakes a similar point, that ìnternational' may mean
just two nations (Gnutzmann 1988) and Chevillet points out that English is not an
international language such as Spanish or Russian, but rather a `world' language
(Chevillet 1991).
3.2 ELF
The ELF acronym could be said to have a longer tradition than the EIL one, even
though its use is more recent. This is becausethe term `Lingua Franca' has been in
more common use for longer than the term Ìnternational Language' and, as such,has
attracted more intellectual interest. Consequently, there seems to be, surrounding the
term ELF, a more properly worked-out notion of what it stands for: its users and
functions have been more clearly defined and investigations have been carried out
into what its existenceis basedon.
3.2.1 ELF is built on the LF construct.
`Lingua franca' is a term which has been used historically to describethe function of
a language as a way of bringing linguistically disparate people together. Samarin
assertsthat a `lingua franca' can be considered such òn the basis of function alone'
and defines its users as `people of different mother tongues for whom it is a second
language'. `Lingua Franca' does not refer to fixable or codifiable grammar,

.....
vocabulary or phonology: 'Since lingua franca indicates an aspect of the use of any
language, it suggests nothing about the structure of that language' (Samarin
1987:370). The official UNESCO definition of `Lingua Franca' is a languagebeing
used ìn order to facilitate communication', by `people whose mother tongues are
different' (Barotchi 1994) and Ammon, discussing diplomatic encounters,suggests
that the term lingua franca holds only where the language used is not the official
languageof any of the countriesinvolved (Ammon 1994:1727).
Insistence on the primacy of function has not been without its critics. Knapp and
Meierkord, for example, consider that the functional conceptualisation of lingua
franca needs reviewing (Knapp and Meierkord 2002:10). Further, the term `Lingua
Franca' itself may often have negative connotations such as the pidginising
tendenciesof what were once thought of as `backward' societies (see Hall 1976 and
Zima 1977, cited in Meierkord 2002: 109-110). More recently, the term has been
treated with distaste by Phillipson, who considers it to be part of a `linguicist
discourse', because it implies a myth of cultural neutrality and has a globalising,
crusading image (Phillipson 1992: 55 and Chapter 3).
3.2.2. Users of ELF
As has been said, the discourse of EFL and ESL is far from absent in the newer
acronyms: native speakers seem mostly to be included in EIL but are rejected by
most people from ELF.
Firth describesELF as a `contact language' between personsfor whom English is the
chosen foreign languageof communication (Firth 1996:

241). and House defines ELF
interactions as those `between members of two or more different linguacultures in
English, for none of whom English is the mother tongue' (Housel999: 74). Seidlhofer

.....
confirms that ELF is a system that serves as 'a language by means of which the
members of different speech communities can communicate with each other but
is
which not the native languageof either' (Seidlhofer 2001:146), but concedesthat
there are lesspure forms of ELF, interactionsin which may sometimesinclude native
speakers(Seidlhofer 2004: 211).
Following the restrictive definition, Lesznyäk studies only language produced in
interactions between non-native users, having defined lingua franca as a language
which is `mother tongue to none of the participants' (Lesznyäk 2002: 166). She later
defines ELF communication, in opposition to EFL communication, as being among
'NNS of English' (Lesznyäk2004:50).
Knapp is slightly less confident. He sets out, in his `Case study of unco-operative
lingua franca communication' by defining his participants as non-native speakers,
interacting with each other in a language which is their `second' one. As he moves on,
however, he perceives that the speakers in his study have varying levels of
proficiency and that some of them are `near-native speakers', while some may have
`become' native speakers. He eventually concludes that, within the ambit of ELF,
there are likely to be many encounters where `real native speakers' interact with
`non-native' and `near-native' speakers (Knapp 2002: 220-221).
3.2.3. ELF is a Function.
It has been noted (Section 3.2.1) that the team Lingua Franca has traditionally been
to a language being used ìn order to facilitate communication'. Since all

applied
language use is concerned with communication, the emphasis here must be on the
facilitation: the implication is surely that, whereas a community or national language

is used in an atmospherewherenormal communicationwill be consideredan obvious
outcome,no such obviousnessattendsthe use of a lingua franca.
ELF may therefore be said to describe the function of a collaborative attempt to
achieve communication, despite the different backgrounds of the users. Knapp and
Meierkord review researchon how LF is `characterisedby a striving to collaborate':
the focus is on accommodationand on strategies,rather than on languageforms and
pragmatics.(Knapp and Meierkord 2002:16). Meierkord extendsthe characterisation
to style, noticing that analysesof LF interactions show a style characterisedby `co-
operation leading to successful communication rather than misunderstanding'
(Meierkord 2002: 120) and Lesznyäk seemsto put strategy and style together to
suggest that ELF may be considered as a process of gaining common ground
(Lesznyäk 2002:165).
3.2.4 ELF and the concept of `common core'
For the process of ELF to take place, for cross-cultural speakers to be able to
communicate with each other and gain common ground, there must be a basis on
which they succeedin doing so. This basis has been referred to as the `Lingua Franca
Core' (Jenkins 2000b).
The idea of a `core' within a language can be traced back to Samuel Johnson's
dictionary in the eighteenth century and the Oxford English Dictionary in the
nineteenth. This dictionary-making tradition embodied two principles - the potential
of dictionaries for `fixing' the languageand standardising it and the identification of
a core (Bolton 2003:26).
Hockett used the term `common core' to describe the most fundamental possible
factor in explaining how people communicate in speech. The `common core' of a

person's idiolect is the `total set of features' shared with another person or other
people thus making communicationpossible (Hockett 1958:336). It is this core, then,
which provides the basis for accommodationand finding common ground.
Bex extendsthe notion from the personal,idiolectal domain, to the idea that a whole
group of people ('native speakers'in Bex's paper), `have an idea of a 'core' variety of
Englishto which theycanmove(to a greateror lesserextent)in unfamiliardiscourse
situations' (Bex 1993:257).
Preisler, too, takes up the `core' theme, widening it out still further, and using it to
describeELF whose function is, in turn, a `function of what speakersof English have
in common' - the common 'core'. For Preisler, the `core' stands for the `structural
and cultural propertiesof English sharedby native speakers'(Preisler 1995:343).
With Jenkins comesthe idea that the common core permitting EIL communication is
not what is sharedby native speakers,but rather what may be sharedby the people
Jenkins identifies asNBES (seeabove,Section 3.1)
When accommodationfails, when usersof English find that they cannot find enough
shared ground to communicatewith each other, then they must resort to a common
core, renouncing their own varieties or idiolectal forms and giving up on attempts at
accommodation.Unlike native Jenkins'

speakers, participants in EIL, all non-native
users of English, do not have a ready made common core of phonological elements,
so one needs to be contrived, based on sounds which are comprehensible to all
(Jenkins 2000b: Chapter3).
By suggestingthat a core needsto be contrived, Jenkins provides a role for teaching
EIL. Learners of English, and `native speakers' who want to use English
internationally, need to be trained in the contrived core in order to have somethingto
Chapter3 57 English as a language

.....
build on for their own idiosyncratic pronunciation, and to fall back on when an
idiolectal variety fails.
Seidlhofer, in her apologia of Ogden and `Basic English', seems to agree with
Jenkins: rather than try to derive a core from `natural' English, as Quirk attempted,
she would rather prescribe a core, thus making a `programmatic statement'
(Seidlhofer2002:275-6).The corewould, presumably,be basedon resultsfrom her
ELF databaseand go beyond the limits Jenkins imposed on herself by choosing to
deal only with phonology.
3.2.5. ELF as a language variety
Despite the apparent clarity with which Ènglish as... ' labels appear to refer
exclusively to language use (and therefore to users), there is a tendency to use the
term ELF to refer to a thoroughgoing languagevariety, on a par with the Englishes
dealt with in the previous chapter. The tendency is not limited to ELF either: in a
recent paper, Crystal finds himself referring to EFL as a `native language' (Crystal
2001:56).
The tendency may stem from the searchfor, or establishmentof a `core', mentioned
above, or may derive from the tradition of identifying a Lingua Franca as an
autonomous language or variety. One of the key criteria for ascribing either full
languageor languagevariety status to a Lingua Franca is that of stability, discussed
in Chapter 2. Knapp and Meierkord note that, where a Lingua Franca in the past has
stabilised in specific plurilingual areas, it has become a variety (Knapp and
Meierkord 2002: 9-10).
Seidlhofer, whose use of `system' to describe ELF has already been mentioned,
seemsto suggest that ELF has indeed stabilised, or is at least stabilising, since she

.....
makesmention of `the international ELF speechcommunity' which is developing its
own language norms (Seidlhofer 2002:273). She wonders whether she will find
`commonly used constructions and sound patterns which are ungrammatical in
StandardL1 English but generallyunproblematic in ELF communication' (Seidlhofer
2000) and eventually wants to make ELF a 'feasible, acceptable and respected
alternative to ENL' as a resourcefor teaching (Seidlhofer 2001:150) - when the ELF
community provides the norms it has developed.The existenceof an ELF community,
similar to, but more limited than the World English community envisagedby Brutt-
Griffler (See Chapter2, Section 5), has been questionedby McKay, who is confident
that there is no such thing as an EIL group (McKay 2002:29) and by House, who
considers that any such community is constituted anew every time there is an
international encounter(House2002b:259).
Mauranen is more forthright in her use of the word `variety' to describe ELF, but
rather than attach the word to one community, she suggests that there are many ELF
communities and therefore many varieties of the variety (Mauranen 2003: 516). This
view chimes with Gramkow Andersen's definition of ELF, where each combination
of interactantsnegotiatesits own variety (Gramkow Andersen 1993:108).
Mauranen's consideration of some ELF communities as `fairly-well established'
resonateswith the stability criterion. Interestingly, Mauranen begins her paper by
using the word èstablished' to describe `Native' or `World' varieties in opposition to
ELF ones. Her claim that ELF is a variety seemsto rest on the idea that, following
Jenkins,thereis an ELF `core' which, shethinks, it is reasonableto supposewill be
different from a Native Speakerone, and also on the stabilisingtheory: advanced
learners of English use, for example, structural simplification as a strategy to help
facilitate communication;continueduse leads to the establishmentof simplified

.....
structures as part of the ELF core (Mauranen 2003: 515). As with Jenkins and
Seidlhofer, Mauranen is interestedin setting proper targets for learners of English.
The difference is that she favours different norms for different ELF discourse
communities,rather than one international set of norms.
Promodrou is scepticalabout Seidlhofer's project. He does not mention Mauranen's,
but his critique might just as easily apply. He considers it unlikely, firstly that
learners from diverse European backgrounds will display the same common core
features, even where they are supposed to be part of the same ELF discourse
community, and, secondly, that they will display the same degree of simplification
when communicating in ELF. He therefore considers it implausible that ELF will,
even eventually, have endonormativestandards,

arguing that it is not, and will never
be, an indigenized variety reflecting the identity of a community (Promodrou,
forthcoming a).
A more convincing argument,perhaps,leadsback to the conclusion of Chapter2 and
the acceptanceof World English as a performancevariety. It is the one put forward
by Meierkord and Knapp who feel that Ènglish as a lingua franca is a variety in its
own right' because of a number of common strategic characteristics (Knapp and
Meierkord 2002: 19). There is a risk, however, that this claim may serve merely to
make matters more complicated than ever. If ELF is considered a variety in its own
right on the strength of strategies,then to which variety are people using who do not
use the same strategies but who nevertheless attempt to communicate across
linguacultural boundaries? In other words, is a strategy-determinedELF the only
variety used in Lingua Franca settings?Or doesthe setting determine, in the end, that
what is happening is the use of a Lingua Franca?

4. EWL - English as a World Language

Although the EIL and ELF formulae are more satisfactory than Àdjective +
English' oneswhen attemptingto accountfor the kind of data this thesis is interested
in, there are neverthelesstwo problems.As hasbeen seen,the problem of exclusivity
is not solved since ELF doesnot iclude native speakersand, at least for Jenkins,EIL
is basedon the Lingua Francacore: whereasthe terms have arisen partly becauseof
the myth of native speaker supremacy, the native speaker construct still informs the
thinking behind them.
Secondly, there is a risk that EIL and ELF will represent attempts to find a new
variety of English. Leaving aside the somewhat dubious category of `performance
it
variety', seemsquite clear that no such variety is likely to emerge.The use of the
two acronymsmay neverthelessprove confusing in this respect.
That said, the function approach to ELF seems to be a fruitful one in terms of
accounting for what happens when people use English across linguacultural
boundaries,resting as it doeson the notion of a `common core' which all usersdraw
on.
In order to mark out this thesis from the work so far referred to, it is therefore
to
proposed employ the tinder-used acronym EWL - standing for English as a World
Language.This term standsin the useful set of Ènglish as.... ' formulae but, given
its relative lack of use, doesnot carry with it the exclusivity of those dealt with so far.
EWL will now be critically defined in terms of the people it includes. Following that
EWL will be considered in terms of its functions and how its users might perform
them; argumentswill also be further rehearsedfor not considering EWL a variety of
English.

.....
4.1. EWL Users
The reasonsgiven or implied for excluding groups of people from EIL and, more
particularly, ELF seemto have to do, eventually, with languagelearning. In order to
provide learners with realistic, politically correct targets, native speakermodels, or
ÈSL' models, are thought to be inappropriate. While this may be true, almost
obviously true, if by `native speaker models' the kind of interaction is intended
whereby natives of Birmingham (either in the West Midlands, UK, or in Alabama,
USA), Bombay or Mombasaspeakor write in English to other natives of those towns.
But where native speakersare part of an international or intercultural discourse,the
logic is not quite so strong for at least two reasons.The first has to do with clarifying
who native speakersare, in an ever more globalised, fluid world and the second,
connectedreasonhas to do with the weight of numbers.
4.1.1. Difficulties defining `native-speakers'
Davies, in his key work on the native speaker, provides a list of defining
characteristics,one of which is 'the unique capacity to produce fluent, spontaneous
discourse'(Davies 1991a:150). He goes on to gloss this assertionby stating that `The
native speaker is also expected to exhibit normal control especially in fluent
connected speech' (op. cit. 164). It seems clear that, by 'control' Davies means
competence,
communicative or knowledgehow: in a previousarticle he makesthe
connection betweenthe two terms explicit (Davies 1989).
It ought, therefore, to be possible to monitor the speech of a random selection of
users of English and to determine who among them are native speakerssimply by
paying attention to their 'capacity to produce fluent, spontaneousdiscourse', their
'normalcontrol'.

.....
Plainly, however, this endeavourwould not be likely to succeed for at least three
reasons.
i. ) Firstly, since proficiency is measurableon a cline, rather than in a digitalised,
polar way, it is very difficult to determine at which point along the cline non-native
speakerproficiency becomesnative-speakerproficiency (Or, using Davies' term, at
which point a speaker's control could be said to be `normal') (McNamara 1996).
Some formal proficiency tests, for example, the Cambridge ESOL Certificate of
Proficiency in English, once gave `near native-speakerproficiency' as a yardstick
against which to measurethe performanceof top-rated learners of English but it was
not clear where the line had to be drawn betweentheseand thosejust under them.
ii. ) Secondly, `normal control in fluent speech', unless limited to monologues
delivered into a void, supposesinteraction with other speakers.The extent to which,
therefore, a speaker can be said to exercise 'normal control' and to be a fluent
communicator, may only be judged by how effectively he or she conveys the
intended message,which clearly also dependson the ability of the hearer to make
senseof what is heard (Deen 1997: 15).
iii. ) Thirdly, the capacity to produce fluent, spontaneousdiscourse can only be
guessed at by attending to actual instances of speech, which are likely to be of
different quality when delivered by the same speaker in different contexts. Deen,
reviewing the work of Spitzberg and Cupach (1984), points out that àn individual's
competencecan only be measuredthrough his performance in a certain context. This
performance may vary from context to context and thus it cannot be objectively
determinedfor the individual' (Deen 1997:14-15).
It is clearly not possible to define native speakersby their performance. Kachru and
Nelson succinctly conclude that `being labelled a native speakeris of no particular a

.....
priori significance, in terms of measuring facility with the language' (Kachru and
Nelson, 2001:14).
Perhaps competenceis a more reliable marker of the native speaker: Coppieters
certainly seems to think so. Having, he claims, identified a group of non-native
speakerinformants whose English had becomethe functional equivalent of a mother
tongue (presumablythey had the capacity to produce fluent, spontaneousdiscourse)
he shows that they lacked native-speakercompetence,intuitions about what Davies
might have called their 'personal English grammar' Davies 1991a:150) and any
differences between this and 'the shared grammar of their English-speaking
community' (Coppieters:
1986). The problem, however, is that Coppieters first
identified his informants as 'non-native speakers'and then conductedhis investigation
into their competence.This begs a number of questions: how did Coppieters know
that his informants were non-native speakers?If he had first investigated competence
and then deducedthe native or non-native statusof his informants from an analysisof
this, would the results have beenthe same?How significant was his sample of native
speakers providing the reference point for showing native-speaker competence?
These questions are important, particularly as Coppieter's findings have been cited in
defence of using native speakers to teach English in preference to non-native
speakers (Quirk 1991b).
Coppietersseemsto think that his 'near-nativespeakers'will never achievenative-
speaker competence.Davies (op.cit. ) is slightly more generous, limiting himself to
wondering whether "second language speakers have access to a second language
langue" and, if they do, whether it is "the samelangue as the langue of first language
"
speakers? (Davies 1991a: 19) but neverthelessopining that there is no reason why
Chapter3 64 English as a language

.....
they should not have this access or why it should not be 'the same langue'
(op.cit.: 150).
Using language performance as a measureof whether or not someone is a native
speaker turns out to be a dead end; using competence seems to require the pre-
identification of native speakers.A sociological definition might therefore, be more
useful. Severalwriters have affirmed that to be a native speakermeansto be part of a
social, rather than a linguistic system (see, for example, Coulmas 1981:18, Mey
1981:76) and severalothershave implied as much by saying that native languagesare
social constructs rather than objectively distinguishable linguistic realities (Corder
1973,cited in Pennycook 1994:27, Annamalai 1998: 149-150).
The most straightforward way of defining native speakers in sociological terms is by
seeing them as a group of people who were born into an English-speaking
environment or who at least learned English in early childhood. This 'folk wisdom'
view of the native speaker is supported, among others, by Davies, who gives early
childhood acquisition as one of the characteristics of native speakers, by Prabhu and
by Mufwene (Davies 1991a: 150, Prabhu 1998, Mufwene 1998: 111). The situation is,
however, far from clear cut in all cases,as Medgyes points out:
Let us take Juan, for example, aged 9, who has been living in the United States
for five years.His father is a Mexican immigrant, his mother comes from Norway.
They both speak to Juan in their own mother tongue. Which is his native
language, English, Spanish, or Norwegian? All three of them? None of them?
(Medgyes 1992:341)

Similarly, Leung, Harris and Rampton show up the inconsistencies with the term
`native speaker' becauseof Creoles, languageswitching and bilingualism and assert
that a simple `category' approach to defining the native speaker does not work.
(Leung, Harris and Rampton 1997)
While it may not be possible to define native speakersby the circumstancesof their
birth or early childhood,it seemnevertheless

to be accepted,for betteror for worse,
that the term still has to do with social groupings. Without the 'birthright' criterion,
three possible social definitions of native speakerremain.
i. ) The first of thesedefinitions may, after Bloomfield, be termed the 'tribal' definition
(Bloomfield 1933). Bloomfield suggeststhat tribal attitudes are gradually built up as
the result of people with similar languagescoming together and agreeingon rules and
regulations which, eventually, make up a 'native language',allowing its users to call
themselves 'native speakers' of that language (see Bartsch 1988, cited in Davies
1991a:125). Speakersthen agree to surrender their individual linguistic identities,
modifying their own languagesto createthe new social reality of the native-speaking
group, with which they now identify.
The group finds it sharescritical attitudes about languageand its use, or rather about
the way people use language;theseattitudes include the norm of excluding thosewho
do not `surrender'. These attitudes and norms have to do not only with the grammar
and vocabulary of the created language, but also with accent, registers, pragmatic
forms and conversationalstyles (Davies 1991a:123).
The problem would appear to be the application of the term 'English', coupled with
'native-speaker',to whatever is spoken by these 'tribes'. There may well be
communities (or tribes) whose members refer to themselves as 'native speakersof
incomprehensible
English'who are nevertheless to other'native speakersof English'

.....
and whose use of the term 'native speakers'applied to themselvesmay be questioned
by others. Canagarajah, for example, points out that members of postcolonial
communities would consider themselvesnative speakersof indigenised variants of
English (Canagarajah1999a). But learners and teachersof English seem to contest
this, as Amin's experienceattests:her learnerswere convinced that she could not be
a native speakerbecauseof her skin colour; her colleagueswere equally shockedat
sense,'one of them' (Amin

her claim to be, in a language-acquisition-and-competence
1999).
Kandiah reports a chilling episodewhere a teacherconsidering himself to be a native
speakerwas refuseda post becausehis prospective employers were dubious about his
ethnicity and changed the relevant job advertisement so that it included the word
'Caucasian',lest there should be any doubt about who is or is not a 'native speaker'
(Kandiah 1998).
This leads logically to the second social definition which may be termed the
'ideological' definition.
ii. )
Skin-colour-basedracism apart, one of the reasonsfor some people being rejected as
native speakersof English, or, indeed, for their refusal to consider themselvesas such,
has to do with the obvious connection between 'native' and 'nation' and, therefore, in
this case, between 'English' and 'England'. Blank shows that language and national
identity were associated as early as 1414 when Ènglish representatives at the
European Council of Constancecited the "difference of language" as one "by which
divine and human law is the greatest and most authentic mark of a nation and the
essenceof it" (Aston, 1968: 41, quoted by Blank 1996: 1). The creation of a standard
tongue, under the auspices of King James 1" of England, and associatedwith the

.....
construction of nationhood led, in succeedingcenturiesto a strong sensethat native
speakerswere those people, presumably the majority living within England, who
identified with the English nation and consideredthemselvesusers of the national
language. It is a curiosity that the standard tongue is called 'English' rather than
'British', given that the constructed nation resulted from the union of Scotland to
England and Wales.
The importance of nationhood and the national tongue in the construction of
colonialism and imperialism is underlined by Pennycook who refers to those leaving
England to settle in North America and, later, in the dominions of South Africa,
Australia and New Zealand as having taken with them, in the first instance, a strong
senseof Englishness(Pennycook 1994); the notion that the language widely spoken
in the latter three countries is anything but English has only recently been contested
with the publication of Australian English, New Zealand English and South African
English word books. The case of North America is slightly different, for obvious
reasons.After the War of Independence,

the national language of the United States
was constructed as the badge of the new nation and a new set of native speakers(of
English, perhapsagain curiously, rather than of 'American') came into being.
Native speakersmay thus be identified as those who identify with the nations that
have created national English languages, starting with England. As a mark of
must,
nationhood, native-speakerness perforce, imply monolingualism: according to
the ideology which created the notion of Standard English and, therefore, native
speakers,it is not possible to identify with more than one nation or more than one
culture at a time (Annamalai 1998:152).
The ìdeological definition' therefore excludes those such as Amin (see above): if a
distinction is to be drawn, it is not between those who have joined a speech

.....
community and those who have not, but between those who have conjoined nativity
and speechhabits, who have `bought' the national line in one way or another and
those who have not. In order to becomea native speaker,then, or in order to acquire
nativity, one has also to accept, and be acceptedinto, the national cultures of the
traditional monolingual English-speakingnations and then to believe in the whole
complex creation. U.N. Singh (1998) calls it à club which people may try to
gatecrash' but a club which, importantly, expandsits membership with colonialism,
imperialism, industrialism and commercialism(Singh 1998).
iii. ) Singh seemsto have been referring to those people, not necessarily of English
(or British, American, South African, Australian, New Zealand) origin who became
native speakersby espousing a theoretically alien culture. These may have been
people holding or seeking positions of power in colonies or the British Empire or,
more recently, those holding or seekingpositions of power in international, 'Wester'
industrial and commercial companies.The group may also include those who feel a
strong affinity for English or who have decided to assimilate into a traditionally
native-speakingcommunity (Rampton 1990).
Florian Coulmas, in his editor's introduction, dedicatesthe work to 'those who made
me a native speaker', thereby presumably declaring his strong affinity with one of the
national cultures traditionally associated with native speakers or the international
culture representedby English (Coulmas 1981). Connor becomes a native speaker
writer of English by giving up her Finnish culture as she seeks to assimilate into
American academicsociety (Connor 1999).
Thus,througha processof 'imaginednativity', servingto `keepthe other[i.e. the non-
native speaker] excluded from the self on ideological grounds' or simply through a
process of 'self-ascription', anyone can, potentially, become a native speaker, can

.....
'gatecrashthe club', with the obvious rider that he or she must be acceptedas suchby
the other, longer-established members (Annamalai 1998:151, Davies 1991a:8,
Kramsch 1998:19).
Sociological definitions of the native speakerare, then, fraught with problems: each
definition seems either to leave some obvious candidates out, or to include some
obvious non-candidates.
4.1.2. Native speakers are a significant minority
Although native speakermay be hard to define, an approximative use of the word has
allowed many to create statistics showing the relative numbers of native speakersof
English compared to non-native speakers of English. Perhaps most notably, and
certainly very influentially, Graddol has estimatedthat the number of native speakers
of English is currently the sameas the number of ESL speakersbut that the number
of EFL users is the same as both numbers combined, putting native speakersinto a
clear numerical minority compared to the overall number of English users. He
predicts that, in the future, the proportion of native speakersto the overall number of
English userswill be even smaller (Graddol 1997).
The fact of this minority has been used by those wishing to describe ELF as a non-
native speaker phenomenon and as a possible reason for excluding native speakers
from their ELF data (most of the contributors to Knapp and Meierkord, 2002).
The minority constituted by native speakersis, however, a very large one. Graddol's
figures suggest that the number is currently around four hundred million (and
Graddol, for whatever reason, excludes most African or Indian native speakers,
including them in his `L2' category).While many, even most of thesenative speakers
may never use English cross-culturally, it may still be supposed that a significant

.....
number of them do: foreign travel, international business,educational exchanges,for
example, are as much a part of the life of many native speakersas they are of non-
native speakers.It is therefore obvious that significant numbers of native speakers
in
participate cross-culturalconversationsin English.
Furthermore, the number might easily grow. The existence of people such as
Coulmasand Connor,who think of themselvesas having becomenative speakers,
has already been indicated. Those whom Gnutzmannhas identified as shifting from
being EFL usersto ESL usersmay also think of themselvesone day, or be thought of,
asnative speakers,again, swelling the numbers.
4.1.3. Four case studies
Akin to the fictional case of `Juan' cited above, some real participants in the
conversationsconsituting the data for this thesis should serve to confirm both points
about EWL users.
a) Two women of Nigerian nationality describe themselves as native speakers,
having spoken English from birth, in their families, including to their grandparents,
and having learned another language only patchily. That other language is Yoruba
and is identified by the women as `their' language.In other words, the languagethey
identify with is spokenwith considerablyless fluency than English. This seemsto be
a clear casefor applying Rampton et. al. 's distinction between ìnheritance' (Yoruba,
in this case) and èxpertise' (English), rather than keeping to the `native-speaker'
category (Rampton 1990). Other Africans in my study questioned the right of the
two women to call themselves native speakers, arguing that this represented a
travesty of their roots (Appendix K, D1 and D7).

.....
b) A woman born and brought up in the United Kingdom, speaking English from
birth and unable to speak any other language, is reticent to call herself a native
speakerbecauseshe considers`her' languageto be the languageof her parentsand
ancestors,which is not English. (Appendix K, D3).
c) Several people born in Nigeria describe themselvesas non-native speakersbut
haveusedEnglishsinceearlychildhoodfor all purposes,including communication
within their homes.(Appendix K, D 1)
d) An Amharic- and English- using man consideredthat he would not be a suitable
participant in the data-collection exercisebecausehe had lived for a long time in the
United Statesand was thereforemore of a native-speakerthan a non-native speaker.
He had lived in the U. S.A. for four years.
People of this sort are becoming more typical by the day; they are part of a growing
majority of users of English who cannot easily be classified as either `native
speakers' or `non-native speakers'. Knapp appears to have come to a similar
conclusion when collecting ELF data: he found that many of the participants in so-
called ELF interactions were on the edge, as it were, neither `true' native speakers,
nor non-native speakers; some participants had `become native speakers'. For
convenience,Knapp refers to thesepeople under the term `near-nativeness'.He also
makes the point that there are many international gatherings where English is used
as a non-native languageby many participants but where also `true' native speakers
are present (Knapp 2002: 220-221).
Either way, then, whether the term `native speaker' should be consignedto history as
in
meaningless today's world or, conversely,should be retained and applied to
swathesof new `club members', there seemsto be no reasonto exclude vast numbers

.....
of people from the study of English as it is used in international and intercultural
settings.
4.2. EWL as a function
Following the researchinto ELF, it seemslogical to consider EWL as a function, to
focus on the use made of English, by whomever and in whatever form, in order to
facilitate communication on a world scale. In this way, the overall researchquestion
posed in this thesis may now be refined, or broken down into more specific sub-
questions.
Traditionally, cross-cultural spoken interactions have been studied by looking at the
way non-native speakers interact with native speakers (Varonis and Gass 1985,
Zuengler 1991, Deen 1997). Many recent studies have instead focused on ELF
interactions, amongnon-native speakersonly (Tarone and Yule 1987, Meeuwis 1994,
Firth 1996, House 1999 and 2002b, Meierkord 2000). Lesznyäk has provided a study
where the two sorts of interaction are contrastedwith each other (Lesznyäk 2004). In
most of thesestudies,the focus is on interactants' strategiesand styles, in other words
on the way they make their languageresourcesfunction.
The purpose of this thesis is to follow in those traditions, with the difference that the
terms of comparison are to be shifted away from the NS-NNS dichotomy. Instead, a
comparison will be drawn between the way people make language function in
international settings and the way they use their languageresourceswhen interacting
with co-nationals.Thus EWL is seenfirmly as a function of English and any patterns
emerging will be performance patterns having to do with interactants' strategiesand
styles.

It may well be that EWL is characterisedby collaboration and co-operativeness,just
as ELF has been (See above, Section 3.2.3.). But whereas this behaviour has been
identified by comparing ELF interactions with those between native speakersand
non-native speakers, it may be that, by including native speakers in EWL and
comparing this with English used in more homogeneous settings, no such
characterisationis possible. It is equally possible that all users of English become
more collaborative and co-operative in international settings than they are when
interacting with their co-nationals,or that their level of comity is the same,whatever
the situation.
4.3 EWL and the variety and `core' questions
It has already been made clear that there can be no intention here of seeking to
identify an EWL variety. The very choice of an Ènglish as... ' label was motivated,
in part, by the futility of that exercise.The loophole by which a claim might be made
for a performance variety will not be explored for reasons given above (Section
3.2.5).
The less futile route, that of looking for a `core' might seem more appealing except
for two prominent objections. The first of these is a practical one: this thesis is
limited in its length and scope by the parameters laid down by the University of
Nottingham and by the extent of data it has been possible to collect. Conversational
data amounting to 50,000 words, the backboneof this thesis, cannot competewith the
corpusunder by
construction Seidlhofer
and could thereforenot be relied on to yield
weighty information regarding what is or is not a `core' item.
The second objection is more theoretical but has practical implications similar to
those posedby the first objection. Having decided to include all languageusers in the

.....
researchproject, there can clearly be no question of isolating an ELF core which is
different from an ENL one. An interestingquestionmight be the extent to which each
language user exploits his or her idiolectal core for international purposes or,
following Bex, his or her `national' core. While it should be possible to identify
differences between the way individuals use English in homogeneous and
internationalsettings,a muchlargerset of datawould be requiredin orderto match
thesedifferencesto `cores'.
5. Conclusion
This Chapter has completed the preliminary study of research in the area of the
international use of English by rejecting the two most frequently used formulae for its
definition. By insisting on exclusivity, the EIL and ELF labels have been seento be
inadequateas frameworks within which to place the current research.Instead, a new
framework has been suggested,using the acronym EWL for English as a World
Language.By showing why it is impossibleto exclude native speakersfrom the EWL
scene it is hoped that a firm basis has been established for moving ahead and
describing how spoken English functions internationally. The next Chapter will
therefore attempt to anticipate matters,using previous researchresults to predict what
in fact may happen, and so to refine the research question still further.

Chapter Four
Hypotheses concerning EWL

interactions
Chapter4 76 Hypothesesconcerning
EWL interactions
1. Introduction
Having concludedin the precedingtwo chaptersthat the object of this study is best
characterised under the EWL heading, the thesis will now focus on existing
literature and research results which might indicate how EWL is likely to be
characterised.It must be borne in mind that, within the parametersset out for the
present project, little or no previous researchhas been conducted: what data has
been gathered has been either ÈFL' in nature - native speakersinteracting with
non-native - or ÈLF' - non-native speakersinteracting with each other. Hypotheses
concerning EWL must therefore be drawn from these two areas, rather than from
more specific precedingresearch.
This chapter will therefore begin by examining reported ÈFL' interactions and
relevant literature concerning what may or may not happen when an EWL
conversation includes native speakers. This kind of interaction has often been
characterised as asymmetrical, with native speakers being considered (or
considering themselves) as the owners of, and authorities over English, thereby
making non-native speakers accommodate to them. At the same time,
accommodationgoing in the opposite direction has been characterisedby Foreigner
Talk, thought of aspatronising, even humiliating.
The second part of this chapter will deal with ÈLF' interactions and examine the
notion that these are somehow symmetrical in nature. While the use of English in
settings where all users are non-native speakers,

with different cultural and linguistic
backgrounds, may lead to communication difficulties, growing research results
suggest that the potential for this is largely overcome by co-operative and
collaborative attitudes and universal mechanismsfor accommodation. Finally, brief
4
Chapter 77 Hypotheses
concerning
EWL interactions
mention will be made of how languageforms may develop in this kind of setting,
whether as a result of `pooling dialects' or agreeingon a `core'.
2. EWL àsymmetrical' conversations

In Chapter 3, an attempt was madeto deconstruct`native speaker' in order to justify
including all users of English in this research.The point was also made, however,
that the native speaker myth continues to pervade the whole area of EWL. As
Kandiah conclusively argues, the native speaker is not dead (Kandiah 1998:90).
Having decided not to exclude native speakersfrom EWL, it is important to take
stock of their presumedposition in it and of the difficulties traced in the literature.
A common assumption has it that when a conversation takes place in English in
which one or more of the participants is a native speaker, the conversation is
asymmetrical (Ammon 1994:1727,Annamalai 1998: 149-150).
The lack of symmetry arisesbecauseof the presumedcentrality of the native speaker
in the construction of Standard English and the entailed perception of the native
speaker's exclusive ownership of, and authority over English. This in turn leads,
according to much of the literature, to non-native speakershaving to accommodate
to native speakersand to a potential loss of confidence as a result. The situation may
be exacerbatedby native speakers'attemptsto accommodateto non-native speakers
via Foreigner Talk.
EWL interactions
2.1. The centrality of the native speaker

It was pointed out in Chapter 2 that, until very recently at least, most learners of
English, whether learning within an English-speaking environment in an ÈSL
country' or in traditional ÈFL' settings, have, at one stage, or even at most stages,
been greatly exposed to `Standard English' and to the cultures of the two main ENL
countries. Stereotypical scenes involving a Mr. and Mrs. Smith drinking tea on the
lawn have been used in order to present Standard British English grammar and
vocabulary and, at a successive stage, along with the fashion of teaching `functions',
learners have been introduced to language-behavioural norms appropriate to British
or American middle classes (e.g. Eckersley 1959, Abbs & Freebairn 1979, Gairns &
Redman 1996).
Given the ubiquity of British and American models of language and behaviour in
teaching materials, it is hard to counter complaints about linguistic and cultural
imperialism. Further, given the ubiquity of the native-speaker myth, people learning
English will traditionally see themselves as non-native speakers and, therefore, as
not belonging to the social group representing the target of their learning. Davies
highlights the problems faced by `foreign' speakersof English when they `wish to
identify with the community which they regard as defined in terms of target native
speakerness'; the problems have to do with acceptance, not with (language)
knowledge (Davies 1991:69).
For some, the problem continues beyond non-acceptance and becomes one of
effective disempowerment. Annamalai asserts that "the social construct of native
language is to sustain an asymmetrical power relation through the ideology of
otherness. It ensures that some speakers of a language are defined (...... ) as others in
Chapter4 79 Hypotheses
concerning
EWL interactions
contrast with the self'. Annamalai emphasisesthe individual, local and national
levels of this ideology (Annamalai 1998: 149-150).
The difference in levels of power may be analysed in terms of the conflation of
native speakerswith `StandardEnglish'. The overlapping of the two constructswas
dealt with in Chapter 2 (Section 2.3) and the perception of disempowerment
from it has led to pleas to districate one from the other. Wong called for
resulting
'non-native Englishes'to ceaseto be referred to as 'non-standard',(Wong 1982:263f1)
and Ahulu reiterates the request, calling for the concept of Standard English to
recognise and accommodatethe 'developmentsbeing described as New Englishes'
(Ahulu 1997:19).
The fact that, as ChapterTwo attemptedto show, there is no logical connection at all
between the two terms - indeed, it is difficult to find logical support for either term,
let alone a connection between the two - does not affect perceptions of
connectedness which therefore confer power upon `native speakers' to the detriment
of `non-native speakers'. Nair-Venugopal sums the situation up thus: "The
is that `native speakers' are in a position to wield power where cultural

suggestion
hegemony can be exerted as linguistic hegemony by using a standard dialect or a
native-speaker variety" (Nair-Venugopal 2003: 40).
2.1.1 Ownership
Given the historical and actual connectednessbetween StandardEnglish and native
speakers,these have traditionally been considered to be the òwners' of English.
Kandiah considersthat the notion of proprietorship appearsto 'be built centrally into
the term' (Kandiah 1998: 82-3) and Modiano's critique of Kachru's three-circle
model points out that the presence of Èngland and its former colonies' in the ìnner
concerning
EWL. interactions
circle' means that the inhabitants of these countries possess the language
(Modianol999a: 23). Modiano's use of `former colonies' seemsa little misleading
here since English users in Britain's former colonies, where they fit into Kachru's
circles, seemto be in the outer circle, the only former colonials in the ìnner circle'
being Americans. Samimy and Brutt-Griffler note that a 'non-native speaker'
candidatefor an English-medium MA in TESOL felt subservientto native speakers
since he considered his British and American colleagues to be the owners of the
language(Samimy and Brutt-Griffler 1999:139).
Ownership includes the right, or the power to be able to use language resources
unavailable to the dispossessed

and to be a partner in the processof languagechange,
from which non-native speakersare excluded.The question of which languageitems
to
are available native speakersonly is discussedby Janicki, who cites Marton and
Preston's personal possession hypothesis (Janicki 1985:13, citing Marton and
Preston 1975), while Kandiah points out that language change has always been the
preserve of native speakers. He makes his case by comparing on the one hand the
way New Varieties of English (NVEs) are described by inner circle linguists using
such terms as ìnterference', `simplification', èrror', `deviation' and `deficiency',
with, on the other, the way that Shakespeare'sEnglish is derived from Middle
English. WhereasNVE languagechangeseemsto be about deviation, native speaker
language change is a straightforward, linear process (Kandiah 1998:99). Ironically,
perhaps, native speakershave also traditionally been accorded the right to include
`borrowed' words from other languages,including languages spoken by bilingual
users of NVEs. When users of NVEs include words from other languages,this is
generally consideredto be a caseof ìnterference', rather than `borrowing'.
Chapter 4 81 Hypotheses concerning

EWL interactions
Just as many voices have been raised calling for Standard English to encompass
non-native speakers,there have, since the early 1990s,been voices contestingnative
speakerownership of English. Many of theseprotestationscome from among native
speakersthemselves(e.g. Smith 1991, Bryan 1994, McArthur 1996:4) while others
spring from the disempowerednon-native speakers(e.g. Braine 1999).
Higgins attemptsto prove objectively that non-native speakersare indeed owners of
English: she usesConversationAnalysis procedureson a set of conversationsabout
the correctnessor otherwise of a range of sentencesand finds that her participants,
from India, Africa, as well as from the United States,refer confidently to their own
use of English when correcting or confirming the correctnessof English usage. She
does however concedethat, when in the companyof native speakers,her informants
might have a less confident approach and accept the authority of the òwners'
(Higgins 2002).
2.1.2 Authority
Ownership obviously entails authority. Native speakers have traditionally been
accorded the authority to decide what the languagenorms are and what constitutes
an error. Davies confirms, or perhapsopines that à foreign languagespeakercannot
be appealedto for authoritative pronouncementsabout the language's rules and its
use. First languagespeakersof course can be;' (Davies 1991a: 23) and Kachru sees
the ìnner circle' as norm providing (Kachru 1985:17, Kachru and Nelson 2001:15).
This is not to deny, of course, that some of the most eminent compilers of English
grammars have been non-native speakers.From Jespersento Svartvik, there is a
long line of authoritative, non-native grammarians.While exercising their authority
to codify English, however, their sources have always been the language of native
concerning
EWL interactions
speakers:their job has been to find instancesof what native speakerssay, never to
provide instancesthemselves(Jespersen
1933,Leech & Svartvik 1994).
There is considerable evidence to confirm the authority conferred upon native
speakersas sourcesof what is acceptableEnglish. Ventola and Mauranen examine
the 'native revising' of non-native writing (Ventola and Mauranen 1991) and Connor
enlists the help of native speakers to make her writing more 'appropriate'
(Connor:1999). Many research projects have examined non-native speakers'
'mistakes',or learner English, by measuringthe performanceof these againstnative-
speaker performance or, more contentiously, by directly asking native speakers
which mistakes they felt were tolerable and which not (Hultfors 1987, Norrell 1991,
Lorenz 1998). As with the conflation of native speakerwith Standard English and
the question of native speakerownership of English, the idea that native speakers
have exclusive authority over English language matters has been hotly contested.
Widdowson, for example argues strongly against accepting the authority of native
speakerswhen it comesto regarding

pronouncements StandardEnglish (Widdowson
1994) and Graddol argues that authority will shift from native speakers as the
realisation sets in that they constitute a minority of users of English. (Graddol
1997:3). The weight of evidence at present, however, shows that native speaker
authority is largely unquestionedbeyond the protestationsof applied linguists.
2.1.3 Learner preferences, teacher predilections
Given that ownership of and therefore authority over English is seen as a native
speakerprerogative, it follows that people in what Kachru calls the outer circle and
the expanding circle should look to native speakers, in the inner circle, for their
norms. (Kachru and Nelson 2001). Several pieces of research have shown, for
concerning
EWL interactions
example, that learnersand teachersof English prefer native-speakeraccentsto non-
native speakerones.
Mazzon reports researchcarried out in the 1980s on how acceptablelocal English
accentsmight be to those studying the language.Whereas studentsin India seemed
to accept local models, Thai studentsand studentsin Singaporewere all in favour of
British models, finding the British standard `more pleasant' and `more correct'
(Mazzon 2000:83).
In the 1990s,Chiba et. al. found that Japaneserespondentswere more positive about
native speakeraccentsthan they were about non-native speaker ones, ranking UK
accents as most acceptable, followed by USA. Hong Kong, Sri Lankan and
Malaysian accents were given low ratings (Chiba et. al. 1995). Similar research
conductedin Austria found that non-native speakeraccentswere accordedlow status
and that preferencewas given to native-speakeraccents(Dalton-Puffer et.al. 1997).
Recent research carried out by Hannam suggests that teachers in Greece reject
Greek- accented English in favour of native-speaker accents (Hannam 2005).
This widely felt preference is reflected in a similarly widespread preference for
native-speaker teachers. At the beginning of the 1990s, Quirk opined that native-
speaker teachers of English were required since they are the ones with intuitions
about the language (Quirk 1991b); his opinion is not supported by research done
among staff at a Hong Kong educational institutions, who thought that non-native
speakerteachershad a better command of the grammar than native speakersbut who
nevertheless found native speakers to be superior with respect to communicative
aspectsof English (Norton 1997).

EWL interactions
Research in the USA, in the UK and in the Basque Country has all found that
learners prefer native-speakerteachersto non-native ones (Lippi-Green 1997:124;
McKay 1995;Amin 1999;Timmis 2002, Lasagabasterand Sierra 2002; Abe 2004).
According to the results of a striking piece of research reported by Lippi-Green,
students actual performance seemsto be negatively affected if they perceive their
teacher to be a non-native speakerinstead of a native speaker (Lippi-Green 1997:
127).
2.2 NNS and NS in EWL interactions
2.2.1 Accommodation
The comparative power of the native speakermay have an incisive influence on the
nature of cross-cultural communication. Lesznyäk refers to several studies which
have concluded that native speakerstend to dominate in cross-cultural interactions
and her own study seems to confirm the tendency (Lesznyäk 2004:76 and 229).
Where native speakers are present: non-native speakers may, given a sense of
inferiority, try to accommodateto native speakers.They may fail in their attempts
and eventually lose their confidence.
Accommodation theory was used to some extent in the 1980s and early 1990s to
explain why cross-cultural conversationsinvolving native speakersare likely to be
different from those involving only non-native speakers (Gallois et.al 1988).
According to Janicki, native speakers' attitude will define the behavioural norms of
conversational interaction, thus restricting any negotiation among non-native
(Janicki
speakers. 1985:14-5).
Not only behavioural norms but linguistic norms are also, it seems, set by native
in this type of interaction, again constraining non-native speakers who

speakers
concerning
EWL interactions
to Zuengler
make efforts accommodate. notes the results of Young's 1988 study in
which proficient Chinese speakersof English used plural forms when talking to
native speakersbut used non-marked plurals when talking to other Chinese non-
native speakers, even where these were also highly proficient users of English
(Zuengler 1991:226).
Some non-native speakers fail to accommodate sufficiently to native-speaker
behavioural and linguistic norms, leading to a breakdown in communication, usually
the result of pragmatic failure (Varonis and Gass 1985). Typical examples in the
areaof pronunciation differencesreceived wide recognition in the 1980sthrough the
work of Gumperz (Gumperz 1991) and, in the same period, the phenomenonwas
closely studied by Thomas (Thomas 1983 & 1984).
Communication breakdown also occurs when non-native speakers òver-
accommodate' to native speakers. `Maximal convergence', a term coined by Janicki
to describe non-native speakers' using the full range of native speaker language and
behaviour, is unwelcome in interactions between native and non-native speakers.
According to Janicki, native speakers do not warm to non-native speakers' using all
the idiomatic expressions and strong linguistic indices of their identity. This
rejection is explained by reference to Marton and Preston's personal possession
hypothesis (Janicki 1986: 171. Seealso above, Section 2.1.1). Promodrou highlights
the use of taboo words in `maximum convergence' attempts, confirming that they
àre notoriously difficult for non-native speakers to manage without risking socio-
pragmatic failure' (Promodrou forthcoming a). Promodrou provides a further
example of pragmatic failure involving a non-native speaker attempting wordplay
while interacting with a native speaker. He reports a non-native speaker as follows:
concerning
EWL interactions
Às a nonnative speakerI am not as free as native speakersto use the language
creatively and idiomatically. For instance, yesterday I said something to a
group of teachers and one of them commented 'you can say that again!'
Humorously, I said 'OK, I'll say it again' and repeated myself more
emphatically - embarrassingly,she said, 'no, I actually meant that I agreedwith
you.' The assumptionwas, of course, that the meaning of the idiom had been
lost on me! (Promodrouforthcomongb).
While non-native speakers are attempting to accommodate to their native-speaking
interlocutors in behavioural and linguistic terms, native speakersare, it seems,trying
to reciprocateby accommodatingpsychologically to their interlocutors. When a non-
native speakermishandlesa conversationalritual, any negative reaction on the part
of the native speaker will be modified by what Janicki calls the
`congenial/uncongenial' dimension. Àmusement', àcceptance' and even
àppreciation' are possible attitudes developed by native speakers while
accommodating to non-native speakers (Janicki 1985: 44-45). Dirven and Piitz
reiterate Janicki's point, defining intercultural communicative competence as `not
only the nonnative speaker's competenceto deal with his limited proficiency but
also the native speaker's ability and willingness to accommodatethem (Dirven and
Pütz 1994, cited in Deen 1997:17). Lesznyäk's study provides a couple of examples
of native speakersaccommodatingpsychologically to their non-native interlocutors
(Lesznyäk 2004: 203 and 214).
Deen finds, however, that psychological accommodation goes both ways: in her
study, the non-native speakers did as much clarifying as the native speakers.
Negotiation of meaning through psychological accommodation seemsto be not just
one-sided.

EWL interactions
2.2.2 Foreigner talk
Native speakers'accommodationto non-native speakersis not necessarilylimited to
psychological posturing: there may also be an elementof linguistic accommodation,
in the shapeof `ForeignerTalk'. `Foreigner Talk' was once thought of as a way of
native speakersmaintaining the social distancebetween themselvesand `foreigners'
(Hall 1976:8, cited in Valdmann 1981:43, Ferguson 1975:1-10) or as a marker of
native-speakerdominanceand non-native-speaker
subservience(Clyne 1981:
77) or,
again, as a pedagogicdevice to provide languagelearnerswith comprehensibleinput
(Terrell 1990). Freed, however, suggeststhat `Foreigner Talk' may representnative
speakers' attempts at maintaining conversational flow (Freed 1981), a stance also
taken by Long (Long 1981b). Zuengler, in the same vein, suggeststhat Foreigner
Talk may be one of several strategiesused by native speakersto accommodateto
non-native speakers in order to achieve understanding. She even implies that
maintenanceof social distance gives rise to the opposite of Foreigner Talk, noting
that, where native speakers may feel under threat from non-native speaking
interlocutors, they may well switch away from `Foreigner Talk', the desire to
accommodatebeing subordinatedto the need for asserting or confirming identity. In
a similar way, Zuengler reports a study by Beebe in which non-native speakerswere
observed to decreasetheir `native-like' pronunciation when they felt under threat
(Zuengler 1991:225).
The term `Foreigner Talk' seems to be under-used in more recent research, and
Crystal, writing in 2001, reports a version of it that indeed seemsto move away from
the traditional view, so much so that he refuses the `Foreigner Talk' label. Giving
only anecdotal evidence in the environment of the European Commission in
concerning
EWL interactions
Brussels, Crystal notes that native speakerdiplomats, politicians and civil servants,
all accommodateto their non-native speakerpartners,by using increasingly syllable-
timed rhythm,simplifiedsentence andfewerphonologicalelisionsand

constructions
assimilations (Crystal 2001:

57).
Whether or not `ForeignerTalk' is delivered as part of an accommodationstrategy,it
may still be perceived as a sign of dominance,while a reversion to normal and less
comprehensible speech on the part of native speakers may be, perhaps more
correctly, perceived as social distancing. In either case, a likely result will be the
non-native speaker'ssenseof confidence-loss.
2.2.3 Loss of confidence
While they are attempting to accommodateto their native speakerinterlocutors, non-
native speakersmay be doing so in an atmosphereof loss of confidence.
In terms of conversationalbehaviour, it is possible that non-native speakerswill use
an ìndependentpolitenessstrategy' whereasit is likely that native speakers,where
the term refers to, for example, American, Australian, British, Canadian, Irish or
New Zealander users of English, will use a different, `high involvement politeness
strategy' and therefore appearto take control of the discourse(Scollon and Scollon,
1995:87). Native speakerstherefore appear to have the upper hand, causing non-
native speakersto feel less confident.
Even more radically, it has been suggestedthat dominant partners in a conversation
mayclaim not to understandsubordinateonesasan a priori posturemakingall non-
native efforts
speakers nugatoryand entailingfurther lossof (Wolf
confidence 1959,
cited in Jenkins, 2000b:14; Davies 1991a 118-20).
Chapter 4 89 Hypothesesconcerning
FWf intnrortinne
Anecdotal evidence suggests that non-native speaker feelings of inferiority when
confronted by native speakers are widespread. Thomas reports her experience of this
and how it lead to her stammering, making mistakes and eventually giving up during
college seminars (Thomas 1999). Ventola and Mauranen recall their worries about
whether or no their texts might appear, to a native-speaker, to be 'impolite, clumsy,
stupid or naive' (Ventola and Mauranen 1991:459). Japanese respondents in Abe's
survey, however, report the contrary: they feel more confident speaking with native
speakers than with other non-native speakers from the same geographical region
since, in the latter situation, they have greater fear of loss of face (Abe 2004).
3. EWL `Symmetrical' conversations
One type of `symmetrical' EWL conversation might be where the interaction is
between two native speakersfrom different countries. Toolan suggeststhat users of
`global English', which he characterisesas `the language of international high
flyers' accommodate to each other (rather than one group doing all the
accommodating). He does not provide any hard data but cites an anecdote
concerning an Indian and an Irish person, who `slip into' global English by the
processof mutual accommodation(Toolan 1997:

9).
Little research has been carried out, it seems, on this type of EWL interaction
involving people who have acquired or learned English in early childhood.
Conversations among non-native speakers have attracted rather more interest,
usually under the ELF banner, discussedin Chapter 3. From the ongoing researchin
this area, is may be claimed that while àsymmetrical' interactions between native
speakers and non-native speakers are characterised by (i) the dominance of the
former, (ii) the latters' often failed attempts at accommodation and (iii) the

EWL interactions
production of Foreigner Talk, `symmetrical' conversations among non-native
speakersonly should show the opposite characteristics:equality of all speakersand
successfulaccommodation,including a more sharedform of language.
3,1 Equality among speakers
Janicki's assertion that non-native speakersare restricted in their interactions with
native speakershas already been noted (See above, Section 2.2.1). Conversely, in
ELF settings involving no native speakers,there is opennessand freedom (Janicki
1985:14-15).
Lesznyäk notes that in ELF encounters`the psychological posture of interactantsis
more favourable than it is in other types (i. e. native speakerto non-native speaker)of
intercultural communication', presumably due to `the mutual recognition of
nonnativeness' (Lesznyäk 2002: 189 and 2004: 58). She goes on to take things a good
deal further by suggesting that the basis of ELF meetings is equality of cultures
(Lesznyäk 2004: 235). This may seem a little utopian given that there is very little
equality among cultures in the world and that, as many have noted, the use of
English often carries with it the notion of belonging to an elite (McArthur 1996: 14).
3.2 Accommodation
Given caution regarding equality, it should nevertheless follow that EWL
interactions without native speakerswill be characterisedby more equal and more
successful accommodation than those between native speakers and non-native.
Towards the end of her article arguing for models of ELF, Seidihofer describessuch
accommodationas `mutual' (Seidlhofer 2001: 147).

EWL interactions
3.2.1 The bases of ELF accommodation
It is obvious that "when speakersof more than one country or culture interact (in
English), more than one set of social and cultural assumptionswill be in operation"
(Campbell et.al. 1983:36-7), but, as will be seen,the effects of cultural differencesin
ELF interactions must be seenas constructedon someuniversal foundations.
Tarone suggests,for example,that there is a universal aspectto strategic competence
"used to bridge the gaps between (.... ) two linguistic or sociolinguistic systems".
Although this type of competencemay be universal, the types of strategy used for
bridging purposesmay well be culturally determined(Tarone 1980:422).
Sifakis appears to agree and identifies, in EIL communication among non-native
"the
speakers, ability to processeachother's performanceto account for the needsof
the specific situation and of one another" (Sifakis 2004:240-1)
Meierkord provides psychological underpinnings for these universals by referring to
, two principles' usually governing ELF conversations: the first is that participants
wish to save face and so avoid putting their interlocutors into embarrassing situations
and the second has to do with interactantsreassuring each other of their benevolent
attitude (Meierkord 2000).
3.2.2. Culture transfer
Given that participantsin ELF interactionsfeel relatively free and equal and that
they make use of psychologically-motivated, universal capacities, a point of
contention is the extent to which they draw on their background cultures in order to
fulfil conversationalaims.
On one level, it has been noted that interactants deliberately avoid the use of
culturally bound information which listeners would be unlikely to know (Tarone and
concerning
EWL interactions
Yule 1987:57), although Meierkord observesthat speakerstranslate proverbs from
their home cultures (Meierkord 2002:114).
On the level of culturally-determined interaction strategies,however, the situation is
not so clear. On the one hand, House's ELF conversation data seem to show that
participants in these interactions "do not let their native linguacultural norms come
to the fore" and so fail to make "appropriate use of routine pragmatic phenomena"
(House 1999:80,84). On the other, several people have assumed that "users of
English bring elements of their own culture to their interactions" (Medgyes
1999:188). Both Smith and Davies agreethat when an intercultural or international
group of people interact, different discoursestrategiesor performance strategiesmay
be in play (Smith 1987a:3, Davies 1991:156), and the more proficient people
becomein their use of English, the more they are likely to transfer sociolinguistic or
pragmatic norms from their home culture (Takashi and Beebe 1987).
These assertions have been supported by a small amount of field research. In
Meeuwis' study of interactions in English among Flemish, Tanzanian and Korean
users, there are examples of pragmalinguistic transfer from other languages into
English with Korean speakers, for example, giving prominence to topic over
grammatical subject and often using a word central in a previous utterance as a
backchannelling signal (Meeuwis 1994: 64). Meierkord has also found cultural
in
elements play in cross-cultural conversationswhere she has noted, for example,
diverse forms of a number of routine formulae (Meierkord 2002:114).

EWL interactions
3.2.3. ELF Strategies
Whether or not non-native speakersrefer to their background cultures for strategic
resources in their interactions, there is much agreement on the emergence of
strategieswhich seemto be specific to ELF conversations.
As mentioned in Chapter 1, Firth was one of the first to identify phenomenaas
typical of ELF interactions. In his paper on the discursive accomplishment of
normality, he puts forward the idea of the `let-it-pass' procedure. In his ELF data
(recorded telephone conversations between a Danish export manager and
international clients) he finds interactantscoping with misunderstandingsby letting
them pass as interactionally irrelevant. He also finds that interactants construct
interactional order in the conversationsby managingtopics jointly, distributing turns
in an orderly way and relying on the assumption that utterances are sequentially
linked. It is the `let-it-pass' strategy which seems,however, to be characteristic of
this type of interaction, and not to be imported from a background culture (Firth
1996).
In her 2000 paper, Meierkord identifies in NNS-NNS small talk several strategic
phenomena which she considers to be typical of ELF interaction. These are:
overlapping turns where the speakersdo not see their utterancesas competitive but
rather as collaborative; absenceof extractors (such as Ì'd better be off now) to link
opening and closing phases to the core phase of conversations; frequent and long
pausesboth within and between turns; high occurrenceof cajolers or verbal appeals
for the listener's sympathy; considerableuse of politeness strategiessuch as routine
formulae in opening and closing phasesand backchannelling. Meierkord notes that
the back-channelling behaviour is very similar to what has been observed with
British English native speakers, suggesting perhaps that it is not imported from

EWL interactions
interactants' home cultures but rather learned from native-speaker sources
(Meierkord 2000).
Lesznyäk's book-length study focuses on topic management. She contrasts this
strategicphenomenonin ELF meetingswith the way it is handled in meetingswhere
non-native speakersare in a minority comparedto native speakers.She finds that in
the ELF meetings, "a process is observable in which divergent communicative
behaviours converge towards each other" (Lesznyäk 2004:234). Seemingly
heterogeneous behaviour among participants was actually part of a dynamic
development from chaos to orderliness (ibid 197). By contrast, the EFL meetings
followed a clear pattern from the beginning and did not exhibit any convergent
behaviour.
3.2.4 Community construction
The idea that ELF interaction is characterisedby convergent behaviour chimes with
the idea that participants in ELF conversationsare concerned,whatever the aim of
the conversation, with the construction of a community. According to Hüllen, ELF
speakersactually create speechcommunities as they proceed to interact with each
other. As a consequence,
rules and norms gradually emerge (Hüllen 1982:
86).
In the paper mentioned above, Firth's `let-it-pass' procedure is central to convergent
behaviour which demonstratesthe ability of speakersto àttend and disattend to a
range of anomalies and infelicities in their unfolding interaction.' (Firth 243).

1996:
Firth is careful to note that this ability is basedon local considerations- the need to
achieve interactional goals alongside the transactional goals inherent in business
deals.

EWL interactions
The distinction between local and global goals is taken up by House who notes that
the ELF interactants in her data co-construct a speechcommunity only where the
focus is on à local affair' rather than on the presentationof an àrgumentativepath'
(House 2002b: 260). The samephenomenonseemsto be presentin data provided by
Knapp in his study of interactions among non-native speakers,some of whom were
`near-native speakers'. He finds that behaviour was convergent until the
competitiveness of the situation became paramount, at which point interactants
becamemore divergent: the near-nativespeakersleft the others behind. Knapp goes
on to cite accommodationtheory as an explanation for the changein behaviour: non-
native speakersuse a convergentstyle with each other in informal contexts to `signal
(.... ) that they share common ground, even if it is only for shared incompetencein
the language'. With a shift to more formal and competitive situations perceived
differences among speakers triggers a more divergent style. He sums up by
suggesting that the co-construction of community aspect of ELF interaction may be
restricted to certain types of situation (Knapp 2002: 240-241).
Many other studies stress the cooperative nature of lingua franca communication
without this rider (Schwartz 1980, Yule 1990, Gramkow Andersen 1993, Meeuwis
1994, Varonis and Gass 1985, Meierkord 1996 and 1998, Wagner and Firth 1997)
and Meierkord, in particular, goes further by suggesting that participants in ELF
interactionsare engagedin the construction,through cultural contact,of a lingua
franca culture (Meierkord 2000).

EWL interactions
3.2.5. A shared form of language
While native speakersmay use Foreigner Talk to addressnon-native speakers,who
may in turn get into difficulties in attempting to accommodateto native-speaker
norms, in ELF interactions, all parties might share language forms on a more equal
footing. Phillipson notes, for example, that "in many international fora, competent
speakers of English as a second language are more comprehensible than native
speakers because they can be better at adjusting their language for people from
different cultures and linguistic backgrounds"(Phillipson 2003:167).
A less generous view might have it that ELF interactions are easier because all
participants have an equal and limited range of vocabulary and cultural referencesto
share.Seidlhofer reports a student,writing in an essay,thus:
"... Compared to conversationswith native speakersI felt much more at
easein this group of ELF speakers.About half a year ago, I had spent an
evening together with several native and non-native speakers of English.
In comparison, I contributed much more in this ELF conversation than in
the conversation with the native speakers.While it sometimes happened
that, due to lacking vocabulary or cultural knowledge, I did not understand
some of the native speakers'utterances,this never happenedin the group
of ELF learners..." (Seidlhofer 2002)
A limited rangein grammar,as well as in vocabularyand cultural reference,may
lead to a sharedform of languagewhich is simpler than one of the national forms of
English. Mauranen seemsto think so: she considers that ELF `tends towards some
kind of structural simplification, or generally unmarked features,becauseas a global
language it has an exceptionally rich variety of Lls among its users'
(Mauranen :2003: 515). Yano agrees with her, characterising EIL in three ways. The
concerning
EWL interactions
first characteristic is generalisation: according to Yano, EIL is not `high-blown',
there are parallels between it and the Plain English campaign. The second is
regularisation: Yano notes the existence in EIL of regularised plurals such as
`mouses', `formulas' and `symposiums', etc. Thirdly, Yano identifies the
characteristic of commonality, giving the example of the cancellation of /0/ and of
the growing ubiquity of ìsn't it' as a generaltag (Yano 2004).
4 Symmetrical interactions including non-native
speakers and native speakers; asymmetries among
non-native speakers
Given the tradition of conflating StandardEnglish userswith native speakersand the
attachment
consequent of languageownership and authority to these people, it is not
surprising that interactions in English without native speakersare often considered
to be more successful than those with them: the `trespassers' can form leagues
among themselves without fear of what the `proprietors' may say. Empirical data
seems to suggest that a spirit of creative co-operation and collaboration informs ELF,
with the obvious corollary that when English is used in àsymmetrical' situations,
where native speakers are present, a superior-subordinate, perhaps conflictual
atmosphereobtains.
If empirical data suggestsas much, it is surely in part becausethe data itself has
been collected with divisions, imbalancesand symmetries in mind. On the one hand,
data has been gathered based on conversations between native speakersand non-
native These
speakers. conversationstend to be either one-to-one or in settings
r
EWL interactions
where the non-native speakeris in the minority. An exception to this is to be found
in Knapp's 2002 paper. On the other hand, conversationsproviding ELF data have,
perforce, excluded native speakersand, in Lesznyäk's case,been compared directly
with an EFL conversationwhere, again, the non-native speakeris in a minority.
The question has not been addressedof what may happen when native speakersare
in a genuinely international conversation,on the samefooting, aside from the native-
issue,
speaker-status as the other participants.
It may well be that, in international or cross-cultural settings,the way speakersrelate
to each other in terms of superior-subordinateor group solidarity has to do with
many factors of which native-speakerstatus is only one. It may even be that native-
speaker status is not a factor. It is possible that accommodation, convergent
behaviour and the co-construction of community characterise any kind of EWL
conversationwhere the focus is `local', where there is no competition and no global
to
goal achieve. It is also possible, that native speakersare not the only ones to use
Foreigner Talk: Haegeman suggests that non-native speakers may use it when
addressing other non-native speakers whose level of linguistic competence they
perceive as being lower than their own (Haegeman 2002).
At this point it is worth returning to Firth's 1996 paper which, it will be remembered,
reports findings from telephone conversationsbetween a Danish businessmanand
his international counterparts.Firth does not use a `Foreigner Talk' framework to
account for the way the Danish user of English modifies his speech during the
conversation: in an ELF context, he naturally wants to emphasisethe symmetry of
the interaction. Yet the transcripts of the conversation show that it is the Danish
speaker who is in the dominant position, linguistically: his accommodation to the
concerning
EWL interactions
Indian is on a par with the accommodation, noted by Crystal, and, similarly,
distancedfrom Foreigner Talk (Firth 1996, Crystal 2001).
There may be a case,then, for thinking of interactions between native speakersand
or
non-native speakers, among heterogeneous
groups of people including both native
and non-native speakers, as being symmetrical. At the same time, asymmetries
clearly exist in interactionsinvolving only non-nativespeakerswho do not, of
course,constitute a homogeneous in
whole contradistinction to native speakers.
If instances of convergenceand divergence, symmetry and asymmetry are to be
it
sought, might be more fruitful to consider what Pennycook calls "the connections
betweenEnglish and various forms of culture and knowledge that are far less readily
localizable". Pennycook refers to the "dominance of English in the domains of
popular culture, international academic relations, and other forms of information
transfer" and to the international use of English as a social practice, as a way of self-
positioning within these international domains (Pennycook 1994: 19,33). In this
sense, any asymmetries are more likely to be between people who think of
themselves, or who are thought of, as the òwners' of the domain and those who are
the òutsiders'; this has nothing at all to do with the traditional categories of native
speaker and non-native speaker.
5. Conclusion
This Chapter has attempted to organise predictions concerning spoken EWL into
three groups: EWL seen as interactions between native speakers and non-native
speakers,as interactions among non-native speakersonly and as any international or
cross-cultural interactions.

EWL interactions
By referring to previous research,it has become clear that, in the main, interactions
between native speakersand non-native speakersare likely to be different from ELF
interactions. Given that this researchhas largely centred on interactions involving
non-native speakersin a minority vis-ä-vis native speakers,

or in a setting where they
are clearly relegated to an inferior position, the likelihood becomes a hypothesis
worth testing under circumstanceswhere, on the contrary, there is no clear majority
and where little or nothing in the setting implies native-speakersuperiority.
Interactions among non-native speakers, ELF interactions, have been characterised
by co-operativenessand the co-construction of community. The present research
may further confirm this characterisation and may go further by allowing for the
same levels of comity even where native speakersare present. Interactions among
ELF speakerswho all share a cultural and language background may register the
same,or different levels of comity.
Finally, the third group of predictions seems very fragile, given that little data is
available deriving from situations where the native speaker and non-native speaker
divide has not been an underlying factor. One of the purposesof the presentresearch
is to investigatejust such a possibility.
The following chapterwill presentthe methodsused in order to (a) identify and
select data-providing participants, (b) gather useful spoken data and (c) analysethe
collected data in an appropriateway.

EWL interactions
Chapter 5
Gathering and analysing data
Chapter5 102 Gatheringandanalysingdata

I Introduction
This thesis has so far attempted to establish that the search for àn' International
English, or any other describable, fixed entity, does not form any part of the aim of
the present research which, instead, focuses on how people in intercultural or
international contexts use the English language resources available to them. The
acronym EWL, standing for English as a World Language, has been shown to be a
useful term for this. In the preceding chapter, a brief review of previous research
showed that there has been some attention given, on the one hand, to international
interactions between native speakers and non-native speakers and, on the other, to
interactions involving non-native speakers only. The latter interactions have been
dealt with under the Ènglish as Lingua Franca' heading. The chapter concluded on a
speculativenote, making hypothesesconcerning conversationsin English as a World
Language - among speakersof all kinds, irrespective of whether or not they are
native speakers. Conversations between native speakers and non-native speakers
have been largely thought of as asymmetrical and, therefore, problematic. ELF
conversations have, on the contrary almost, been characterised in terms of co-
operativeness,collaboration and the co-construction of new cultures.
Whereas the central aim of this thesis remains a general one, having to do with a
characterisationof English as a World Language, greater specificity is now possible:
two more detailed questionsseemto presentthemselves,along with a framework for
analysing conversational data. The first question revolves around the notion of
symmetry or asymmetry: will EWL conversations which include native speakers
appear asymmetrical, compared to all other conversations? Will there be evidence of
asymmetry elsewhere? The second, connected question follows from the ELF
referred to in the preceding chapter: are co-operativeness and convergence

research

characteristicsof ELF conversationsonly, or may other conversations,whether EWL
or among national groups be characterisedin the sameway?
Previous research has allowed for these questions to emerge; it also suggests
analytical frameworks within which to study EWL conversations.Given the focus on
the holistic issues of symmetry and co-operation, it appears obvious that
conversationaldata needsto be analysedin an explanatoryway rather then in a flatly
descriptive one. Such an approachis justified at length by Candlin who affirms that
it "rests on the assumption(... ) that participants use language,employ strategiesof
communication, and infer particular meanings often without any conscious
awarenessof how such in

usages, particular social conditions, act to betray sectional
interests, beliefs and values" (Candlin 1987:25). The latter part of this chapter will
present, following Candlin, the analytical methods which have been used to account
for what happens in the EWL conversations central to this research.This will be
preceded,however, by two preliminary sections.
Firstly, descriptions will be given of the methods used in order to identify data-
providing participants and, secondly, there will be an account of the in

ways which
the selected participants were screened in order to find out their perceptions of
themselves and of each other as different language users. This screening was
consideredessentialsince, if symmetriesand asymmetrieswere to emerge,or if a co-
operative spirit was to be observed in conversations, it was thought important to
ascertain the extent to which collaborative or antagonistic attitudes, as well as
perceptions of linguacultural superiority or inferiority, were present before any
interaction took place. A third preliminary section will be devoted to the methods
for
adopted collecting and transcribing the conversationaldata.
Chapter 5 104 Gathering and analysing data

2 Selecting data-producing participants

Previous studies of international or intercultural conversations have used data-
providing subjects in business settings or in university or student-related ones. In
some cases, real business or commercial transactions have been recorded
(e.g.Varonis and Gass 1985a,Firth 1996) while in others, researchershave gone to
specific sites where English has been used internationally, by students or student-
aged people. (Knapp 2002, Meierkord 2000, Lesznyäk 2002,2004). Knapp made
recordings at a European Youth United Nations conference (Knapp 2002), while
Lesznyäk used a similar setting of a European youth forum (Lesznyäk 2002 and
2004) and Meierkord a University Hall of Residence(Meierkord 2000,2004).
Other researchers have turned to their own students and colleagues in order to obtain
data (e. g. Smith 1979, Varonis and Gass 1985a, Tarone and Yule
conversational
1987 and Jenkins 2000). This option was chosen for the present research: a `pool'
of participants was gathered from among students and staff at the University of
Hertfordshire. The prime motivation for this choice was one of convenience:having
in theory, to English speakersfrom sixty different countries constituted an

access,
which could not easily be ignored. At the same time, using the
opportunity
University to provide participants ensured that some of the variables in the
data would be reduced: the participants all had similar levels of

conversation
from backgrounds where tertiary-level education is considered

education, came
had some shared information, knowledge and experience in common.

normal and
They were all, therefore, educatedspeakersof English.
In selecting the pool of participants, there were none of the constraints and
difficulties experiencedby previous work which required participants to

concomitant

be characterisedas native speakersor non-native speakers.That said, care was taken
to ensure that those chosen came from a wide range of language and cultural
backgroundsand that monolingual usersof English as well as multilingual oneswere
included.
In order to make sure that participants were more or less equally proficient in
English a procedure was adopted based on that used by Smith in his research into the
comparative intelligibility of different users of English (Smith 1979): as part of a
questionnaire designed to examine attitudes and perceptions, participants were asked
to compare their command of English to that of other participants (see Appendix D).
Potential participants were excluded, who were rated as being much more or much
less proficient than anyone else in the pool.
Ultimate proof of participants' equality of proficiency was the conversational task
itself: all interactants were able to participate actively in their respective
conversationssave one. The complete pool of participants is given in Appendix A.
3. Exploring language-based attitudes and
perceptions
The previous researchreferred to in Chapter 4 seemsto have taken perceptions of
antagonism as givens in native-speaker/non-native

asymmetry, antipathy or even
speaker interactions. Similarly, ELF researchfocusing on co-operativenesssuggests
that this too is either a given (non-native speakershave a natural bond among them)
from international conversationalsettings in which all participants are equal.

or arises
In order to be clear as to whether theseperceptionsprecedeacts of communication or
arise from them (or, indeed, whether preceding perceptions become firmer or are

overturned in them) it would seem to make sense to attempt to gain a little
preliminary information before proceeding to the gathering of conversationaldata.
In particular, it was considereduseful to find out (a) whether participants thought of
themselvesas native speakersor non-native speakersand what this entailed in terms
of linguacultural attitudes and (b) the extent to which some participants perceived
others as similar to themselves (entailing, presumably, more symmetrical
conversations)and potentially co-operative.
Data concerning participants' attitudes was collected in two separate stages: group
discussions at the first stage were followed, at the second, by a `listening and
questionnaire' procedure.
3.1. Discussions
Group discussionswere chosen as an appropriate method for gauging participants'
language-related perceptions of themselves and attitudes towards others. This
approach was adopted following Hyrkstedt and Kalaja's suggestion that, for the
required information, a discourse analysis approach to examining participants'
talking is a more efficient substitute for more traditional, positivist questionnaires
(Kalaja 1997, Hyrkstedt and Kalaja 1998).
Mixed groups of participants were asked to discuss issues related to the research
topic. The issues had to do with how native speakers can be identified, the
connection between English and speaker identity, the question of who òwns'
English and who, therefore, might or might not be considered an authority over it.
Participants were also asked to situate themselves on the native speaker/non-native
speaker divide. The complete set of discussion questions is given in Appendix B. 35

people from the participant pool were involved in seven discussions. Discussion
groups are given in Appendix E.
The discussions were recorded, roughly transcribed and analysed. Explicitly-
expressed attitudes and perceptions were highlighted as were those emerging in
responseto other utterancesor appearing to be constructed in collaboration with or
in opposition to other participants. The following extract from one of the transcripts
and following analytical notes may serve as an example of the adoptedapproach:
David: Yeah. One main reasonwill be that my English will be like what can I say? I'll
... ...
say polluted with my other languages,yeah. Theirs will be pure English becausemaybe they
only... they were brought up speakingEnglish in their ... native English...
Ian: That's not entirely true becauseyou don't havejust English in this country, you have got
English, Scottish, Welsh and Irish and a lot of people from Ireland..... (Appendix K, D5: 66-
70)
In analysing this interchange, explicit attitude markers `polluted' and `pure' are
noted first: David (who thinks of himself as a non-native speaker), does not perceive
himself to be an authoritative user of English. Subsequently, John's response is
examined as a response to David. It is the act of disagreement, rather than the
content of what John actually says (which does not make much sense) that partly
constitutes the expression of his attitude which seems to be rather defensive. As such,
John seems to refuse the implication that his own English may be unpolluted, that he
may have more authority than David.
The discussionprocedurewas terminated after the seventh discussion since, by then,
general attitude and perception tendencies had begun to emerge. The perceptions and

attitudes of particular individuals were then assessedfollowing a listening and
questionnaireprocedureoutlined below.
3.2 Listening and questionnaires
Following the preliminary discussion procedure, 45 participants were selected from
the pool and placed into ten international/intercultural groups where various
nationalities and cultural backgrounds were represented in each group. In each case,
every effort was made to ensure that no one nationality or cultural background was
represented by more than one participant. Groups were of mixed sexes and, with a
couple of exceptions, consisted of people of roughly the same age. A list of all
international groups is given in Appendix C.
Each selected participant was then interviewed individually according to a set
procedure: (1) he or she was askedto read a text silently, the same text being given
to all participants; (2) he or she had to summarise the text, orally, without looking
back at it and the resulting monologue was audio-recorded;(3) he or she had to give
oral instructions for getting from the University of Hertfordshire to central London,
which was also recorded. In this way, there was a recording of eachparticipant's use
of English in two different sorts of monologue.
When all membersof each group had been recorded in this way, the recordings were
spliced together and each group member was asked to listen to all the recordings of
the other membersof the same group and to use a questionnaireto rate speakersfor
friendliness, reliability and proficiency in English. Participants were not, at this stage,
given any other information about who they were listening to, nor had they, in most
cases, met the people concerned. The questionnaire is given in Appendix D.

This procedure was based on the `Matched Guise' technique and on questionnaire-
based approaches to assessing language-basedattitudes (Giles 1970, Giles and
Powesland 1975,Ryan, Carranzaand Moffie 1977, Kalin and Rayko 1980, Chiba et.
al. 1995, Dalton-Puffer et. al. 1997, Lasagabasterand Sierra 2002, Timmis 2002).
While previous questionnaire-basedresearch has sought to uncover and affirm
attitudes in order to make general, sometimes absolute points, the use of
questionnaires in this project was aimed solely at ascertaining the feasibility or
otherwise of the proposed international groups and, where appropriate, at helping to
analyseconversationaldata by referring to participants' attitudes towards eachother.
Given the limited aims of the questionnaires,as well as the relatively low numbersof
respondentsin each case,there was little need for sophisticatedstatistical apparatus.
The important results concern only the members of each conversation group,
independentlyfrom any other group.
Results of both the discussions and the listening and questionnaires are given in
Chapter 6.
4. `Homogeneous' groups for recording
conversations
The final stage of preparation consisted of putting participants into homogeneous
groups. Using the same participant pool, groups of four to five people (in one case
only three) were formed, who shareda national background. This does not mean, of
that
course, they shared
necessarily a deeplycommoncultural or ethnicbackground,
but does mean that they were brought up, educated or instructed in English within
similar or identical In
systems. some cases,nationality, ethnicity and culture all
seemedto overlap: the members of the American group, for example, all considered
5
Chapter 110 Gathering data
andanalysing
themselvesto be ethnically white Americans (despite claiming to use different labels
for themselves in different circumstances) and the Norwegian group members,
likewise, thought of themselvesas culturally very similar. The same can not be said
of the Nigerian and Indian groups whose members did not necessarilyshareall their
other languagesand came, in some cases,from different faith backgrounds.
Every effort was made to ensurethat at least one member of each `homogeneous'
group was also a member of one of the cross-cultural groups since this would make
comparison more significant. A list of all homogeneousgroups is given in Appendix
F.
5. Data-gathering and transcription methods
5.1. Data-gathering
A pool of participants had now been identified and grouped into both international
and homogeneous groups and a preliminary investigation of attitudes had been
undertaken, which might be used to explain conversational results; actual
conversationaldata could now be gathered.
Several options seemedavailable to do this: Discourse Completion Tasks (DCTs),
Communication Games, Field Notes, Recall Protocols, Recording Naturally
Occurring Authentic Discourse (NOAD), Role-Plays and Simulations.
DCTs and Communication Gameswere eliminated since they appearto be valid only
for the analysis of isolated speech acts and, in any case, tend to elicit symbolic,
than real pragmatic action (Golato 2003: 92, Lesznyak 2004:86). Field Notes
rather
were also eliminated for the reasonsgiven by Yuan, who points up problems with
accuracy, by Lehrer, who notes that many pragmatic discourse markers are not
5
andanalysing
recalled and by Golato, who makes a general point about the dubious quality of
noted material. (Yuan 2001: 285-287, Lehrer 1989: 105, Golato 2003: 95-6) Recall
Protocols seemto suffer from similar disadvantages(Lehrer 1989: 105).
The decision not to use recorded NOAD was taken mainly for reasons of
convenience: given that the intention to

was collect a large amount of data and, in
to compare international and cross-cultural speaking with speaking in

particular,
homogeneousgroups, it was felt that it would not be realistically possible to find
enough opportunities where relevant groups of people would come together and talk
about the same thing. This fits a general criticism of the procedure, reported by
Golato (Golato 2003: 97).
It was eventually decided, therefore, to use a simulation, which in this case was
taken to mean an unreal situation in which participants were not required to play
roles. A compelling reason in favour of using a simulation was that, by dint of
having all groups tackling the same simulation task and therefore, presumably,
having to perform similar speech acts, the likelihood increased of gathering
comparable data across all the recordings (Kasper 2000). At the same time some
setting factors were controllable and, since the focus of interest was interactional
aims, rather than transactional ones, it did not matter that the participants had no real
commitment to the outcome of the simulation.
Having taken, therefore, a lead from Fant who pointed out the utility of simulation in
that it may imitate reality with regard to relevant parameters (in this case
interactional features)care was taken to consider his six warnings (Fant 1992:65).
Warning 1. `subjectsmay not understand their task and may, therefore, abandon
their roles in order to discuss the instructions and the meta-activity or role-play'

The task was discussed with participants before the recordings were begun;
participants gave that

assurances they understoodexactly what to do.
Warning 2. `subjectsmay not befamiliar with the simulated activity in real life and
thus may not act in a typical way'
There was no evidence in the recordings of participants showing lack of familiarity
with the simulated activity.
Warning 3. `subjectscannot identify with their roles'
Participants were not required to play roles. In the simulation they had only to be
themselves,albeit in an unusual situation.
Warning 4. `subjectsdo not take their role seriously due to lack of motivation'
This did happenon occasion,usually when participants seemedto have decided that
they had had enough and wanted to bring things to a close. On one occasion, one
participant seemedto be taking things flippantly, less becauseof lack of motivation
and more becauseof enthusiasmto stimulate interesting debate.
Warning 5. `subjects' behaviour may be influenced by the fact that they are being
observed'
This objection of Fant's, recalling Labov's observer's paradox, would also be true,
of course, with NOAD (Labov, 1972:113).. In this case, the artificiality of the
simulation was complemented by the artificiality of the recording process.
Participants' awareness that they were having an ùnnatural' conversation was
matchedby their awarenessthat they were being recorded.
Warning 6. `time constraints may create an unnatural progression of the activity'
No time constraintswere imposed. The resulting recordings are therefore of varying
lengths. Participants seemedto come to a natural end after around ten minutes in
each case.

The simulation required the participants (who, it will be remembered, were all
students or staff members at the University of Hertfordshire) to imagine that their
university was about to be visited by an international dignitary. The groups were to
discuss how the budget set aside for the event was to be spent. Each participant in
each group of four or five was given a `cue card', outlining the situation and
proposing a way of spending some of the money. Adding all the proposals together,
the total sum was greater than the total budget allowed. Participants were therefore
required to argue their own corner first, trying to achieve the sum proposed for their
suggestion, but then to make concessions so that all agreed on how the overall
budget should be spent. `Cue cards' for the simulation are given in Appendix G.
5.2. Transcription
Given that the aim in this researchis an examination of EWL in terms of interaction,
an approach to transcription was chosen which is in line with those proposed for
Conversation Analysis. Rather than using conventions which might focus on
structural patterning, a transcription of talk as social action was considered to be
more useful, using conventions that capture features of talk which are interactionally
important. The conventions adopted by Meierkord and by Lesznyäk were chosen,
combined with a couple of features from the Jefferson system, refined by Atkinson
and Heritage, on which Lesznyäk's approacheswere largely based (Atkinson and
Heritage 1984, Meierkord 1996,2000; Lesznyäk 2004).
Lesznyäk's conventions for recording turns, inaudible and unclear segments,best
hesitation,
guesses, laughter, cut offs and overlapping speechwere all followed.

Re-starts, pauses, lengthened vowels and comments were not recorded since these
phenomena did not seem relevant to the research questions and every effort was
made to keep things simple.
Unlike in Lesznyäk's work, it was decided to maintain commas and full stops to
show continuation and stopping respectively. Although it was felt that Lesznyäk is
right when she argues that spoken language does not reflect grammar rules
(Lesznyäk 2004:100), thesepunctuation marks make the conversationseasierto read
do
and represent,as they did for Jefferson,the ends of tone groups.
The transcription conventionsare given in Appendix H.
6. Handling the conversation data: word counting and
annotation
Transcription showed that there was a total of about 50,000 words of conversation,
roughly divided into 25,000 words each for international and homogeneous
conversations.In order to achieve a high level of thoroughness,a Corpus Linguistics
approachwas taken for the initial analysis of the data.
A brief lexical analysis was first carried out, using a lemmatised list (Kilgarrif 1997).
Cobb's `Compleat Lexical Tutor' was used for this, providing not only a general
frequency list of words, but also a division of words into four groups: those
belonging to the thousand most frequent, those belonging to the second thousand
most frequent, words in the Academic Wordlist and Òfflist' words. (Cobb 1997 and
http://www. 132.208.224.131/accessedJune 15`h2005).
Subsequently, and following Leech's injunction, the transcripts were annotated
(Leech 1991). Nvivo software was used for annotation purposes since this allowed
for easy retrieval of conversation extracts under each notation heading (Richards
5
andanalysing
2000). Given the parameters of the research questions as refined at the end of
Chapter4 (Chapter 4, Sections4 and 5), annotation was limited in order to focus on
typical patterns connected to contextual factors influencing variability (Biber
et.al. 3).
1998: Annotation headingstherefore reflect the thrust of the analysis, which
concernedparticipants' interactional goals. The choice of headings,along with their
associatedlinguistics tradition is presentedin below, followed by an overview of the
explanatoryproceduresadopted.
6.1 Superordinate annotation headings: Convergence and
Divergence
Communication Accommodation Theory (CAT), referred to briefly in Chapter 4
(Section 2.2.1) has been successfullyused in several researchprojects similar to this
one (e.g. Young 1988, Knapp and Meierkord 2002: 19-20). It was used here as an
overarching analytical framework because of its in-depth approach, viewing
conversationalcontributions as constrainedby participants' perceptionsof eachother
and looking at how utterancesare affected by speakersrelating to each other as the
conversation evolves. In particular, CAT's broad categories of convergence and
divergenceseemto correspondto the issuesof symmetry and asymmetry which were
of prime importance in the investigation (Giles 1973, Giles and Smith 1979,
Lesznyäk 2004:78).
6.2 Convergence: accommodation
Occasionally, instances of convergencein the conversations were coded simply as
àccommodation', overtly referring to CAT. This was done only in the rare instances

where speakersclearly copied eachothers' speechpatterns,in particular their lexical
patterns. Otherwise, convergencewas coded using labels from SpeechAct Theory,
Pragmaticsand Interactional Sociolinguistics.
6.3. Convergence: speech acts
Many conversation turns were coded as pragmatic speech acts. In particular, turns
were annotated when speakers appeared either to be agreeing explicitly with other
speakers or to be making concessions; in both cases, the assumption was that
participants were sacrificing transactional aims for the sake of interactional ones in
an attempt to achieve convergence.
The `Concession' and Èxplicit Agreement' headings are clearly consonant with
Speech Act Theory and Pragmatics, considered together by Eggins and Slade as
`logico-philosophic' methods for the study of spoken language (Eggins and Slade
1997:24). Taken together, they provide an analytical framework which is both
explanatory and allowing of personal input.
6.4 Convergence: discourse markers
Functional labels conventionally given to certain speechacts, were, then, considered
useful in dealing with conversational moves where the speaker's intention is either
explicitly or implicitly stated. Some of thesemay be grouped as Pragmatic Discourse
Markers and various turns in the conversation data were accordingly annotated as
`Hedges', `Downtoners' or `Backchannels'. These terms are used by Carter and
McCarthy, who point out that discourse markers such as these not only indicate
relationships between utterances but that, significantly for this thesis, they also

ìndicate social relations relating to power and formality'. (Carter and McCarthy
2006: Sections 105-112). Carter and McCarthy have a section in their grammar for
`Hedges' which, they explain, are expressionsused `to downtone the assertivenessof
a segment of discourse' (ibid). Expressions in this conversation data were
accordingly annotatedas `hedges' where they seemedto be fulfilling this function,
again, as part of an overall strategy aimed at achieving convergence.The coding
`downtoner' was used for adverbs and adverbial expressions used to downtone
adjectives or adjectival expressions. `Hedging laughter' was also coded in this
context: on many occasions, some participants appearedto be forcing laughter in
order to lighten their utterances:as with hedging expressionssuch as `sort of, `you
know' and Ì mean', laughter was often used to accompany an utterance, so that
convergencemight be achievedwhile the participant in question might still fulfil her
or his transactionalgoal.
Carter and McCarthy's grammar also contains material under the sub-heading
`Discourse marking in responses',which deals with the use of discoursemarkers by
listeners to indicate their involvement with what is being said. Markers such as okay
and right show that the listener is channelling back support for what the speakeris
saying (ibid. 34.16). This is a clear referenceto `backchannelling', first identified by
Fries, which constitutes yet another way in which speakers attempt to achieve
convergence(Fries 1952, cited in McCarthy 2002:51). The label `backchannel' was
used extensively in annotating the conversation data. Laughter was also coded as a
form of backchannelling, following Meierkord's finding that participants in ELF
employ
conversations laughteras a substitutefor verbal back channels(Meierkord
2000: 120). Backchannelling laughter was coded separately from hedging laughter,
which participants used to accompanytheir own turns.

6.5 Convergence: other strategic phenomena
Apart from attempting to achieve convergenceby accommodating to each other's
speech patterns, performing specific speech acts and using specific discourse
markers, including laughter, participants also seemedto make strategic use of jokes,
colloquial and vague language, inclusive `you' and `we', inclusive questions and
collaborative turns: in annotating conversation transcripts for convergent behaviour,
all of thesewere noted.
While the use of jokes to achieveconvergenceamong conversationparticipants does
not seem to have been studied, colloquial language, as a marker of casual style, has
been identified by Joos as paying an interactant a compliment by treating him or her
as an insider (Joos 1967: 23ff). Joos's work pre-datessystematic frameworks for the
analysis of conversation but colloquial language, like jokes, might be usefully
considered as an interactional strategy. It is in these terms that Jucker et al. present
their work on vague language which, they conclude, may serve various social
functions, among which is that of providing à way of establishing a social bond' or,
in CAT terms, convergence(Jucker et. al. 2003: 1766).
An interactional sociolinguistics framework was also used in noting instances of
inclusive `you' and `we' under the general convergenceheading. This follows Tao's
conclusion that generic `you' expressions`serve to indicate high involvement'. (Tao
1998:37). At the same time, inclusive questions were coded as further indication of
in
convergenceattempts; particular, questions such as `Right? ', 'OK? ' and Ìsn't it? '
uttered with a fall-rise tone change were noted, following Brazil et al. 's finding that
this intonation pattern is often an indicator of convergent behaviour (Brazil et al.
1980).

Finally, and following Meierkord's work, collaborative turns were also coded as
showing convergence(Meierkord 2000): there were many casesof overlapping turns
where it was clear that speakerswere supporting each other rather than trying to take
the floor for themselves.
6.6 Divergence
Speechacts were coded as divergent when participants appearedto prioritise their
transactional goals without any concessions to interactional ones. Thus, bald
assertionsand unhedged suggestionswere coded as plain `divergent'. Challenging
questions, often making rhetorical use of negative forms, were given a divergence
category of their own: the pragmatic force of, for example, "Don't you think
that....? ' seemsclearly to be a divergent one.
Certain modal verbs were also noted as indicating divergence, in a parallel to those
used to hedge suggestions in attempts at convergence: while `might', `could' and
`would' could be seenas ways of toning utterancesdown, `must', `have to' and, to a
lesser extent `need' seemed to reinforce transactional goals at the expense of
interactional ones.
Paralleling the coding of downtoning adverbs and adverbial expressionssuch as à
bit' and `quite', intensifying adverbs and adverbial expressions were coded as
divergent: when participants marked their utterances with, for example, `really' or
àbsolutely', divergent aims were attributed.
A similar converse was identified in the prosodic features of the conversation data:
where fall-rise tone changes were coded, in connection with question tags, as
marking convergentbehaviour, rise-fall ones were marked for divergence recognised
by Brazil et. al. as a marker of the speaker'sauthority (Brazil et. al. 1980).

Finally, closure moves were also coded under divergence:when participants seemed
to signal that no further discussionwas to take place on a particular issue, they were
consideredto have put their transactionalgoal above any interactional ones (Burton
1978).
6.7 Conversation annotation: summary
After a brief quantitative lexical analysis,the conversation data was annotatedusing
a Communication Accommodation Theory framework and coding derived from
SpeechAct Theory, Pragmaticsand Interactional Sociolinguistics. Turns were coded
as either convergent or divergent, two broad categories,under which a number of
subcategorieswere used for more specific coding. The complete set of coding labels
is given in Appendix J.
7. Analysis of the conversation data: explanation
A Pragmatics-informed annotation of the conversations permitted an explanatory
commentary which, it was hoped, would serve to answer the research questions: by
examining the construction of convergence and comity, and their opposites, in both
sorts of conversation, similarities and differences were thought likely to emerge, and
EWL might be characterised,with a particular focus on symmetries and asymmetries.
In order to focus on patterns which were characteristic of EWL, other explanations
had of course to be taken into consideration and perhaps even eliminated: instances
for
of symmetryandasymmetry, example,
might haveseveralexplanations,andnot
only that put forward by Meierkord, who concludes that ELF conversation, unlike
conversation between native and non-native speakers, is characterised by a highly

consensualinteractional style. Meierkord draws, for support, on the work of Firth
and of Gramkow Anderson (Meierkord 1996,2000 section 3.2, Firth 1990,1996,
Gramkow Anderson 1993).
The explanatory frameworks provided by Conversation Analysis (CA), Ethnography
of Speaking (ES), Systemic Functional Linguistics (SFL), Interactional
Sociolinguistics(IS) and Critical DiscourseAnalysis (CDA) were all used in an
eclectic attempt to account for participants' behaviour in the conversations.
7.1 Setting/Ends
Setting and Ends, two of the analytical categories envisaged in Ethnography of
Speaking were brought to bear on the data with a view to explaining conversation
patterns (Hymes 10-15).

1986/2005: The Setting category seemed particularly
appropriate since it has been asserted,by Lesznyäk, that this may well determine
whether or not English is being used as a Lingua Franca, therefore entailing co-
operative conversational behaviour (Lesznyäk 2002: 166). All the conversations, it
will be remembered, represented simulations in which participants had to put
forward individual proposals and then come to collective agreement.As a result,
some conversation groups may have made their conversations resemble formal
meetings, providing settings which could account for instances of convergenceor
divergence, perhaps impelled by interactants' perceptions of how meetings should
run.
Similarly, the Ends,or purposeof the simulatedconversationcould perhapsprovide
some participants with a reason for divergent behaviour: in order to play the
simulation game according to the given rules, they clearly needed to fulfil
transactionalgoalsand, behave
therefore, in divergent
ways.
Within the meeting setting, the `Chair' figure has particular significance and the
establishing or emergenceof a Chair figure was a further factor to be considered
when attempting to account for convergent or divergent behaviour. A Chairperson
might well tend to use more convergentturns than an ordinary member of the group,
irrespective of whether or not the conversation is an EWL one or one among more
homogeneouspeers.
7.2 Interpersonal explanations
While the meeting setting and the transactional goals could be used to account for
some of the convergent and divergent behaviour in the conversations,interpersonal
explanationsprovided an analytical framework which was potentially much richer.
In dealing with conversation turns, attention was therefore paid to personal factors
such as age, sex/gender and background culture. While in some instances, these
factors might be considered deterministically, in an Ethnography of Speaking
framework, they might also, following a more SFL approach,be thought of as being
constructedby participants' use of language(Eggins and Slade 1997:

51).
7.2.1 Deterministic approaches
In a crudely deterministic way, the results of the preliminary discussions and
listening and questionnaires were referred to in the process of explaining
conversational behaviour: where participants had perceived of themselvesas native
speakersor non-native speakers,as having or not having ownership of and authority
over English, as being more or less proficient at English, more or less friendly and
5
andanalysing
mature than other participants, due considerationwas taken in accounting for turns in
the conversations.A brief account is given, in Chapter 6, of the discussions and
questionnaireresults in order to clarify later referencesto them.
In an equally essentialistmode, participants' gender and background culture might
be used to explain conversationalbehaviour: as with aforementioned factors, it was
thought important to consider these areas as possible explanations, alongside the
impelling factor constituted by the difference between EWL and English used in a
more homogeneous group.
As far as gender is concerned,some reference was made to Lakoff inasmuch as a
convergent speech style might be accounted for by her findings concerning gender
and power. Lakoff specifically remarks on, for example, tag questions and hedges,
two of the annotatedphenomenain the presentdata, under the convergenceheading;
for Lakoff, these features mark out a more feminine, less powerful style (Lakoff
2003).
Other conversational features might be accounted for by in a similar way by
reference to participants' cultural backgrounds. Scollon and Scollon, referred to
above (Chapter 4, Section 2.2.2) identify culturally-located speech styles: a
`solidarity' style is associatedwith `Western' cultures, while a `deference' style
belongs to `the East' (Scollon and Scollon 1983). `Solidarity and deference' have
also been referred to as ìnvolvement and considerateness' (Tannen 1984,1989,
1990) `connection and autonomy' (Green 1992), àcceptanceand self-determination'
ìnterpersonal and personal interpersonal face' (Janney and Arndt 1992). Scollon
or
Scollon also consider what they call `discoursestrategies', identifying two basic
and
types, corresponding to the binary division between conversational styles. A
`deference' (considerateness)
style complementsan ìnductive' discourse strategy, in

which speakersgive reasonsand explanations before making their main point. This
contrasts with a `deductive' strategy, aligned to a `solidarity' (involvement)
conversational style, in which speakersfirst make their point and then back it up
with explanations.
Where a number of scholars have focused on the difference between Èast' and
`West' stereotypes,less attention has been paid to `North' and `South' approachesto
conversation style. Nevertheless,it was felt important when explaining turns in this
data to take some account, however crudely, of a stereotypical difference which
might inform participants' attitudes towards each other and, therefore, their use of
convergentor divergent behaviour. Many of the contributors to Boetsch and Villain-
Gandossi, for example, highlight and document the `North-South' construction and
the concomitant oppositions of civilisation/barbarism, modernity/archaism,
progress/stasis which might affect turns in international conversations where
participants identify each other as belonging to the `North' or the `South' (Boetsch
and Villain-Gandossi 2001).
Finally, participants might choose to play different cultural cards in an attempt to
assert power over each other. Social structures might be called into account,
therefore, to explain divergent conversational behaviour, where participants are
clearly unequal or where they strive to highlight their inequality to their advantage.
Thus the Interactional Sociolinguistics framework suggested by, for example,
Scollon and Scollon, might be supplantedby a Critical Discourse Analysis one, in
which discursive differences express power relationships (Fairclough 1989,1992,
1995aandb).

7.2.2 Constructionist approaches
The CDA approach to explaining conversation sequencesseems to lead towards a
constructionist approach: while participants' turns might be determined, in an
essentialistway, by their gender and cultural backgrounds,they might also construct
their gender and cultural identity in the process of conversing, assuming or
relinquishing power into the bargain while they assertthemselvesas they wish to be
seen.
Instances of convergent or divergent conversational behaviour were, therefore,
occasionally considered as part of attempts to construct gender or cultural
background, following studies more recent than Lakoff's and Scollon and Scollon's
(e.g. Holmes and Meyerhoff 2003) and following Stokoe's advice to attempt to
understandhow participants `constitute the world, themselves, and other people' in
their talk (Stokoe 2005).
In this way, a return is suggestedfrom Interactional Sociolinguistics and Critical
Discourse Analysis to the earlier-establisheddisciplines of Conversation Analysis
and Systemic Functional Linguistics. The former, with its focus on micro-analysis
and sequencepatterns, preferred and dispreferred second parts in adjacency pairs,
might provide a framework for explaining gender and cultural roles under
construction, where Interactional Sociolinguistics and CDA offer explanations
according to pre-established world views (Garfinkel 1967, Sacks 1974, Goffman
1967,1971,1981). The latter similarly acceptsthat social life `requires the continual
renegotiation of our places within the world' (Eggins and Slade 51).
1997:
In explaining participants' convergent or divergent behaviour by referring to the
construction of cultural identities, the overall processmay neatly be brought back to
the major aim of the the

analysis: characterisation
of EWL: whetherparticipantsin

EWL conversations seek to construct cultural identities at variance with other
participants or in harmony with them must surely be a central concern. One of the
hypotheses examined in Chapter Four suggests that, in the absence of native
speakers, an ELF culture is constructed in international settings. Additional
hypotheses,inclusive of native speakers,might now also be tested and, taking due
account of essentialistdetermining factors, proved or disproved.
8 Three explanatory formats
In order to explain the data as rigorously as possible, three different formats are
given in Chapters 7,8 and 9 respectively. In Chapter 7, two conversations are
analysedin contrast to each other. The first is an international conversationwhile the
second is a homogeneousone. One of the speakersin the international conversation
is also present in the homogeneous one. The two conversations are analysed turn by
turn, with reference to the annotation coding and the explanatory categories given
above. Preliminary conclusions are then drawn.
In Chapter 8, the remaining nine international and eight homogeneousconversations
are considered, firstly by showing quantitative results of the annotation effort:
instancesof convergent and divergent behaviour in all conversations are listed and
tabulated. In the second place, the explanatory framework explored for the two
conversations in Chapter 7 is used to account for behaviour in the remaining
conversations.
Chapter 9 concludes the analytical series by proposing a third format in which six
individual participants are contrastedwith each other in three pairs. The individuals
have been chosen as broadly representative of familiar cultural categories. Each
individual appearedin both an international and a homogeneousconversation: their
conversational in
behaviour eachof theseis logged,tabulatedandexplained.

These three analytical chapters are preceded, it will be remembered, by a brief
overview of the preliminary discussionsand questionnaires.
Chapter S 128 Gathering and analysing data

Chapter Six
Results: preliminary discussions and
questionnaires
Chapter 6 129 Results: preliminary discussions

and questionnaires
1. Introduction
This Chapter reports the results of the preliminary discussions and of the listening
and questionnaireinvestigations.
The discussionswill be dealt with first, and issues arising from them, pertaining to
native speakersand non-native speakersand their ownership of and authority over
English will be highlighted. This first section will close with a view of how the
participants in the discussions viewed interactions between native and non-native
speakers.
The secondpart of the chapterwill look at the perceptionsand attitudes of individual
participants in international conversations.It will examine the results of preliminary
questionnairescompleted by participants while listening to their co-interactants.
2. Discussions
It has been noted that the native speaker construct is central in arguments about
symmetry and asymmetry in EWL interactions (Chapter 4); native speakers are
thought to have ownership of and authority over Standard English, leaving non-
native speakers feeling disempowered in unequal relationships. To an extent, this
explains the exclusion of native speakersfrom ELF researchwhere it has been found
that, freed from feelings of inadequacy and the requirement to accommodate to
native speakers, freed from negative feelings aroused by native speakers' use of
Foreigner Talk, non-native speakers can create and sustain comity and co-
operativenessamong themselves.
In this thesis, the difficulty of identifying native speakers has been signalled, as well
as the possibility that interactions between native speakers and non-native speakers
may not necessarily be characterised by asymmetry. With this in mind, seven
Chapter6 130 Results:preliminarydiscussions

and questionnaires
preliminary discussionswere set up among people, many of whom were later to take
part in the data-providing EWL conversations(Appendix E). As outlined in Chapter
6, the discussionswere prompted by questions (Appendix B), recorded and roughly
transcribed(Appendix K). Someanalysis of thesediscussionsnow follows.
2.1 The native speaker construct
While a couple of self-described native speakers concluded that they could not be
otherwise, given that they were monolingual users of English (D3: 4,7), others
grappled with the notion of native-speakerness by referring, explicitly or implicitly,
to inheritance and nationality on the one hand, and language proficiency on the other.
2.1.1 Inheritance and nationality
Many participants followed the received wisdom that being a native speakerhas to
do with where you were born, who your parents are or were and the fact that you
absorbed English as a child. People describing themselves as non-native speakers
differentiated themselves from native speakersin this way, either by lack of early
childhood acquisition (D4: 1-3,151-2; D5: 68), by lack of inheritance (D3:36-37;
D7:42-44) or by nationality (D1:2-6; D2:4,20; D3: 3,6,15; D4:95; D6: 57-60; D7:
20-22,30). The connection between native speakers and the traditional ENL
countries transpired in many discussions:participants identifying themselvesas non-
native speakerstended to associatenative speaker features with the countries of the
UK, with Canadaand with America (D1: 116-7; 188-201; D4: 79-86; D5:76-77).
Whereas for some people the three ideas (of early childhood acquisition, inheritance
and nationality) were simply interconnected (D2: 4-6,19-20), for others, they were
quite separate: one African participants considered herself a mother tongue user of

and questionnaires
English becauseof the early childhood acquisition criterion (D7: 1-7); her claim was
rejected by co-nationalsbecause,it seems,she failed the inheritance and nationality
criterion (D1: 34-46). The rejection seemedquite vehement, as if the participant in
question had somehow betrayed her own roots by calling herself a mother-tongue
speaker. She did not, however, consider English to be `her' language. A similar
differentiation between native language and owned language was evident in other
participants, one born and brought up in England, with African parents (D3: 28-33),
another,an African bringing up a child in Britain (D1:24-33)
The complexities surrounding the terms, reported in the discussions,reflect those in
the scholarly material reviewed in Chapter3, Section 4.1.1
2.1.2 Proficiency
Participants identifying themselves as non-native speakers gave a range of
explanations for their self-attributed status,which had to do with their proficiency in
English relative to native speakers. Alternatively, in more general terms, those
calling themselves non-native speakers identified particular language traits as
markers of native speakers.
Specific items were mentioned such as accent (D4: 85; D5: 73-81), use of idioms
(D1: 188-201; D4: 35-49) and breadth of vocabulary (D5: 106-111) and in more
general terms, these people thought of themselves,variously, as being less efficient
readers (D4: 29-38) and has having less control and a smaller ability to paraphrase
(D1: 335-337).
At a deeperlevel, self-identifyingnon-nativespeakersconsideredthat they did not
have the intuitions that native speakershave (D4: 51-53; D7: 448-449) and that their
use of English came less naturally (D4: 44-49; D5: 108-11). One self-styled non-

and questionnaires
native speaker disagreed,however, claiming that her use of English was perfectly
natural (D 1:342).
These observations concur in part with the pronouncementsof scholars. It will be
rememberedthat Coppieters actually identified native speakersfor his researchby
attending to their accent and then decided that intuitions provided the key, while
Davies used the `control' criterion as a major way of separatingnative speakersfrom
non-native (Chapter 3, Section4.1.1)
2.2 Ownership and authority
The basis for asymmetry in international conversations involving both sorts of
speakerhas been held as having to do with native speakerownership of and authority
over StandardEnglish. This entails native speakersbeing the models for learners of
English.
The question of ownership, when it came up in the discussions,was, naturally, often
tied to the notion of inheritance. Mention has already been made to participants
referring to `their' languageas the languageof their parents, even where they did not
speak it themselves,or spoke it less well than English (D7: 1-7). A self-defined non-
native speakerconfirmed that she might learn how to speak English naturally by dint
of staying in an English-speaking environment without ever calling English `her'
language(D 1:59-64).
One non-native speakerfelt that, after a lengthy stay in America, English becameher
language (D4: 29-35) but gave the impression that this was a temporary state of
affairs; another participant, already referred to above, considered herself a mother-
tongue user of English while disowning the language almost vehemently (D7: 368-
73).

and questionnaires
Some self-declarednative-speakersclaimed ownership through the inheritance path,
but, at the sametime, wanted to clarify that they did not have sole proprietary rights:
in using the possessiveadjective, they indicated that English had been, as it were,
handeddown to them, but that this did not mean people from other countries were to
be somehowexcluded from rights:
R Can I pick up on this? She said, sorry I am not using your namesbut.. too
many names to learn. You said you don't think of it as 'my'... you don't
say 'my'. Anyone else feel that? So that's the second question. Do you
think of English as your language.Do you say'my' language?
S7 Yes I do.
S6 No becausepeople from other countries still speakthis languageas well so
it's not just our language,it's... everyonecan speakit...
(...........)
R And for you, English is 'my' language.
S7 Well I supposeit's becauseit's what I was brought up with, both parents
coming from London. It's... that is their language, their native language.
R But there are another two people here who come from the same situation
as you I guess,for whom English is... they wouldn't say it's 'my' language.
Sus Well I didn't mean it in that way, I meant, you know obviously it's my
native tongue, it's what I speak but I don't feel that I have ownership of
that language, that other people from other countries can't speak it, can't
use it, you know, I wouldn't be critical if somebody
S7 Oh no, I don't think that's what we meant.
Sus You know, it's a leading question really.
S7 I don't think we meant that it was just ours and nobody else should use it.
It's just the languagethat we speakand therefore it is our...
Sus your nativetongue.

..
D3 21-47

and questionnaires
In Chapters 3 and 4, the evident connection between ownership and authority was
noted to further emphasiseperceptions that interactions between native and non-
native speakersof English are likely to be asymmetrical in nature. The issue of
authority arose fairly frequently in the discussions,particularly where it affected the
question of good or bad languagemodels.
One American native speaker opined that she and her co-nationals, unlike non-native
users, spoke English correctly (D2: 546-8) and another argued, albeit jokingly, that
Americans should have authority over English since they constitute the majority
group of native speakers (D2: 161-71).
One self-styled non-native speaker affirmed that native speakers should assume
responsibility for the language:
Ric I think it's the responsibility of the native speakers. They have the language, as it were.
To us it's kind of handed down. So they have it at its source. And I think it's in the
interests also to proceed to the preservation of the language you know in terms of how
it is spoken....
D1: 355-9
In the main, however, participants on either side of the native speaker/non-native
divide agreed that native speakers did not provide the best models for good
speaker
English. On the one hand, many non-native speakers considered the native speakers
in the participant pool to be dialect users, ineffective in Standard English (D5: 304-
324) and many more paid particular attention to native speakers' pronunciation,
which they considered, again, to be nonstandard (D 1:265-275), while their own
had been perfected by schooling (D3: 201-207; D5: 123-4; D7: 81-84).
pronunciation

and questionnaires
On the other hand, the native speakersin the discussionswere mostly in agreement
that they were not to be countedon as authoritative models. Some said that they had
been taught StandardEnglish at school (D3: 192-197; D5: 346-349) but that they did
not habitually use standardpronunciation and would not normally want to (D3: 212-
9).
Finally, when asked if native speakers could correct their English, non-native
speakers gave mixed replies: while some thought that any native speaker might
correct them (D7: 283-287), others were more choosy (D7: 277-280) and one
consideredthat she could correct most of the native speakersshe came into contact
with (D4: 105-110).
2.3 Asymmetrical interactions
As has been seen, the discussion transcripts show that most participants could
substantiatedifferences between native and non-native but

speakers, that the criteria
for doing so were far from clear-cut. They also show that authority over and
ownership of StandardEnglish was not, in the main, attributable to native speakers.
Nevertheless, perceptions of possible asymmetry in international communication
involving native and non-native speakerswere many and varied.
Some native speakersadmitted to frustration at hearing non-native speakersamong
the participants: their use of English was thought of as òff' or `slow' (D2: 250-254;
545-550) but others said they were tolerant of all users, provided they could be
understood(D2: 552-555).
Some non-native speakersfelt that they were rejected by native speakers:in line with
the `social club' view of native-speakerness, expressed by Singh (Chapter 3, Section

and questionnaires
4.1.1), they felt that there were social barriers in place becauseof their accent (D1:
145-153;D4: 79-86) or becauseof lack of cultural awareness(D7: 202-205).
Others simply felt discomfort in using English, seemingly paralleling the loss of
confidence in the anecdotesreported in Chapter 4, Section 2.2.2. (D6: 361-362; D7:
24-29). The discomfort became, for some, a stronger senseof feeling ridiculed by
native speakers,especially becauseof pronunciation differences (D3: 201-211; D5:
117-124).
Perhapsin reaction to this, and other similar native-speaker attitudes, some native
speakersseemedto stand aloof, considering themselves superior users of English
(DI: 265-275,289-295). One non-native speakerexpressedher senseof superiority
by reporting an experiencein which she felt she had been patronised by a native
speaker(D4: 359-369).
2.4 Summary
For the participants in this research project, there were, then experiences of
asymmetry in international and intercultural communication involving native and
non-native speakers. Reflecting scholarly interest in the area, the participants'
perceptions of what does or does not constitute a native speaker were varied and
sometimes confused while there was general consent that, however they may be
defined, native speakersdo not have much authority over spoken English, inasmuch
as they do not provide good models of the standard version. The perceptions of
asymmetry seemed,in the main, to have more to do with simple social distinctions
and perceived linguistic inadequacies.

and questionnaires
3. Listening and questionnaires

As a further preliminary to gathering conversation data, participants were put into
international groups and then asked, individually, to record two monologues. The
monologues were then played back to other group memberswho had to rate each of
their co-membersin terms of friendliness, responsibility and maturity. They also had
to say if they thought their own level of English was higher or lower than what they
heard from each co-member(See Chapter 6, Section 3.2 and Appendix D). As far as
possible, participants had not met each other at this stage. Results of the process are
given here for each group: only opinions where there was consensus, or near
consensusare reported.
3.1 Group I
Nickname Language. Nationality Sex Age Dis Horn

Stavros Greek Cypriot M 23 X GR
Gauri Hindi Indian F 19 X X
Comfort English Nigerian F 25 D1, D5 NI
Anne German German F 25 X GE
Stavros and Gauri were both accordedhigh levels of maturity and responsibility by
their co-members. Comfort was considered to have a lower language level than all
other group members.
3.2 Group 2
Nickname Language Nationality Sex Ape Dis Hom

Derek Dutch/Flemish Belgian M 21 X X
Ba eh Amharic Ethiopian M 37 X X
Pin Chinese Chinese F 29 X CH
Yang Chinese Chinese M 20 X X

and questionnaires
Derek was consideredto have a high level of maturity and responsibility while Ping
was consideredto have a higher languagelevel than any of her co-members.
3.3 Group 3
Nickname Language Nationality Sex Age Dis Hom

Ahmed Arabic Lebanon M 27 D7 x
Lina Norwegian Norwegian F 25 x NO
David Yoruba Nigerian M 19 D5 x
Ke Chinese Chinese M 21 x x
Lina was consideredby her fellow group membersas having a higher languagelevel
than them. Ke was singled out as seeming of relatively low maturity and
responsibility and as having a lower languagelevel than the other group participants.
3.4 Group 4

Richard English Nigerian M 24 D1 X
Bai Chinese Chinese F 19 CH
Kris Norwegian, Norwegian M 23 x x
S3 Japanese Ja anese F 19 x x
Pallu Hindi Mauritius F 22 x x
Richard and Pallu were considered,by their fellow group members, to have a high
level of maturity, responsibility and friendliness. Bai was thought of as having a
lower level of languageproficiency than the others.
3.5 Group 5
Nickname Language Nationality Sex Ag Dis Hom

e
Ana Cantonese Malaysian F 43 D6
Chat Cantonese ChineseHK M 21 D6
Sus English British F 30 D3 EN
Lei Chinese Chinese F 23 X CH

and questionnaires
Ana, Lei and Susy were all considered to have a high level of maturity and
responsibility; Susy and Ana were also thought of as being very friendly and Susy
was singled out as having a higher level of language than the others. Chat was
thought of as having a lower languagelevel than the others.
3.6 Group 6

Milne English American M _19 D2 AM
Hao Chinese Chinese M 23 x x
Sofia German German F 25 x GE
Javier Spanish+ E. Guinea M 23 x x
Hedda Norwegian Norwegian F 19 x NO
Both Milne and Hao were thought of as having a higher languagelevel than the other
group members.Hedda was consideredto be not very mature and responsiblewhile
Sofia was consideredto be not very friendly.
3.7 Group 7
Anja Russian Russian F 30 x X
Claude French French F 22 D7 X
Greta German German F 53 X GE
Magda Polish Polish F 38 D4 X
Betty English British F 55 X EN
Betty was considered to have a high level of maturity and responsibility. No other
group member was singled out in any special way.
3.8 Group 8
Nickname Language Nationality Sex Ae Dis Hom

Ha1uk Turkish Turkish M 19 X X
Mary French Mauritius F 20 X X
Yon Chinese Chinese M 20 X X
Joseph Ibo Nigerian M 19 DI NI

and questionnaires
In this group, Joseph and Mary were both thought of as having a high level of
responsibility and maturity, while both Mary and Yong were consideredto be very
friendly by their peers.Yong was judged to have a lower level of languagethan the
other membersof the group.
3.9 Group 9
Nickname Language Nationalit Sex Age Dis Hom

Boon Thai Thai M 22 X X
John Czech British M 31 D3 EN
Sammi Farsi Iranian F 24 X X
Angela Greek Greek F 19 X X
Xing Chinese Chinese F 19 X CH
John and Sammi were consideredby their co-group membersto have a high level of
responsibility and maturity and to have a high language level. John was also thought
of as being friendly, a characteristic he was perceived to share with Boon. Angela
was thought of as having a low level of responsibility and maturity, while both Boon
and Xing were consideredto have a low languagelevel.
3.10 Group 10

Shray Hindi Indian M 23 X IN
in Chinese Chinese F 19 X CH
Sacra' Sinhalese SriLankan M 23 X X
Fan Chinese Chinese F 19 D7 CII
Mala Farsi Iranian F 22 X X
In this group, Mala was considered to have a high level of responsibility and
maturity, while Sarraj was thought of as having a high language level.

and questionnaires
3.11 Summary
While the opinions of individual group members will be useful in explaining
upcoming conversation data on a group-by-group basis, it is neverthelessinteresting
to note a few patternsin the questionnaireresponses.
Firstly, there are very few consensualnegative opinions given. This outcome might
support conversationdata which shows tendenciestowards co-operativeness

and
comity.
Secondly, high language status is accorded to a disparate set of participants: Susy
and John are traditional native speakers, while Sarraj would be traditionally
classified as an ESL user and Lina, Ping and Sammi as EFL users. Other native
speakers(Betty, Milne and, possibly, Mary, are not accredited with high language
by
status a consensusor near consensusof peer-group opinion.
Those thought of as having a low language ability are nearly all from Eastern
countries, mainly China, but also Thailand. The exception is Comfort who,
paradoxically, gave English as her predominant language.
4. Conclusion
This chapterhas given some results of the preliminary surveys, intended to help with
the upcoming analysis of the conversationdata.
The first survey seems to show that asymmetries are likely to be present in
international conversationsinvolving both native and non-native speakers,while the
secondsurvey does not reveal any overarching tendencies.It may help, nevertheless,
to account for conversationalbehaviour in conversationsamong particular groups.

and questionnaires
The following three chapters will report the international and homogeneous
conversations,making occasional use of the foregoing results in the analysis and
explanation.

and questionnaires
Chapter Seven
Results: A comparison between an
international and an intranational
conversation
Chapter 7 144 Results: A comparison betweenan inter-

-national and an
1. Introduction
This chapter is concerned with an analysis and comparison of two of the recorded
conversations:one international and one homogeneous.
The EWL conversation is the one among members of group 6. A transcription is
given in Appendix L; the homogeneousconversation is conversation NO, among
Norwegian speakers; a transcript is given in Appendix M. Conversation 6 was
chosen since it included a native speakerand other speakersfrom the `North' as well
as one speakerfrom the `South' and one from the Èast'. Table 1 below gives a little
more detail.

Milne English American m 19 D2 AM
Hao Chinese Chinese m 23 x x
Sofia German German F 25 x GE
Javier Spanish + E. Guinea M 23 x x
Hedda Norwegian Norwegian F 19 x NO
Table 1: Membership of EWL Group 6
Homogeneousconversation NO was chosenbecauseit includes one of the speakers
in Conversation6. Full membershipis given in Table 2 below.
Nickname Language Nationality Sex Age Dis Inter

Birgit Norwegian Norwegian F 23 x x
Gerda Norwegian Norwegian F 24 x x
Hedda Norwegian Norwegian F 19 x 6
Karen Norwegian Norwegian F 22 x x
Lina Norwegian Norwegian F 25 x 3
Table 2: Membership of (homogeneous) Group NO
For analyticaland explanatorypurposes,use will be madeof the annotationlabels
and explanatoryapparatusoutlinedin Chapter5 (a list of annotationlabelsis given

-national and an intranational conversation
in Appendix J); where appropriate,referencewill also be made to any points which
emergedin Chapter 6.
Each conversation is examined in detail and at the end of each analysis an attempt
will be made to summarisethe findings. Finally, comparisonsand contrastswill be
drawn betweenthe two conversations.
2. EWL Conversation 6
The conversation begins between Sofia and Javier. While Sofia seems to want to
establish an maintain an atmosphereof comity, Javier's approachis not so clear-cut.
Sofia's style is characterised by involvement, while Javier begins by distancing
himself:
1 Sof: My idea is to spendthe money for clean and decorateall the areas
2 the VIP will visit.
3 Jav: Yes. How much that going to cost?
The two participants continue with their different approaches.Sofia's persistence
with a more collaborative conversationalstyle is expressedby her use of hedging (4:
Ì think') and vagueness (4: àbout ten, five to ten') and her use of humour,
constructed by the prosody of line 11, which is heavily rhythmic: the rhythm of the
first five syllables is repeatedexactly in the secondfive, there is a rising pitch change
and high key on `me', a falling changeand low key on `you' (11). Later, she starts to
offer a concession(16-19) and explicitly agreeswith Javier's point of view (20-21).
She continues to use laughter (24) in what seems to be a continued attempt at
establishingcomity.
16 Sof @If the area is not too big, that we, then we don't need@... uh... so much
17 money-
18 Jav: Uhm. Uhm.
19 Sof to decorateand clean the places.

20 Jav: Yeah. I mean decorate and clean is important but the security staff is
21 important too.
22 Sof: Yes, it's important..
23 Jav: Becausethey have to feel safe, isn't it?
24 Sof:
Sofia continues in the samevein, deploying several meansby which she presumably
hopes to establish greater comity: hedging (29 `maybe') laughter (30) and an
ìnclusive' question (29-30). This is followed up, a few turns later, with a suggestion
(38-39) clearly marked to show how her argument against spending money on
is
security overlaid by her desire for comity: the suggestion is delivered in question
form, it containstwo negativesand it has a fall-rise pitch changeon the tonic òwn'.
38 Sof: You don't think that the VIP, important person uhm not has security of
39 his own?
Meanwhile, Javier appearsto complement the distancing style of his opening by the
use of an inductive strategy for putting forward his own suggestion.He does this by
laying out his reasonsbefore stating how he wants to spend money and how much
(5-9). His use of Ì think' (8) is, prosodically, clearly not a hedge but rather a
reinforcement of his opinion and his use of `might' suggests genuine uncertainty
rather than studied vagueness.
5 Jav: Uh, I think it's important to... do arrangementfor the security

6 becausethis (is a visit) important. the person, so you needmore
7 security in casethere'san accident...might happen(by the)
8 University. So I think we should spendmore money on the security
9 staff. That might cost around five thousand,so it can combine it
10 between (what you're talking and... -
Despite occasionalsigns of wanting to build a relationship of comity through the use
of inclusive questions (35 ìsn't it? ', 41 `yeah?'), concessions (40) and seeming
linguistic accommodation to Sofia's `security of his own' (29) which emergesas
`security at your own' (41), he neverthelessmaintains his distance by speaking too
7
Chapter 147 Results: aninter-
between
A comparison
quickly for Sofia and appearingto want to close his casewithout further discussion
(55-6), indicating that he has not really respondedto Sofia's interactional openings
and prefers to remain quite distant at this stage.
45 Sof: @Slowly please@

46 Jav: is going to be-
47 Sof I can't understand
48 Jav: Yeah. It's going to be the University who going to get the blame. If the
49 security is not very good
50 Sof: Uh-uh
51 Jav: even if the person haves his own security and he come to visit this
52 University, is University's obligation to keep the person safe. Now I
53 think uh-
54 Sof: OK.
55 Jav: So that (I know I think) That is my idea, so Five thousandis very good
56 price. For the project ten thousandso-
57 Sof: How, how many people do you expect?
58 Jav: Uh How many people?
59 Sof: @@@
60 Jav: I don't know how many people So one very very important person and
61 many of his staff.
Hao enters the conversation with 'OK' (69) to create the space for his turn and
perhaps to establish his right to an opinion. He appearsto use a deductive strategy,
starting by stating how much he wants to spend and going on, at a second stage,to
explain why. His style is clearly an ìnvolvement' one: he goes on to make
approving remarks in responseto others and seemsto want to engagetheir `positive
face'.
Hao clearly establisheshimself as a co-operative interactant by making a joke (80)
and the others respond in a similarly co-operative way by laughing (81). liao
continues in a similar vein, using hedges (83,91 `kind of) vagueness (83 òr
something')anda downtoner
(84 `just') to continuewith his topic.
80 Hao: I think the job is done

81 ALL: @@@
82 Jav: (...... ) Soyou aregoingto have-
83 Hao: My plan was kind of to raise a banquet or something but you spent

84 all the money so it just-

85 Jav: Well that this is our, our idea, yeah, where we think-
86 Hao: No matter how, how securethe security are or as how beautiful the
87 is,
place you still needsto some food or drink. You don't want your
88 just important guestto just stay around and spendall day and thirsty
89 very hungry. You don't want that.
Hao's general style is characterisedby involvement, made explicit by his insistent
use of inclusive `you' and by his responsein high involvement terms, appearingto
approve Javier's position (101 `Yeah, security is important, I think') and then to
establish (male? ) solidarity between them by suggesting that, together, they turn on
Sofia (105).
Javier adopts a more co-operativebehaviour towards Hao than he did with Sofia: he
offers a concession(78), which he reiterates (93-4) and agreesexplicitly with Hao's
proposal. Even when he challengesHao he does so in a co-operative style, typified
by his use of an inclusive question (96 `yeah'). A possible explanation for this
changeis that Javier is now dealing with another man and may feel that he can make
a concession to him, where he could not have done to a woman. It is also worth
noting that Hao was given a high languagerating by the rest of the group, possibly
also accounting for Javier's possible deferenceto him (See Chapter 6 above).
Hedda enters the conversation (111) expressing her desire for convergencethrough
vague language (114,117), by distancing herself from her cue-card, by explicit
agreementand a collaborative turn (121-2).
114 Hed: I don't know how much we are going to spend on the movie because
115 we need a crew.
116 Jav: So how much is it going to cost?
117 Hed: Yeah, it saysabout five thousand.
118 Hao: Five thousand.
119 Jav: That's quite a lot, isn't it?
120 All: @@@
121 Sof: Quite a lot-
122 Hed: For a movie.

Hedda is followed into the conversation by Milne, who, by initially using an
inductive strategy(155-165), perhapsto establishhis authority, seemsrather distant.
The laughter of the other interactants (162) may be a way to try to engagehim, as
may Javier's immediate expressionof agreement(166-167).
155 Mil: I think we need to consider the reason that the very important person is
156 coming, and that's to attend a conference and do a presentation event
157 and that that needs to be I think at the bottom of the budget what we
158 spend. And the conference and presentation, uh, together with the
159 microphones and speakers and everything, and everything that you
160 need to have at something like that is going to cost two thousand
161 dollars, or two thousand pounds, I mean-
162 All: @@@
163 Mil: So we uh, we have to have the presentation event; that's why he's
164 coming and maybe we can cut costs on it, less than two thousand,but,
165 to start, we've got to have that, at the bottom.
166 Jav: Yeah. I think that's a very important thing becausethe visit is for the
167 conference,yeah. Then the-
When he re-enters the discussion (180) Milne seems to assume a position of
authority: he may deem himself to have gained this through his earlier inductive
strategy; he did not share some of his co-nationals' senseof authority vis-ä-vis non-
native speakersduring the preliminary discussion phase (D2) but was thought of as
having a relatively high language level (Chapter 6, Section 3.6). He seems to be
leading a summary of the proposed expenditure (180-192) and Iiao, Javier and
Hedda seem to collude in his self-positioning, or at least to provide the background
atmosphere of agreement: Javier gives several backchannelling signals (181,185,
189) while Hao provides a collaborative turn (183) and Hedda supporting laughter
(190).
180 Mil: We spendtwo thousandon the banquet, right?

181 Jav: Yeah
182 Mil: Two thousandon the presentation.
183 Hao: Two thousandon the presentation
184 Mil: And we haveto havesomesort of security.

185 Jav: Yeah

186 Hao: Yes.
187 Mil: We have to have something.Becauseyou don't want to embarrassthe
188 University.
189 Jav: The University, yeah. So-
190 Hed:
Milne persists in his authority role at a later stage,presenting himself as playing the
pivotal role of the person who will sort out the way the overall sum will be
distributed. `Let's say' (201) seemsto set the tone of conciliation and his question
`What's your bottom line?' (201-2) is suggested, by the negotiation-meeting
framework, as one fitting a no-nonsensenegotiator.
There then follows a sequenceshowing Milne accommodatingto Sofia, with a little
support from Hao (203-210). Sofia does not understandthe expression `bottom line',
leading Milne to repeat it, echoedby Hao in a collaborative turn (204-5), and then to
rephrase it, firstly in a fairly direct way (206 `the least you can do it for') and
immediately afterwards in a more indirect way (206-207 `The least it can be done
for'). Sofia checks her understandingwith a clarification question (208-210) which
Milne confirms (209).
201 Mil Let's say we need decorationsand we need it cleanedup. What's your
202 bottom line?
203 Sof: What's my what?
204 Mil: What is the bottom line. What, what's the
205 Hao: bottom line, yes
206 Mil: le you can do it for? The least
207 it can be done 112
208 Sof: The lowest, uh-
209 Mil: Yeah
210 Sof: price? Four thousand.
211 Mil: Four thousand.
212 All: @@@
At this stage, Sofia reiteratesher proposal in a decidedly divergent manner, picking
up perhaps from her bald rejection of Hedda's suggestion a little earlier (198-200).

She highlights her proposal without recourse to softeners of any kind (214-218),
setting a tone which is taken up by Javier, who replies in an equally divergent way
(219-220) and by Milne, who, setting aside his negotiator role, which he does not
take up again, defendsinsteadhis own proposal, the budget for which he has reduced.
He seemsto want to give an impression of intransigence on his modified proposal
(223 ìt's got to be at least fifteen hundred').Still tending to towardsdivergence,
Javier reiterates his topic (232), and Hao, in a similar vein, decides to challenge
Sofia (233-4) who persistswithout concessionto comity (238 `we have to... ').
Hao provides a counter proposal to Sofia's which he marks with a vagueness
expression (242 òr something') and a colloquialism (242 `quid', 243 `clean this up').
In deploying these two devices,he may be trying to restore a senseof collaboration.
An alternative explanation is that he is assuming a superior posture and using
vaguenessand colloquial languageto belittle Sofia's position.
He seems to be supported in his anti-Sofia stance by Javier, who produced
collaborative turns with Hao a little earlier (233-237) and who now appears to
collaborate with Hao in belittling Sofia, suggesting she reduce her claim from four
thousand to four hundred, which, after protest, he adjusts up to eight hundred. (244-
246).
Sofia laughs at this in a good-natured way (248) and continues to use laughter
throughout the rest of the conversationto soften her otherwise divergent style: while
she seemedto want to achieve comity with Javier at the beginning, she now focuses
on her transactional goal, perhaps in reaction to the combined forces of IIao, Javier
and Milne. The prosodyof `You have two thousand'(261) makesher soundquite
aggressive,as does her turn in responseto Hao's support for money towards film-
making (273-4).

271 Mil: Is it worth making a movie if we only put a thousandpounds into

272 it?
273 Hao: Yeah. Will be. It will be.
274 Sof: No, we said studentscan do it
She rejects the suggestedswitch from film-making to report-writing in a similarly
divergent way, with a direct assertionand emphatic repetition (305-6).
Hedda's responseto the men is more submissive and she seemsapologetic about her
claim, distancing herself again from the instructions on her cue card (269).
The men, therefore, seemto dominate the last part of the discussion.Milne continues
in his assumed air of diplomatic authority, appearing to want to chair the discussion.
His approach seemsdivergent, typified by laconic, unhedged statements(251) and
challenging questions(271-2).
Hao seems equally dominant but continues in his more collaborative style, using
more tentative questions(260). As before, however, his apparent friendliness may be
interpreted as an assertion of his power: his flippant counter-proposal to Hedda's
film-making suggestion (277), followed by his use of vague language ('your stuff
281) may constitute further belittling behaviour as Hao takes up the Chair role
alongside, or displacing Milne.
271 Mil: Is it worth making a movie if we only put a thousand pounds into
272 it?
273 Hao: Yeah. Will be. It will be.
274 Sof: @No, we said studentscan do it@
275 Jav: Yeah I think the studentscan do it.
276 Hed: Then it's (..) it is embarrassing.
277 Hao: Or you do the music.
278 S/H: @@@
279 Hao: The music. The movie. You 've got to (... ) the light.
280 Hed: @I think you ought to do the music@
281 Hao: Ok. Just give me one more thousand. Just forget abou ur stuff.
282 Yeah, it's sweet, sweet all the light, it's need to hire more security
283 personnel. You know what I mean, huh?

Javier seems happy to support his male co-participants, by markedly
backchannelling and explicitly agreeingwith Milne (252,253,255-6) and then with
Hao, when he seemsto have taken over the Chair role. He also makes a concession
to Hao (286-292), something he was loath to do in his original two-way discussion
with Sofia.
The imbalance in relationships continues until almost the end of the conversation
when the participants give the impression of having had enough and so re-introduce
a generally collaborative tone to their proceedingsin order to wind things up quickly
in
and a congenial atmosphere(319-337).
2.1 Summary and further explanation
1-68.
The conversation begins with Sofia trying to establish an involvement style and an
atmosphereof collaboration in which the involvement style is likely to work. She
has little successwith Javier, who seemsto enter the conversation in a different way,
using a different, contrasting style and a different, inductive discourse strategy. The
pair never really achieve comity at this stage.
It may be that the two are drawing on cultural norms in their contrasting approaches.
It may also be that Javier perceives Sofia to be a European,where he is African and
that he refuses, even subconsciously,to `play her game'. It may also be that he sees
her as unfriendly, as the preliminary questionnaire suggested.Whereas Sofia may
seethe interactionas Javier

symmetrical, may take the view that he hasto establish
his authority and his right to speak in an asymmetrical relationship. Finally, his
Chapter7 154 Results:A comparisonbetweenaninter-

response to Sofia may be explained by his need to establish himself as a man,
opposite a woman.
Sofia, on the other hand, seemsto modify her strategy slightly, perhapsin relation to
Javier, having found that her original plan did not work. Perhapsbecauseshe started
the conversation in the first place, or perhaps because she situates herself as
European,interactingwith an African, she seemsto give herself the authority to
control the conversationalstyle. Javier doesnot allow her to develop this, however.
69-108
Hao uses a deductive discoursestrategy and a high involvement style from his first
entry. He does not fit, therefore, the stereotypefor Chinese (Èast') users of English
who are considered to be typically high consideratenessspeakers and to use
inductive discourse strategies (Chapter 5, Section 7.2.1). He constantly gives the
impression of wanting to establish and maintain an atmosphereof collaboration and
comity.
As such, he achieves signal successwith regard to Javier who seemsto co-operate
with him, where he did not give the appearance

of wanting to do so with Sofia. Since
Hao and Javier are culturally very far apart, the best explanation for the solidarity
between them may be that they are both men who therefore take the opportunity
jointly to opposeSofia.
109-154
Hedda'sentry into the conversationestablishesthat she wants to co-operateto the
point almost of self-obliteration. Javier gives the impression of wanting to protect
her, andHao, in challengingher, is carefulto do so gently. The fact that Heddais a

northern European,like Sofia, seems,then, irrelevant. More relevant, perhaps,is the
fact that she presentsherself as a submissivefemale, a role that the two men respond
to in a more receptive way than they do to the more assertive stancetaken by Sofia.
The preliminary questionnairesprovide some correlation here: where Sofia was rated
as ùnfriendly' by Javier and Hao, Hedda was rated as ìmmature'.
155-337
Milne enters the discussionwith what appearsto be an inductive discourse strategy.
Like Hao, he overturns the stereotype:Hao and Milne both behave in ways opposite
to what might be expected of them. Throughout the conversation, he maintains his
distance,using a consideratenessconversational style, perhapstrying to build up his
authority as the last entrant to the discussion. He gives the impression that he may
want to be perceived as an expert negotiator, a role he eventually shareswith Hao.
He doesnot seemat all uncomfortable about sharing this authority role with Hao.
The fact that Hao and Milne end up by becoming the pivots in the discussion tallies
with the questionnaireresults but upsetsthe hypothesis that native speakerswould be
responsible for asymmetries in this type of interaction. Both speakers were rated
highly in terms of their language competence and potential language model status.
On the periphery, Hao and Javier continue in their relationship of solidarity, Hedda
accepts the authority of Hao and Milne, and Sofia seems to hold out in a fairly
divergent style, making concessions to collaboration and acceptance of Hao's and
Milne's authority only through laughter.

3 Conversation among homogeneous, Norwegian

users of English
Hedda begins the discussion, using a deductive strategy and coming straight to the
point. She clearly thinks she has the right, presumably conferred on her by the
meeting format, to state her position and to do so directly, with no softening devices.
She finishes her turn with a direct, challenging question (8). She is behaving in quite
a different way, then, from how she behaved in the EWL conversation. Lina is
equally businesslike, deductive and direct. She does use, however, one downtoning
device (9, àctually'). Hedda challengesLina head on with a rhetorical question (14-
15) and Lina replies with anotherdirect, unsoftenedquestion (16-17).
1 Hedda OK, uhm.... I'm going to start this, I'm going to start this, uh,
2 speechby saying that, uh, we need a good, uh, speakersystem and
3 loud system and visual effects, uhm and also Khofi Annan needs
4 some interpretation service because there's going to be several
5 foreign people in the conference and he needspeople to understand
6 him of course. Uh, we also need headed paper, and appropriate
7 seatingfor everyone.I think this is very important. I think we should
8 spendsomemoney on this. What do you think?
9 Lin I need a film crew actually because, uh... we need to have some
10 future references and if we th, invite a film crew who can film
11 everything during his visit, we can use it as a pr, propos, promotion
12 later on, uhm, for the school but I think it's important for increasing
13 the studentnumber.
14 Hedda But doesn't, don't you thi that it's more important to have uh, a
15 speakersystem?Becausehe. .
16 Lin Yeah but how muc how much money do you need from
,
17 the budget?
The discussion has thus got under way without any apparent attempt by the
interactantsto establish any atmosphereof collaboration or comity between them. As
the two participants in the discussion make rapid calculations and conclude that,
taken together, their two proposals for expenditure are well under the total budget,
Lina wraps things up simply (23 `we're cool') and no laughter accompaniesthe
realisation of potential between

agreement them.

Birgit entersthe discussionin much the samevein as Hedda and Lina. Her strategyis
slightly less deductive in that she gives reasonsbefore going on to her proposal and
finishes with the price she has in mind (24-29) but her manner is similar to the other
interactants' in that she statesdirectly what she wants. She is slightly less direct than
the other two, with a hint that she is trying more than the others to establish an
atmosphereof collaboration. She attempts to tone down her budgetary request (28
`quite cheap', `just no more than') and tries to engagethe others collaboratively by
her use of `you know' (27-28), the prosody of which clearly shows that it is a marker
of desired inclusion.
Karen and Gerda enter the conversationin a similar way to Birgit, leading up to their
budget proposalsthrough three stages:pre-announcement,topic and proposal (34-39,
43-48). Unlike Birgit, however, Karen makes no attempt at all to soften her turn and
Gerda is barely less confrontational. Karen's style is strongly `divergent' from her
presumption that her co-interactants are `forgetting about something' (34) to her
somewhat confrontational use of `you' (rather than a task-focusedor inclusive `we'),
her emphatic use of amplifiers (34 and 36 `really really', 37 `wholly', 39 àt least')
and her use of the modal of obligation `have to' (38). She rounds off her entry with a
challenging `What do you think? ' (40). Hedda's laughter (42) may be a nervous
reaction to this forthright approach.
34 Kar I think you are all forgetting about something really really
35 important which should be a high priority and that's, during this
36 visit, and that's the security. I these, uh, really really terror (...... )
37 times it's wholly important that we really take good care of Khofi
38 Annan when he come here, when he comes here and have to spend
39 quite a lot of money, at least five thousand to buy good uh security
40 around him. What do you think?
41 Ger Uh, I think the most,
42 Hed

Gerda is also direct and forthright in her opinion and proposal using `have to' (44,
46), emphasising òur school' and `really good school' (46) and by repeating `the
most important' (43,49-50). The only concessionto comity comes with the minimal
downtoning of her proposed expenditure(49 `dependingon how extensive the work
is') followed up, nevertheless,by a divergent closure which does not invite any
collaborative comment (49-50). Her style could be thought of, then, as `high
in
considerateness' the sensethat she statesher casewithout seeking approval from
the others.
This somewhat divergent tone to the whole conversation continues for some time,
with Karen taking the lead: she clearly patronises Hedda (`You know that's not
expensive' 66) and the fall-rise prosody of some of her turns indicates her distancing
attitude (51-52,77-80) as does her use of direct `can', rather than `could' (60); she
dismissesGerda's proposal as `ridiculous' (109).
Gerda and Birgit seem to be a little more collaborative, with the former making a
concession (58-9) and the latter expressing agreement with Karen (68-75,98-99), but
Lina and Hedda continue the divergent trend: Lina challenges Hedda with a direct,
unhedged question (88-89) while Hedda laughs down Gerda's proposal (54)
contradicts Karen, using amplifiers to make her point, (63-65) and after a momentary
climbdown (81), reasserts herself in a sequence of gainsaying which leads to an
attempt at closure.
92 Kar Yeah but they're, they're here, they should know

93 English.
94 Hed No they shouldn't.
95 Kar Yes the hould. If they're students at our Uni all, it's all of them
96 have to-
97 Hed No. It's-
98 Kar speakEqgýish in the school.
99 Birg I mean some of this, this univers' there's not a lot of people not
,
100 Hed Yeah but there, but probably will come some important, more
101 importantpeople-

102 Birg Yeah but if, they probably will know English so I don't think
103 that's-
104 Kar Bring their own translators
105 Birg Yeah
106 Kar Yeah.
107 Hed OK Well I said three thousand.That's my final, final-
A more convergent phase now ensues. Karen hedges two turns, making them
suggestionsrather than further assertions(111 `So if you say five thousand', 113 ìf
'
we.. , `maybe'), and Gerda responds in kind by using hedges (112,116 `maybe')
and a downtoner (116,117 `just' ). Karen and Gerda seem to come to an agreement
at the end of this sequence(118-119).
Lina restates her spending proposal but prefaces it with a concession to Karen's
(120-121). There follows a clarification sequencein which all the participants laugh
at their sharedknowledge of the importance of PR (122-125).
111 Kar So if you say five thousandthen.

112 Ger Well I can go down to, maybe, two and a half.
113 Kar And if we, we don't maybe we don't need tie musical
114 entertainment-
115 Ger Because, uhm, just so that we can
116 use some artificial flowers and maybe just uh repair e, just the,
117 the worst that he's go, in the areasthat he's going to visit, yeah.
118 Kar Yeah, the cleaning at the school is good, already so-
119 Ger Yeah
120 Lin But I think the security part is really important and I think that for
121 the school's future we need that film crew becauseif-
122 Kar We needpromotion?
123 Lin Yeah
124 Birg PR
125 Sev PR
Some divergent behaviour still persists, with Karen once again saying Òh that's
ridiculous' (129), Birgit challenging her with `No but listen', starting a clearly
divergent turn (137-141) and Lina delivering an unsoftened emphasisof her proposal
(170-2). Yet the overriding tone is one of comity: Birgit and Hedda collaborate with
other in a high involvement style, characterised by a collaborative turn and

each
Chapter7 160 Results:A comparisonbetweenan inter-

Birgit's accommodation to Hedda's `first impression' (154-155) while Gerda's
suggestion to use music students for entertainment (158-161) is met with
collaborative laughter (162) and explicit agreement (163,168-9). Lina makes
concessionsto comity by prefacing the assertionreferred to earlier with Ànd again'
(170) and by responding to Birgit in a co-operative way, toning down her budgetary
requestby using ìf we say' and implying that her requirementis the obvious
consequenceof a need which all interactantswill recognise, by her use of `so' after
her inclusive `we' (174). Financial calculations take place in an atmosphere of
laughter (177), agreement and concession (178) and further proposals are hedged
(180 Ìf we... ', 184 `could probably' -a correction to `should', 185 Ì think') and
agreedto explicitly (183) or implicitly through a collaborative turn (187-189).
170 Lin And again (...... ) becausethe film group can really show people
171 what studentshave been done before this meeting. So we needthe
172 film crew.
173 Birg But how much? How much can you go down to?
174 Lin We got studentsas well in the, so if we say three thousandthen.
175 Kar How much are we talking about now?
176 Birg Six. Eight and a half. Your own five. That's thirteen and a half.
177 Sev @@@
178 Ger Ok maybe I'll go, down even more.
179 Birg If we spendtwo on the banquet and two on the-
180 Ger If we, if we, if we use more
181 our students to decorate the school and, uh, use more students to
182 entertain-
183 Sev Yeah
184 Ger Then we should, could probably save some money there.
185 Birg And I think we've got a lot of film camerason this school so only
186 thing we need is, and we have video studio, so all thing we
187 needis-
188 Lin professional-
189 Birg professional wo can, really make the movie
190 afterwards.And we have the studios for them so-
This interval of more collaborative conversational behaviour ends with Gerda
confronting Lina in a very direct way. Her question (200) has a rise-fall tone on the
7
Chapter 161 Results:
A comparison aninter-
between
last tonic syllable ('have'), making her turn all the more authoritarian. Lina's answer
is equally confrontational (201) and Birgit joins in with a bald assertion supporting
her own proposal. Hedda offers a counter-proposal,using inclusive `we' and `for us'
(203-4) but her offer is rejected out of hand by Birgit and Karen, Karen being true to
form in her use of the amplifier `real' (207). Karen's counter proposal for the
banquetingarrangementsis articulatedwith more authority than was Hedda's:
instead of the tentative question Àre we thinking about... ' (203), she makes an
assertion,using `can' instead of the more indirect `could' and puts falling intonation
onto her tag question `can't we? ' (209). Birgit also manages to sound authoritative,
with her use of `have to' and the almost verbatim repetition of her point (213),
backedup with an illustration (215) and further repetition (216) justifying her budget
figure.
200 Ger ' You haven't decreasedanything, have you?

201 Lin Yeah I have. I startedoff on three so, I have decreased.
202 Birg Well food is expensive.
203 Hed Are we thinking about having the banquet in Hatfield House?
204 Becausemaybe they can make a special offer, for us.
205 Ger Well maybe that-
206 Birg That would be expensiv
207 Kar That would be, that would be a real expense..
208 Lin Uh-hm.
209 Kar We can use our auditorium or something instead,can't we?
210 Lin Cafeteria.
211 Sev @@@
212 (Confusion)
213 Birg We have to appeargenerous.We have to seemgenerous.
214 Kar Yes.
215 Birg We can't do like, like you getjust a little bit to eat and a little bit to
216 drink. It has to be generousso I think two thousandreally.
Calculations take place during a brief co-operative sequence(216-231) characterised
by a collaborative turn (217-219), a backchannel (227), inclusive `we' (222),
laughter (223) and clarification pairs (224-5,228-229) which is broken by Karen

who leads the group into its last transactional push: time and tiredness take their
effect and Karen's co-participantsseemto force her to stop insisting.
The discussiontails off with a good deal of co-operative exchanges:Birgit expresses
her solidarity with Karen by assuring her that her reduction is excessive(253) and
Gerda explicitly agrees with Birgit that the latter's spending proposal is quite
justified (257-258). There is laughter (260) and more agreement (268) and an
admission on the part of Gerda that she has been hard to negotiate with (271).
Karen's last stand (281 Às long as he brings his own security') is met with
reassurances, not confrontation (282) and more laughter and agreement ensues to
wind the discussionup.
3.1. Summary and further explanation
1-50
All five interactantsenter the discussionwith forthright statementsof their proposals.
While the first two use a more deductive strategy, the other three follow an inductive
pattern of reasoning leading to a proposal. None of the five makes much of an
attempt at establishing or maintaining comity via linguistic means: there is a
noticeable lack of hedging devices, downtoners, inclusive `you' or `we' or modal
forms `could' and `would', which might soften otherwise harsh assertions.Yet the
overall tone of the conversation, coming through the recording more than the
transcript suggests,is not one of unfriendliness. It is perhaps that the interactants
have no need to expressexplicit co-operativeness,given that they share a common
culture and another language.

51-110
The discussion,now establishedwith five participants, continues in a very divergent
way: each interactant seemskeen to establish her status and authority and there is
very little concessionto comity at all. Participants seemto be entirely focused on the
task, with no desire to achieve any interactional goal: the individual transactional
goals are paramount.
111-288
The rest of the discussion moves a little away from the divergent style, with some
evidence of co-operation creeping in. Any collaboration seems, however, to be in
order to achieve transactional goals: interactants agree with each other and
collaborate on turns in attempts, seemingly, to `gang up' on other interactants.
Alternatively, collaborative styles are used, particularly near the end, to bring matters
to a close when participants may be getting weary of the proceedings.
4. Comparison of the two conversations

In most respects, the two conversations are remarkably similar: the interactants
achieve the transactional goal in a similar number of words, make use of a similar
vocabulary, and maintain an atmosphereof friendliness and relaxation. What appears
to be different is the conversational style: while the EWL conversation is generally
characterised by a convergent style, with comparatively little in the way of
divergence, the Norwegian conversation is more balanced, with as much divergent
behaviour as there is convergent.

4.1 Vocabulary
The total number of words in the EWL conversation is 2,264; in the conversation
among Norwegian users of English it is 2,094, around one hundred and fifty fewer
words.
Comparative word frequency charts of the 50 most frequent words in each
conversationare given below, in Table 3. The number of occurrencesis given before
eachword.
Word frequency (EWL Conversation Word frequency (Norwegian

Conversation
93 the 24_important 13_but 85-we 22_that's 14 school
_
70_thousand 24_think 12_clean 74_and 21_important 14 three
66 to 24_for 12_much 61_i 19 he 14 because
_
65 yeah 22_need 12_person 55_that -
19-so _
13_people
_
61 it 21 of 12 what 49 the 19 is 13 students
_ 21 lust 12_decorate -
46 to 18_of 13- much
49 is
48_and 19_uh 12_all 45_yeah 17_well 13 do
_
43_you 19 how 11_then 42_you 17_if 12 then
41 we 17 very 11_in 37_thousand 17_security 12- some
40 so 16_hundred 11_movie 36 think 17 they 12_no
_ i 16it's 11_my 34_ _
16_really _
11_for
38
_ 16- do 11_university _a
33_have 16_five 11 two
36 that
_
30_two 16_or 11 three 27 on 16_it _
11_maybe
15_because 11_on 26_ 16 don't 10be
27five need
26-have 14_don't 10_yes 24 uh _
16this 10_use
14 be 9-will 22_ 15-should _
10_as
25 can
13 9 22_can
but 15 how 10 spend
25 security oin cost
Table 3 Word frequencies in EWL conversation 6 and homogeneous Norwegian conversation
As can be seen, many of the most frequent words in each case are grammatical
words (articles, pronouns, prepositions, conjunctions, auxiliary verbs, modal verbs),
the two discourse marking items `yeah' and ùh' and words tightly connectedto the
theme of the simulation ('thousand', numbers, `security', `clean', `person',
`decorate', `movie', ùniversity', `school', `students', 'spend', `much', `cost'). There
is little noticeable difference except, perhaps, that `we' is more frequent in the
Norwegian conversationthan in the EWL one.
Wordsnot includedin the abovelist are:
EWL conversation: ìmportant', `think' `just' and `very'

Norwegian conversation: `think', ìmportant' `well', `really', `people', `maybe'
Ìmportant' and `think' have similar frequenciesin the two conversationsand `very',
in the EWL conversation is paralleled by `really' in the Norwegian one. `Just' is
considerably less frequent in the Norwegian conversation while `maybe' is the 53`d
most frequent word in the EWL one. General lexical frequency differences between
the two conversationsappear,then to be inconclusive.
Turning to word type, frequency lists of words by category are given in Table 4
below. There are four categories, following Cobb's `Compleat Lexical Tutor' as
outlined in Chapter 5, Section 6. `K1' and `K2' refer to words belonging to the top
and second thousand most frequent respectively. ÀWL' refers to the Academic
Word List.
EWL6 HOM NO
Word Type % of Word Type
No. of words No. of words
total
Kl Words (1 to 1791 Kl Words (1 to
86.27% 1791
1000 1000):
K2 Words (1001 54 1(2 Words (1001
2.60% 54
to 2000: to 2000):
AWL Words AWL Words
48 2.31"%
u 48
academic " (academic):
Off-List Words: 183 8.82% Off-List Words: 183
Table 4 Word frequencies by type
As can be seen,there is, again, very little difference between the two conversations.
The only slight discrepancy appears to be in the number of less frequent words,
which is higher in the Norwegian

homogeneous, conversation than it is in the EWL
one.
4.2 Conversational style
A way in which the two conversations can be said to be different is in the
stylesof
conversational the interactants.

4.2.1. EWL Conversation 6
In the EWL conversation,there is a good deal of implicit negotiation in order to find
a common style, the result of which is a predominantly ìnvolvement' and
collaborative style, where interactantsuse a number of linguistic and paralinguistic
featuressuch as laughter to achievean atmosphereof comity among them.
They explicitly agree with each other very frequently and seem willing to make
They
concessions. make extensiveuse of inclusive questions and often hedge or tone
down their suggestions;where linguistic downtoners or hedges are not used, these
are often substituted by laughter. Finally, they make substantial use of vague
language in order not to impose on each other and give plenty of backchannelling
signals.
Where there is emergentrivalry, between Milne and Hao, for the position of `chief
negotiator', the `loser' in this rivalry does not show any rancour to the `winner'.
Although both contenderswere rated highly, in the preliminary questionnaires, in
terms of their languagelevel, it is neverthelessinteresting that it should be the native
speakerwho climbs down.
The conversation is characterisedby the comparatively small number of `divergent'
turns. There are some `bald', unhedgedturns and some open challengesand several
instancesof insistence, characterisedby the use of the modal `have to'. There are
very few instances of participants' using intensifying adverbs or making their
utterancesmore authoritative through the use of intonation.
4.2.2. Homogeneous conversation among Norwegian speakers
The conversationamong homogeneous,Norwegian users of English doesnot display
any negotiation of styles. All the speakers tend to use a somewhat direct, blunt

by
conversational style which might be called `high considerateness' only the most
sympathetic listener. At the same time, interactants do use almost as many linguistic
and paralinguistic devices as the EWL users in order to achieve comity. Although
they agree with each other and concede slightly less than do their EWL counterparts,
they have as many collaborative turns and make greater use of' inclusive `you' and
`we'. They hedge their turns much less but use downtoners to more or less the same
extent as the EWL participants. They make far less use of vague language, but have
a similar level of hedging laughter and backchannelling.
Unlike the EWL speakers, they make extensive use of divergence markers such as
direct questions and unhedged assertions without downtoners. They tend to use
`direct' modal verbs such as `can' and `will' rather than the less direct `could' and
`would' and make noticeable use of amplifiers, such as `really' to drive home their
points. Fig. 2. below shows the differences: the convergence devices are not
significantly fewer in the Norwegian conversation, while the divergence devices are
markedly more numerous.
. 'I
D EWL 6
0 Norwegian
Convergence Divergence
Fig 2. `Convergence' and `Divergence' markers in F\\'1. Conversation 6 and

Homogeneous Norwegian conversation
Charter 7I (0S Results: A comparison hctwrrn an intcr-

and an intranatiunal conversation
-national
5. Preliminary conclusions
A comparison of these two conversations seems to show that spoken EWL is not
fundamentally different from English spoken among homogeneoususers. The more
divergent nature of the homogeneousconversation may be simply explained by the
fact that the participants sharecommon ground and have less fear of upsetting each
other by being bolder.
The stronger tendency towards negotiation and co-operativenessamong EWL users
is perhapsnot surprising given that participants have less common ground on which
to operatethan do their homogeneouscounterparts.
There is no noticeable strain on the tendency because of the presence of a native
speakerand convergent devices such as vague language, hedges,jokes and humour
are not solely used by the native speaker.There is no evidence of Foreigner Talk but
some evidenceof linguistic accommodation.
The following chapter will look summarily at the other eighteen conversations with
the aim of confirming or modifying the present findings.

-national and an
Chapter Eight
Results: overview of all conversations
Chapter 8 170 Overview of all conversations

1. Introduction
In the previous chapter, one EWL conversation, EWL 6, was compared to one
among a group of more homogeneous speakers, NO. All speakers in the latter
conversation shared Norwegian nationality, a common language other than English
and a very similar cultural background. Using the analytical approaches identified in
Chapter 5, a number of similarities and differences were identified and much of the
conversational behaviour was accounted for using a range of explanatory
frameworks, also reported in Chapter 5, and given as results of preliminary surveys
in Chapter 6.
It was found, comparing the two conversations, that participants reached their
transactional goal in an atmospherewhich createdand sustainedcomity and where it
might be said that interactional goals of further comity were also achieved. While the
linguacultural and social frameworks were different, participants in the two
conversations used a similar range of vocabulary and similar overall discourse
strategies. The difference between the two conversations lay in the area of
conversational style: whereas both conversations contained a good deal of
convergence marking (explicit agreement, concession, hedged or toned-down
suggestions, backchannelling signals, collaborative turns, etc.) the homogeneous
conversation also contained more in the way of divergence marking (challenging
questions,unhedgedcommentswithout downtoners, dominance, ).

etc. Participants in
the EWL conversation, perhaps obviously, made more efforts to achieve
interactional goals than did those in the homogeneousconversation.
In examining the conversations, an attempt was made to explain the differences
between them by referring to the influence of the meeting setting and to speakers'
background, including their cultural background, their attitude towards English, their

sex/gender and any individual characteristics which may have emerged from the
preliminary surveys.
The framework suggestedby the outcomesof Chapter 7 will be used to deal, in less
detail, with the remaining conversations.This chapter will not, therefore, examine
eighteen conversationsin the same way as the previous analysis; rather it will adopt
a more quantitative approach, taking all the EWL conversations together and
comparing them to the homogeneousones. There will firstly be a brief examination
of participants' lexical choices which will be followed by an equally brief
consideration of discourse strategies. At a subsequent stage, the question of
conversational style will be addressed,

with particular focus on the achievement(or
otherwise) of interactional goals. The chapter will conclude with an attempt at
connecting features of conversationalstyle to perceptions of the meeting setting and
to the participants' cultural and individual backgrounds.
EWL conversation transcripts are given in Appendix L; homogeneous conversation
transcripts are in Appendix M.
2. Vocabulary
The two conversations in Chapter 7 were examined in order to establish any
differences in the range and frequency of vocabulary: very little difference was
found. The only figures worth noting referred to the slightly higher use of less
frequent words in the homogeneousconversation,comparedto the EWL one.
The similarity in percentageof high frequency (KI) words and academic (AWL)
words appearsto be repeatedacross all the conversations.Using again Cobb's word
profile the
calculation, of
percentage wordsbelongingto the most frequent
thousand

(K! ) is between 78 and 88 in all conversations,with the averagevery slightly lower
in the homogeneousconversationsand the percentagesof academic words (AWL)
are between 1.9 and 2.8 for the EWL conversations and 1.4 and 2.7 for the
homogeneousconversations,with the two average percentagesvery close to each
other. The higher use of slightly less frequent words in the NO conversation
compared to the EWL one, reported in Chapter 7, is repeated across the remaining
conversations:the percentageof `K2' words is slightly higher (between 2.0 and 5.4%)
in the homogeneousconversations than in the EWL ones, where the percentages
range from 2.1 to 3.9. SeeTables 5-7.
EWL 1 2 3 4 5 6 7 8 9 10
EWL
Kl 83 80.5 82.6 88.4 86.4 86.3 84.6 89.2 86 87.9
Words
K2 3.9 3.6 2.4 2.8 2.5 2.6 3.2 2.1 2.6 2.5
Words
AWL 2.8 2.8 1.2 2.2 2 2.3 2.7 2 1.9 1.2
Words
Off- 10.3 13 13.8 6.7 9.1 8.8 9.5 6.6 9.4 8.3
List
Table S Vocabulary profiles in L WL conversations: percentage of word types
EWL AM EN GR NI NO CII GER SP IN

EWL
Kl 78 87.7 86.9 79.2 86.7 76.6 81.9 91.2 84.4
Words
K2 4.6 3.1 3.8 5.4 3.1 4.1 3.6 2.0 2.6
Words
AWL 1.7 2.4 1.6 1.4 2.2 2.1 2.7 1.7 2.3
Words
Off- 15.6 6.8 7.7 14 7.9 17.3 11.8 5.1 10.6
List
Table 6 Vocabulary profiles in homogeneousconversations
EWL ii011
KI 85.5 83.6
Words
K2 2.8 3.6
Words
AWL 2.1 2.0
Words
Off- 9.5 10.7
List
Table 7 Comparison of vocabulary profiles in the
two sorts of conversation (averages)

The differences in percentages are most easily explained by a commonsense
reference to greater shared knowledge in the homogeneous conversations: where
participants have more in common, they are freer to use slightly less frequent
vocabulary items with each other, there being less risk of misunderstanding. A
further explanation might be that greatercommonality enablesengagementin greater
specificity which correlateswith use of lower frequency words.
The greater frequency of off-list words in the homogeneousconversationsmight also
be explained by the freedom to use more false starts and hesitations, leading to more
`words' unrecognisedby the software.
3. Discourse strategies
There is evidence of both inductive and deductive discourse strategiesacross all the
conversations and there seems to be little or nothing in the way of a discernable
pattern. In EWL Conversations 1,2,5,7,8 and 10, there is near symmetry among
the participants, some of whom use deductive strategieswhile more or less the same
number use inductive ones. The situation is slightly different in the remaining EWL
conversations,with a predominanceof deductive strategiesin all of them.
Three of the homogeneousconversations(AM, IN, NO) are more or less balanced,
with as many participants using a deductive strategy as those using an inductive one.
The other conversations are weighted one way or the other: the Of and SP
conversations are characterisedby deductive strategies, while the EN, GE and NI
show evidence of inductive processes.Table 6 shows clear instances of different
strategyuse. Figures in brackets show line numbers in the relevant conversations.

Participants using deductive strategy Participants min g Inductive strategy

ERWL Stavros (4-13); Comfort (1-32) Cauri (34-47)
E WL Ba eh (61-74) Derek (45-_52,89-129),Van (89-129)
E 3 Lina (1-4); Ke 35-46
EVIL 4 Pallu (6-8); Kris (40-50)
EWL 5 Lei (67-72) Ana 13-181
EWL 7 Anja (2-7); Greta 8-14 Claude (23,11). Betty (32-50)
EWL 8 Mary 1-4 ; Iialuk (42-46) Joseph (33-39); Yong (159-164)
EWL 9 Boon (1-3); Xing (4-7); Angela (12- John (126-130)
14); Sammi (15-19)
EWL 10 Qing (7-12); Mala (14-19) Sarra' (20-28)
AM N111ne(96-102) Sindv (35-41)

CH Ping (1-5)
EN John 93-101
GE Anne (19-22)
GR
IN Pravin (10-14) Shray 17-20
NI David (51-60)
NO Hedda (1-6); Lina 7-10 Birgit (20-24)
SP biontse (10-16); Ines (42-44); Rosario (1-6)
Consuelo 45-48
Table 8 Lºscourse strategies in an conversations
4. Conversational style
4.1. Achievement of interactional goals
Participants in all the conversations, quite naturally, achieve interactional goals
which they were not specifically requestedto try for and there is a general feeling of
This feeling is, however, noticeably stronger in the EWL

comity and convergence.
conversationsthan in the homogeneousones.
The normalised (Biber et.al. 1998:32-3) average number of convergence markers
(Chapter 5, Section 6 and Appendix J) is given in Table 9 below.
CONVERGENCE MARKERS EWL IIOMOG

Total all conversations 877 550
Average 87.7 61
_High 171 (EWL 3) 96 (AM)
Low 40 (EWL 10) 33 (GR
Div. by divergence markers - average 5.8 1.3
Table 9 Convergence Markers in EWL and non-EWL conversations

As can be seen, the average number of convergence markers in an EWL
conversation is 87, with a high of 171, in EWL 3 and a low of 40 in EWL 10. The
same calculations for the homogeneousconversationsresult in an averagenumber of
convergencemarkers per conversation of 55, with a high of 96 in AM and a low of
33 in GR.
This already produces a picture in which convergence is stronger in EWL
conversationsthan it is in homogeneousones. When divergence markers are factored
in, the result is even clearer: by dividing the number of convergencemarkers by the
number of divergence markers for each EWL conversation, an average figure of 5.8
emerges with a high of 19.2 in EWL 2 and a low of 2.0 in EWL 10. The same
operation for the homogeneousconversationsgives an averageof 1.3, with a high of
2.1 in SP and a low of 0.9 in NI. These figures are considerably lower than the
equivalent ones for the EWL conversations.
4.1.1. Laughter and humour
The most obvious sign of a friendly atmosphereis laughter. It was noted in Chapter
7 how the participants' laughter in EWL 6 helped maintain an atmosphereof comity
and all but one of the remaining EWL conversations show the same tendency:
roughly speaking, there are between ten and twenty bursts of collective
backchannelling laughter in eight of the nine remaining EWL conversations.EWL 8
is the exception, with no laughter of any kind.
The laughter is often provoked by the use of jokes and humorous comments which,
to the overall feeling of comity. Three of the participants in

naturally, contribute
EWL 1 make a joke or a humorous comment:

I Sta OK This guy is coming to the University, he is like really well known, he is, uh, Prime
2 Minister of (... ) [Phonerings] This should be him calling now.
3 All @@@
WL 1: 2. Seealso EWL 1: 130,171-3)
three of thoseparticipating in EWL 7 make betweenthem eleven funny statements:
88 Bet He brings his own security, surely. I mean h' not going to come,just sort of step off
89 the 602 bus, is he? He's going to-
90 Anj @Yeah @
(EWL 7: 88-89. Seealso EWL 7: 151-155,240,271,7 308-9,329,338,376-9,386-7,390
There are also humorous turns in conversationsfour, five and nine:
216 Sus I don't think we can cut corners on the budget of the party though. I think that it's got
217 to be lavish, it's got to be really expensiveand impressive. I don't think we should go
218 to Tesco's for the food.
219 All @@@.
(EWL 5: 217-8. Seealso EWL 4: 195-6;; EWL 9: 162-3
Collective laughter is also a feature in the homogeneousconversations,where it is
more frequent than in the EWL ones. All save one of the homogeneous
conversations include at least some backchannelling laughter and some of the
conversationsare constantly punctuated in this way (e.g. AM, EN, SP and NO). As
with the EWL conversations,jokes and humour are unevenly distributed across the
homogeneousconversations: the EN conversation has eight instances (EN: 74,98,
265,318,418,424,454,518) which, even given the extended length of this
conversation, is a high number while the IN and NO conversationshave no jokes or
humorous commentsat all.
4.1.2. Colloquial language and vague language
As well as provoking laughter by the use of jokes and humour, some participants do
like Hao in EWL 6 and use colloquial language and vague language in order to
appear friendlier and to encouragean atmosphereof light-heartedness,propitious to
the achievement of interactional goals. The word `quid', used by Iiao in EWL 6
Chapter8 177 Overviewof all conversations

(Chapter 7, Section 2) is also used in EWL 7 (246) and the word `guy' is used to
refer to the VIP around whom the meetings are centred, (EWL 1:1), to address
fellow participants (EWL 1:125, EWL 4:68) and to indicate the University's security
officers (EWL 7:294).
`Guy' is fairly common in the homogeneous conversations, too, being used
extensively by Sukvinder (IN), and variously in the other conversations referring,
again, to the VIP (AM, NI, EN) the security staff (AM, GE) and each other (IN, AM,
NI). Other than `guy' there is little in the way of colloquial language in the
homogeneous conversations, with the exception of NI, where a few expressions
occur (NI: 2, ìn town', 60,102 `grand', 83 òn deck', 135 `dodgy').
Vague expressions abound in all the EWL Conversations and in many cases it seems
that they are being used in order to make the speaker appear more down-
quite clear
to-earth, less formal and therefore more inclined to constructing a friendly and co-
operative environment (Chapter 5, Section 6.5) (EWL 1: 59 "we can actually do all
this stuff', 161-2 "repairing and the whole image thing"; EWL 2: 32 "some paintings
or whatever"; EWL 3: 206 "I mean the video coverage and all that", 231-2 "the
lights and the microphones and, and stuff'; EWL 4: 36 "the technology and all this
stuff'; EWL 5: 233 "the respect and things"; EWL 7: 194 "profe, visiting academics
EWL 8: 145-6 "pick up litters, those things like that"; EWL 9: 140-1
or whatever";
"buildings, or flowers or trees, something like this"; EWL 10:309 "not for food and
those things").
In other cases,of course, vague language indicates nothing other than a speaker's
hesitation at not finding the word she or he really wants. This seemsto be clearest
when the vague language is accompaniedby a good deal of hesitation, as in EWL 1:
92 "If we could get a goodsetof, uh, microphonesand,uh, consoleandeverything",

EWL 2:91-92 "we should pre, uh, prepare for some, machines like, uh, uh, mi,
microphones, sp, speak, speaker systems or interpreter systems or-", EWL 5:262-3
"Really I think University itself has, be they got this kind of, uh, this, this, this what
they call it? the (..... ) of the, art and design?" and EWL 10: 97-8 "media will be
telecasted either live or later, some other, uh, it will be, programme will be
telecasted".
The use of vague language to help construct an informal, therefore friendly
is
atmosphere, not so abundantin the homogeneousconversationstaken overall. CH
and GE have no such use at all, while GR, IN, SP and NO have only one or two
examples (GR: 95 "We should set up some cameras. Some alarms. That kind of
"
thing. ; IN: 12 "the university looks very good, everything is furnished, well-
decoratedand everything", 43 "When somebody comes and everything is dirty and
everything" ; SP: 15 "You know like a record or something like that", 98 "Or some
of the labs. Some, some kind of labs.

"; NO: 169 "spending money on repairs or
",
something. 209 "We can use our auditorium or something instead"). NI and EN
have a few more examplesof `friendly' vagueness(NI: 3 ". We want like the latest
gadgets in town, you know, technologywise and everything", 22-3 "this thing for
them to come on the film and all that" , 27 "Drinks and all that", 44 "you can just
have like a radio playing or something", 114 "the food and all that"; EN: 150-51
"awards and whatever", 190 "BBC or anything", 200 "Some kind of nourishment",
273 "some kind of compromise", 287 "flowers and stuff') while participants in AM
make substantialuse of this strategy (AM: 59 "the parking lots and things like that",
68 "the receptionand that kind of thing", 86 "some flowers, you know, stuff like
that", 92 "the banquet,thingslike that", 103"Stuff like this will", 221 "everything's

clean and everything", 310 "a punch or something", 328 "the cost of flowers or
something", 360 "about two thousandor whatever").
Vague languageis also used to tone down suggestionsand will be dealt with in the
appropriatesection (4.1.6) below.
4.1.3. 'We', 'Us' and 'You'
A senseof co-operativenessis also maintained in all the EWL conversations,as it is
in the example EWL analysedin Chapter 7, by the persistentuse of `we' and ùs' or
of the ìnclusive you'. The many instancesof `we' and `you' can often be explained
by the simple fact that speakersare discussing collective action, but there are plenty
of turns where it is clear that `we' is intended to signal explicit comity or an explicit
call to co-operativeness.
74 Lin Oh well, I, I, I, you and me, we are sort of like, we have sort
75 of the sameidea, I think. So I could, I could probably agreewith you. I, I, I mean I
76 could, I could cut my budget on food and drinks a bit and we could make like snacks
77 or-
(EWL 3: 74-7. Seealso EWL 2: 165-8, EWL 5: 73, EWL 7: 54-6
The homogeneous conversations also make use of `you' and `we' to signal
inclusiveness.
255 Sin Well, with food and everything like that's something that, I mean, we all know how
256 expensivefood is here, especially like nice food. I mean this guy is not going to
(AM: 255-6. Seealso CII: 21, EN: 303,398, GR: 127; IN: 92; NI: 88-9,122,176; NO: 149,
176,250, SP: 39
The distribution of `we' and `you' acrossthe EWL conversationsshows, possibly, a
pattern not present in the homogeneousconversations. After normalisation, there
seems to be a relatively high use of `we' and a lower use of `you' in the EWL
conversations,while no such clustering occurs in the homogeneousones. Tables 10
11
and give instancesgroupedby frequency.

Times 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100+
used
WE 8 6 10 2 1 4 3
5 7
9
YOU 1 4 6 7 5
2 9 8 10
3
Table 10 Use of `We' and *You' In EWL conversations
Times 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100+
used
WE SP CH NI AM EN NO
GE IN GR
YOU CH GR GE EN IN SP NI
NO AM
Table 11 Use of We' and `You' In homogeneousconversations
It may be that, following the idea that the EWL conversationsare more convergent,
the greater use of `we' in them emphasisesco-operation and the lesser use of `you'
correspondsto situations of minimal personal confrontation.
4.1.4. Accommodation, collaborative turns and backchannels
As with EWL 6, examined in detail in the previous chapter, much evidence of
comity in all conversationscomes not in individual turns, but acrossexchanges,with
participants accommodating to each other, speaking collaboratively and using
backchannelsignals while anotherinteractant is speaking.
Speakersappear to accommodatelinguistically to each other in many of the EWL
conversations, as Javier accommodatedto Sofia in EWL 6 (Chapter 7, Section 2).
This mostly takes the form of one participant echoing a word or words of a previous
speaker. In EWL 3: 241-248, for example, Ke uses the word `volunteers' shortly
Ahmed has introduced it, while in EWL 4: 17-9, Pallu echoes the word
after
`perfect' previouslyutteredby Richard.The fact that it

sheprefaces with Èxactly'
to
seems emphasisethat she wants to use Richard's word in order to show how
attuned she is to him. In the same conversation, Richard deliberately echoes Bai's

use of the word `volunteer' and `student union' in an exchange that is therefore
heavily marked as showing comity betweenthe speakers(EWL 4: 140-144).
In EWL 5, there is, apparently, a more subtle instance of accommodation: Ana uses
the word ìmage' twice in two sequential turns, collocating it with `promote' and
then with `put'. Chat, in the following turn, appearsto want to accommodateto Ana
in that he uses the word ìmage' but collocates it with `give', thereby showing
perhapsthat this is not a word he usesvery often (EWL 5: 137-142).
This type of èchoing' accommodation is also present in the homogeneous
conversations:in AM, for example, Sindy hears Dolores say à good opportunity to
come in' in
and, a move which seemsto be a backchannel, accommodatesto her by
echoing `Coming in' (46-47); Dolores a little later accommodatesto Candice (AM:
94-95). In EN, Betty echoes Susy's `they're learning' (180-181) and Susy echoes
Betty's `He's an intelligent man' (372-5) and in CH, Bai echoesFan (84-5). There is
of
some evidence echoing accommodationalso in GE (85-86,114-5) GR (118-9), SP
(55-6) and extensively in NI:
8 Shi Well I think the food is very important, otherwise be hungry

9 Com Uh... yeah food is very important but
I 8-9. Seealso 11-12,32-33,67-71,117-119,145-146,169-170,175-176,193-194).
EN, alone among the homogeneousconversations,has examples of accommodation
which are not echoing, where one speaker seems to use another speaker's
them back instantly. The word `showcase' for

expressions without repeating
is introduced by Betty (33) and shortly after taken up by Santosh (40); a

example,
little later, Betty usesthe word èsteemed'to describe the university (58) and, again,
Santoshusesit in a later turn (77).
At times, accommodationappearsto take place during collaborative turns and, in any
often overlap with each other in a supportive way. The collaborative

case, speakers

turns identified in EWL 6 in the previous chapter are readily paralleled by many
similar instancesin the other EWL conversations:
195 Cla Well you could ' agine a morning there, followed by a, maybe banquetand then the,
196 visit.
197 Bet Yes. OK. omething fairly scaleddown. Of coursehe will need to give a
198 speech.I mean it would be the main purpose-
199 Cla Yeah he will needto give a speech,the VC will need to give asp ech and-
200 Bet Yeah
201 Cla can think you can have a couple of, uh, gueststhere-
202 Bet Hmm.
(EWL 7: 195-202. Seealso e.g. EWL 1:71-74,155-9; EWL 5: 156-161; EWL 7: 215-224;
EWL 8: 154-7; EWL 10: 103-113
The homogeneous conversations also contain many collaborative turns, with
participants supplying words for each other or supporting each other by
simultaneouslysupplying a synonymousexpression:
171 Lui And the presentation,yeah. Becausemine's alr, all, I think they often sort of-
172 Sof Overlap Yeah.
173 Lui overlap slightly I meany'e.
(GE: 171-173)
140 Bai Yeah and, not only, not only for films music.
141 Lei Five hundred pounds Films but also music.
142 Pin Entertainment
143 Bai Entertainment
(CH: 140-143. SeeAlso e.g. AM: 113-115,267-269,336-339; 150-151; EN: 440-441,
488-490; GR: 87-90; IN: 39-42; NI: 217-221; NO: 159-162, SP: 78-79,
100-102
The most frequent evidence of collaboration, however, comes from the use of
backchannelling signals. Naturally, some participants use thesemore frequently than
others, but they are present in all the conversations. Two examples of rather
enthusiasticbackchannelling in the EWL conversationsmay be found in EWL 3: 99-
107 and EWL 7: 197-208:
99 Lin We, we could make them

100 Ke Yeah.
101 Lin You know what, I have an idea. I think we should to Art and Design
102 Dav Yeah yeah.
103 Lin Ask the studentsthere whether they could make time to make something.That
104 Ahm Yeah
105 Ke Yeah

106 Lin we could do but we should keep it simple becauseit's, it's really, i, it's not very
107 Dav Yeah
EWL 3: 99-110
199 Cla Yeah he will needto give a speech,the VC will need to give a speechand-
200 Bet Yeah
201 Cla can think you can have a couple of, uh, gueststhere-
202 Bet Hmm.
203 Cla So that the university standsfor what it is-
204 Bet Hmm
205 Cla an academicinstitution at the sametime as providing him-
206 Bet Mmm.
207 Cla an insight on the life of the campus.
208 Mag Yeah.
(EWL 7: 197-208)
The very uneven distribution across the EWL conversations (only 3,4,4 and 5
instancesin Conversations1,9,8 and 4 respectively, 9,12 and 17 in Conversations6,
10 and 2, no fewer than 27,46 and 64 in Conversations7,3 and 5) has mostly to do
with the personal idiosyncrasiesof particular What

speakers. is noticeable is how all
save one of the homogeneous conversations make relatively little use of
backchannelling, with normalised numbers ranging from 0.4 (EN) to 16 (SP), while
AM has 33, more than double SP, which is secondin ranking. This is almost entirely
accounted for by the near-obsessiveuse of `right' and ùh-huh' by one of the AM
speakersin particular (Sindy).
There are times in the EWL conversationswhen these three convergencemarkers,
accommodation, collaborative turns and backchannelling, co-occur, producing an
extremely collaborative This

sequence. example comes from EWL 2: 144-160:
145 insteadof going out buying something we just go and rent it.
146 Der Er, indeed, yeah
147 Pin Yeah.
148 Bay That-hopefully that will be cheaper.
149 Der Indeed.
150 Pin Yeah.
151 Bay Uhm The other thing, uh, for my portion of the project, instead of going renting a
152 professional film crew we can-
153 Pin Maybe-

154 Bay get a consultant,just hire a person

155 Pin hire, hire adult students, hire stu ents.
156 Bay Well, we'll be using the studentsbut get so body professional so he can
157 direct the students' activities.
158 Pin Yeah
159 Bay So that will.. end up being a lot cheaper.
160 Der Yeah that's, that is maybe a good idea, like you say........
(EWL 2: 144-160)
Derek and Bayeh express their co-operation with Ping by their use of
backchannelling signals ("indeed" and "yeah" respectively); Ping collaborates with
Bayeh by suggesting students while Bayeh is proposing the engagement of a
consultant and Ping to

accommodates Bayeh by taking up his use of `hire'. There are
no examplesof this three-way co-occurrencein the homogeneousconversations.
4.1.5. Explicit agreement and concessions
Sofia's, Hedda's and Javier's efforts at achieving comity by agreeing explicitly with
each other and the other speakers and by making concessions were noted in Chapter
7, Section 2. Further proof that interactants in all the EWL conversations set out to
achieve interactional goals as well as the transactional one given in their task may be
found in the fact that they too constantly express agreement both explicitly and
implicitly:
16 and becauseand that's can improve our university's image. And a nice
17 banquetwill be, uh, must be nice food with wines and, That budget for
18 this is, uh, around two to th, two to three thousands.
19 Der Hu-hum. Yeah, that's a good idea.
(EWL 2: 19. Seealso e.g. EWL 1:55; EWL2: 39,70; EWL 3: 83,161,242;
EWL 4: 73; EWL 5: 66,147,351; EWL 7: 15,73,142,268; EWL
8: 19,66,151; EWL 10:42,303,402)
They also, like the participants in EWL 6, decide to settle for less of the budget than
they had originally planned to obtain, again with the aim of making interactional
goalsas importantas,or evenmoreimportantthan,the transactional

ones:

298 Ana So how are we going to allocate the expenses then? Because look at it. If
299 the budget givens I can compromise to one thousand, I think that is quite a
300 big step I'm moving, you know to compromise with that expenses.
(EWL 5: 298-300. Seealso EWL 2:36-7; EWL 3: 73,259-63; EWL 4: 60,64,
67; EWL 5: 167,331; EWL 7:302; EWL 8: 27; EWL
9: 213,229; EWL 10: 177,326)
The overall amount of concessionin the EWL conversations is rather more than in
the homogeneousconversations:67 marked turns comparedto 50. When it comesto
the
explicit agreement, difference is far greater: there is almost double the amount of
explicit agreementin the EWL conversations,compared to the homogeneousones -
130 marked turns compared to 75.
What concessionand explicit agreementthere is in the homogeneousconversations
is fairly evenly distributed with SP seeming the most convergent in this way,
showing an overall high (13 explicit agreementturns, 10 concessions)and GE the
least (3 explicit agreement,1 concession).
Speakersin the homogeneousconversationsexpress their agreementby the use of
`yeah' or `yes' (CH: 108,123,154; GR: 40,45,93; NO: 19,85,156), by the use of
adjectives `great', `good', àlright' and `fine' (AM: 336; CH: 79,109; GE: 71; IN:
105; NI: 26, NO: 139; SP: 122), by using the performative Ì agree' (EN: 32,36,175,
436, SP: 83,88) or by repeating,restating or rephrasingprevious turns:
132 Ine Uh, about two or three thousandpounds. It's not very much.
133 Con It's quite cheap.
(SP: 132-133)
140 Mil Yet there's, it's kind of open, it's a little too open.
141 Sin Yeah it's a little dangerous.
(AM 19-23
19 Bai I think uhm, such as painting as the outside of the building uhm, if they
20 come, they will, it is appearedin their mind firstly.
21 Qin Uhm, but I think, uhm, we should do something cleaning. You know
22 that if the VIP come here, when see the very clean walls and the fresh flowers
23 and they will be very gre, great and glad
(CH: 19-23. Seealso AM: 13,141,268; SP: 10,19,23,108,119)

Concessionis mainly expressedby short phrasessuch as Ì know', 'OK', àll right',
`that's true' and Ì see' (AM: 64,170,321; EN: 153,494; NI: 12,99,102,131,140,
216; SP: 17,51,215). Concessionin EN is characterisedby the use of performatives
Ì concede' and Ì can compromise' (EN: 69,201,204,479,498,506).
4.1.6. Maintaining comity despite transactional goals
It is quite clear from the foregoing that participants in all conversations are as
focused on interactional goals as they are on the prescribed transactional one of
making bids for sums of money, the total of which exceedsthe overall amount of
In
money available. some conversationsthey may even be more focused on unstated
interactional goals than on the given transactional one. They use a mixture of
paralinguistic behaviour (laughter), within-turn linguistic behaviour (jokes, vague
language, inclusive `we' and `you'), conversation exchange behaviour
(accommodation, collaborative turns, backchannelling) and illocutionary functions
(agreement, concession) in order to create, maintain and emphasise the spirit of
in which they clearly want and like to work.

convergence
In order to achieve the transactional aims of the meeting, some divergent behaviour
is nonethelessrequired: participants do need to make points and occasionally to take
others. It was noted in Chapter 7 that interactants did this in ways

a stand against
intended not to upset the co-operative atmosphere and the interactants in the
remaining EWL conversations use similar strategies to achieve the same effect.
Suggestions, for example, are often made in tentative ways, using appropriate
hedging devices to render them less threatening:

modals and

54 Hal Well it maybetoo much money to spend,like, that much money in,
55 in security-
(EWL 8: 54-5. Seealso e.g. EWL 1:51-2,62-3,120-1,155; EWL 2: 90-1,
125-8; EWL 3: 7-8,42-3,45,125-6,149; EWL 4: 84-6,88-91,
97,176; EWL 5: 145-6,164,242-3,286,327,358; EWL 7: 12-
3,195-6,209-10,239,279-80; EWL 8: 54-5,145-6,245-6,218;
EWL 9: 167-9; EWL 10: 152,177,249-50)
In the same way as the participants in EWL 6, interactants in the other EWL
conversations also use a range of other means in order to make their suggestions
more acceptableto others and therefore to maintain comity. A significant number of
suggestions are made with explicit downtoners such as `just' and `really' (in
conjunction with a negative:
316 So I, my party is just to have impression for the person that's coming.
(EWL 10:316)
116 think of a budget for promoting and..it as well. We can't really use like five
117 thousand-
(EWL 1: 116-7. Seealso e.g. EWL 2: 12-3; EWL 3: 45-6,231, EWL 5: 99,295,
322-3; EWL 7: 82-3,279-80,323; EWL 8: 9,117,199-201,
245-6; EWL 9: 111
Other suggestions appear to be toned down by the use of vague language.
157 Lin Well, we have to use like a list, I don't know.

(EWL 2: 157. Seealso e.g. EWL 1: 13,15,62-3,95; EWL 2: 74,90-92; EWL
3: 6,69,71,93,191; EWL 7: 14,102-3,246; EWL 8: 42,144-5,
189,222,232-5; EWL 9: 87-8
Other ways of making suggestions while maintaining comity are to add an
ìnclusive' question or to add hedging laughter, much as Sofia did in EWL 6:
72 Ric Well, that's a lot of money to spendon security, don't you think?
(EWL 4: 72. Seealso EWL 1: 77; EWL 3: 172,217-8; EWL 7: 118; EWL 10:
245-8,301-2)
34 Ana Do you think you're able to. Try to reduceyour style @@@
(EWL 5: 34. Seealso EWL 3: 50; EWL 7: 335; EWL 9: 168,213)
There appearsto be a great deal more of this softening behaviour in the EWL
conversations compared to the homogeneous ones. After normalisation, the number

of hedgesin the EWL conversationsis 122, while in the homogeneousones it is less
than half, at 53. While the use of softening laughter and the number of `downtoners'
is roughly the samein both setsof conversations,the use of vague languageto soften
a proposal is, like the use of hedging, far smaller in the homogeneousconversations
with, again, only half the instances present in the EWL ones. The number of
ìnclusive questions' is, similarly, significantly higher in the EWL conversations
than in the homogeneousones.
Among the homogeneous conversations, most hedges are, after normalisation,
present in CH (CH: 21-29,56-63), followed closely by AM (e.g. AM: 38,86,321,
327,331). There is a significant amount of downtoner use in IN (IN: 10,16,28,92)
and, again, in AM (e.g. AM: 89,101,104,140,246). Vague language and laughter
are used to soften turns in severalconversations:
6I think I, the money would be better spent on repairs and cleaning and sort of
painting,
(GE: 6
286 Sus And if we're going to have a banquet, we're going to have a banquet, then
287 I'd like to decorate that hall with like flowers and stuff
(EN: 286-7. Seealso e.g. GE 18,23,40,41,48,127,148; IN: 16,29,31,70,
103; EN: 45,473; NO: 45-6,49,238,248).
There is a scatteringof inclusive questionsacrossall the homogeneousconversations.
4.2. Divergent behaviour
All of the foregoing does not, of course, mean that all the conversationsare free of
unfettered divergent behaviour. There are plenty of examples of participants
challenging each other and making suggestions,or even demands, without recourse
to hedges and downtoning. There are also instances of clearly unco-operative
conversational styles. There tend to be more instances of overt, unsoftened

challenging and bare demands in the homogeneousconversationsthan in the EWL
ones.
4.2.1. Challenging
Alongside the inclusive questions referred to above as helping to maintain an
atmosphere of comity, interactants also use challenging questions where, on the
contrary, they seem to want to make their individual transactional goal more
important than any interactional goals. There are clear examples in most EWL
conversations:
122 Mag Exactly. What about snipers?Don't forget snipers.

(EWL 7: 122. See also e.g. EWL 1: 23,50,75,76,106,152; EWL 3: 25,47-9,91,152;
EWL 4: 9,20,131-2; EWL 5: 80,273,347; EWL 7: 52,149-50,257; EWL
8: 21,87,103,107,237; EWL 10: 66-7,72-3,141,185,202,205,213-4,220-
1-343)
The average amount of challenging questions across all the EWL conversations is
roughly half that occurring in the homogeneous conversations. After the usual
normalisation, two EWL conversations have no challenging questions at all
(Conversations6 and 9) while all the remaining conversations except EWL 1 have
five or fewer. Normalised results for the homogeneousconversationsshow a high of
11 challenging questions in NI, followed by 8,7,6.5 and 6 in GR , NO, EN and CII
respectively. AM, GE, IN and SP are more similar to the EWL conversations
inasmuch as they show only 3,3,5 and 4 instances of challenging questions
respectively.
4.2.2. Demands, not suggestions
Throughout all the conversations there is ample evidence of participants insisting
be spent in the way they want it to be spent and that the sum they
that money should

require should not be reduced.When participants choosenot to hedge their point nor
to tone down their demand, this insistence results in `divergent' conversational
behaviour, often characterised by modal verbs `have to', `must', `need to' and
`cannot' and by interactants intensifying their demands with `very'; ìt's very
important', `definitely' etc.
222 Hal I still want to spendtwo thousandpound, around two thousandpounds for
223 entertainmentbecauseit's really important. Everything should be all right.
(EWL 8: 222-223. Seealso e.g. EWL 1: 4-5,110,150; EWL 3: 38-9,135-6,251; EWL
5: 17,139,195,220,274; EWL 7: 8-9,12-3,61,102-3,302; EWL 8: 22-
3,67-8,88-9; EWL 9: 115-8,135-6,212-3; EWL 10: 14-5,64,299;
AM: 13,56,142,161; CH: 17,56,69; EN: 66-7,113,167,324; GE: 28,
141,167; GR: 18,46,54,85; IN: 33,36,58, NI: 24,54,92,150; SP: 19,
48,83,217)
The averagenumber of unhedgedphrasesusing modal verbs expressing obligation
or impossibility is roughly the same in the EWL conversations as it is in the
homogeneousones. The use of intensifying words and phrases is, however, much
more frequent in the homogeneousconversations, with an average of 11 uses per
EWL (low 7, high 21) compared to an averageof 5.4 (low 0, high 11) in the EWL
conversations.
4.2.3. Closure
Another way in which interactantsexpresstheir divergence from each other is in the
use of `closure' phrases,intended to curtail the treatment of a topic or in any way to
bring a point to a close, without inviting further discussion, thus effectively blocking
comity and There

co-operativeness. are slightly fewer examples of this behaviour in
the EWL conversationsthan in the homogeneousones.
55 Der and I think they, uh, really appreciateit and they are al-also really proud and that their
56 painting is hanging on that wall.
57 Pin Yeah
58 Der So they are really proud and we can reduce costs, so I think, uh, that's a good solution. I
59 hope so.
(EWL 2: 55-59. See also EWL 3: 18,80; EWL 5: 66,112)

22 So we should spendmoney on security. This is my suggestion.

(GE: 22. Seealso CH: 5,29,40,62; NI: 167-8; NO: 118)
4.2.4. Other divergent behaviour
Challenging questions and the use of certain modal verbs and intensifiers do not
account for all the divergent behaviour in the conversations. There are times,
although rare, when interactants express their divergence directly, with the
performative Ì disagree' (EWL 3: 19; EWL 8:40; EWL 9:33,152; EN: 33,139-140)
they add a marking expression which appears to signal divergence (e.g.

or when
EWL 2:75 "But as I already mentioned"; AM: 316-7 "But when it comes down";
IN: 71 "That's I
what said", 82 "So what I say is..", 89 "What I'm saying is"... ). Far
than these lexical markers is the use of prosodic features to show

more common
divergence. Interactants frequently use a rise-fall pitch change on tonic syllables to
indicate the authoritative, and therefore divergent nature of their turn. This is mostly
by greatervolume and a slowing in rhythm:

accompanied
99 Ann Yes, but they won't film for two thousandpounds, nobody films for
100 two thousandpounds.
(EWL 1: 99-100. Seealso e.g. EWL 1: 131-135; EWL 4: 25; EWG 5: 21; EWL 7: 92-3;
EWL 9: 157; EWL 10: 397; EN: 10,42,261,472; GE: 158; CR: 42; NI:
8-59" NO: 61" SP: 20
Apart from turns which are clearly marked for divergence in this way, there are
turns which take the form of bald, unsoftened assertionsand counter opinions
many
others where speakersinsist or at least simply reinforce a previous point, again,

and
with no attempt at softening.
22 Ric We, we needmore money for a preparationsthan the welcome.

(EWL 4: 22. Seealso EWL 1:99-100,146-147, EWL 2: 75; EWL 4: 27,38-39,72,78,131,
136,175; EWL 5: 23-24,234; EWL 7: 85,87,144; EWL 9: 47; EWL 10:97;
AM: 88,118,119-120,191,333; C11:13; EN: 119-120,227-228,253,388,471,
492; GE: 100,123,128-129; IN: 51,82; NI: 47,99,101,105,108,126-127;
NO: 54,56,63-64,113; SP:29

21 Joh So how much are we going to spend?

22 Boo Uh, five thousand pounds.
23 Sam Ohh
24 Ang Five thousandpounds.
25 Boo Five thousand.
26 Xin I think it's impossible becauseif you want to, uhm, uh, suit up a security
27 systemyou need much technology system. So it should be spentmore than five
28 thousand.
(EWL 9: 21-28. Seealso EWL 1: 109; EWL 3: 150; EWL 7: 306-307; EWL 9: 37,
102,131; EWL 10: 159,265; CH: 99,102,113,114,133,138;
EN: 11,32,121,197,239,290,311,357-358; GE: 4,196; GR: 67,68,
84,96-97; NI: 36,39,43,50 128,185,222,225; SP: 54
195 Fan I think two thousandis enough.

196 Sar No. Never never.
197 Shr A thousandis enoughfor your flowers.
198 Fan You say we can bring some,bring flowers to school
199 Sar Then I, then I, then I can decor to only the hall. I can
200 paint only the hall. Then how can get, films, BBC, they com round the university.
201 How can they feel?
202 Fan I think two thousandis enough.
(EWL 10: 195-202. See also EWL 9: 200; EWL 10: 306; AM: 72,105; EN: 310,387; CE:
11; NI: 107,195; NO: 87)
Rarely, the baldnessin assertionsturns to scornful remarks
62 Gra Yes but I think like security is more important than all thesethree causelike
63 Corn You must be joking
(NI: 63. Seealso EWL 8: 6; EWL 10: 276; GR: 48; IN: 53; NO: 108; SP: 86,202
There are three more types of divergent behaviour of which there are no instancesat
all in the EWL conversations but several examples in the homogeneous ones.
Participants in the latter conversationsoften simply gainsay a previous turn, often in
the most laconic way This doesnot happenat all in the EWL conversations.
204 San I think we should think

205 about food after we've taken care of the main priorities such as security and I would
206 say half of the budget would be spentbut I'm going to compromise and say OK a little
207 under a half. We'll say five thousand.
208 Ian I wouldn't-
(EN: 206. Seealso EN 217,289,354,443,449,523,565; CR: 14; NI: 35,66,67,88,129,
163; NO: 52-3,55,77,78,79; SP: 66,74,118)
A further, albeit rare divergent feature of the homogeneousconversationsis the use
of dominant discoursemarkers such as `Look' (GE: 19; NI: 84) and `Listen' (NI: 75;
NO: 115). One of the interactants in NI goes further still, with a defiant reprimand:
"I hopeyou know that" (NI: 89-90).

The instances of all types of divergent behaviour are far more numerous in the
homogeneousconversations(average 14 per EWL; high NI 28, low SP and GE 10)
than in the EWL ones (average6 per EWL; high EWL 10 - 11 instances,low EWL
2 -I instance).
4.3. High involvement and high considerateness styles
The problem with recording convergenceand divergence markers is that it assumes
common ground among speakers,that speakersare likely to follow the same route
towards the achievementof interactional goals but that some are more willing to do
so than others, or more willing to do so in one situation rather than another. Within
this framework, it is perfectly possible that the interactants using the most
divergence markers are also those using the most convergence ones. By marking
their turns in either way, they are providing evidence of their choice of a high
involvement conversational style and perhaps consider that the way to interactional
is
goals through engaging personally with other participants, whether this means
supporting them or disagreeingwith them.
4.3.1. Involvement and considerateness styles In EWL
conversations
EWL 7 is unique among the EWL conversationsin that all the participants seem to
high involvement style. it can be no coincidence that EWL 7 also appearsto

share a
be the most meeting-like of all the EWL ones.
In EWL 8, all participants seem, conversely, to have a distant and independent
It has already been noted that this EWL is totally lacking in

conversational style.
laughter and contains few backchannelsor inclusive questions.

In other EWL conversations,there is evidence of conflicting conversational styles
with some interactants clearly attempting to involve and engage fellow speakers
while others appear to want to maintain safe distances, perhaps endeavouring to
achieve and maintain a senseof comity by allowing others maximum freedom and
protecting their negative face.
In Conversations 1,5 and 10, most participants use a high involvement style, with
consideratenessspeakers in a minority of one in each case. Stavros, Anne and
Comfort (EWL 1) are clearly keen to engage, making free use of `you' and
approving and rejecting others' ideas. Gauri, on the other hand, uses a lot of
impersonal language,uses `we' (rather than, `you') to refer to other people's ideas
and, with few exceptions, does not tackle other interactants head on. Ana, Susy and
Lei, in EWL 5, and Mala, Shray, Sarraj and Fang in EWL 10 are also high
involvement speakerswhile Chan (EWL 5) and Qing (EWL 10), with their more
approach,
considerateness tend to becomemarginalised.
In Conversations 2 and 9, the bias seems to be the other way, with Derek trying very
hard to keep everyone going on his terms by giving a great deal of approval and by
enthusiasticbackchannelling:
17 will be, uh, must be nice food with wines and, That budget for this is, uh, around two to th, two
18 to three thousands.
19 Der Hu-hum. Yeah, that's a good idea. I also think uh, that uh, the guest has, yeah certainly
20 have to drink something, so-
(EWL 2: 17-20. Seealso EWL 2: 70,160)
95 Der Excuseme?
96 Yan Appropriate seating.
97 Der Uh-huh
98 Yan Yeah, I think it's essentialfor, for the meeting and will, some machineswill make the,
99 uh, make the conferenceclearly-
100 Der Yeah
101 Yan and, uh, I think they will helpful, uh, f, for the conferenceto be successful.
102 Der Yeah
uh we canpre,preparesome, uh, drink
103 Yan And, uh, I think, uh, uh, in, in a conference,

104 soft drink.

105 Der Uh-hum
106 Yan Uh, , and uh someint.. interpreter systemto help others to hear clearly.
107 Der Uh-hu
(EWL 2: 96- 6. Seealso EWL 2: 66,143-149)
Two at least of the other three participants seemto want to move matters forward in
a different way: Bayeh deals with Derek's suggestionsfactually, not personally and
his own proposal forward without attempting to engage other opinions on it;
puts
Yan participatesin a similar vein.
In EWL 9, it is John who, albeit only occasionally, seemsto want to engageand to
encourageothers to do likewise while everyone else is content to use a less personal,
more independentstyle:
79 Joh How much are we, how much have we got to

80 Sam money. So it's very great.
81 Boo We have, uh-
82 Sam Becausewe, uh, we want to spendour money in, uh, five case.so you can't,
83 uh, you can't spendtwenty-five percent of our money for security system.
84 I'm sorry.
85 Joh How much money have we got together if we don't pay anything on
86 security? How much, Sam.No, Sam, how m.
87 Sam No. How mu for your case?
88 Xin About, uh, five thousand.
89 Boo Five-
90 Sam Five thousand
91 Ang Five thousand.
92 Boo Oh. Come on!
93 Joh @@@ How much for you?
94 Ang One thousand.
95 Joh Yeah?
96 Sev @@@
97 Ang And I think it's better not to spenda lot of money for this.
(EWL 9: 79-86)
In EWL 4, Richard and Pallu appear to be at opposite ends of the involvement-
distance spectrum with Richard constantly supporting or combating other speakers
and Pallu respondingby mirroring, rather than confronting.
in
Whereas all these speakers
conversations with different styles seemto coexist
happily with each other, the situation in EWL 3 is not quite so felicitous:
quite

Lina's high involvement style seemsto overwhelm the other participants who treat
her as a leader and take it in turns to get beatendown:
47 Lin You, you want to spendfour thousandpounds-

48 Ke Yeah
49 Lin on, on, on microphonesand, and seatsfor people?
50 Ke No, I just @@@, I don't meanthat. Uh, I, I think the most important is let the lecturer
51 sharehis or her ideas-
52 Lin Yeah, yeah OK
53 Ke for the audience.And if he or she didn't have the chanceto do that, it is not, uh, worth,
54 uh, so much money.
55 Lin Yeah, but, I, I seeyour point, really I do. But still we are, we have to,
56 you know, this person needs to feel welcome because we-
57 Ke Yeah
58 Lin we invited this person,and thi person actually agreedto come here and
59 Ke Yeah, of course, yeah
60 Lin that's, that's a big thing itself owe need to make this person feel comfortable and
61 welcome.
62 Ke Yeah, comfortable and welcome. ý
63 Lin That's why I want us to ma, to, to give something in return, like-
64 Ke Um, yeah,
(EWL 3: 47-64. Seealso EWL3: 72-83
Yet, in her own way, Lina seemsto be trying to achieve co-operation and seems
unaware that her approach is not shared by the others. Ke, using a style at variance
with Lina's, seemingly tries to hold his ground without tackling her but is then
silenced by her forthrightness (EWL 3: 111-150), causing laughter among all
participants, perhaps reflecting the awkwardness of the conflict (EWL 3: 151).
4.3.2. Involvement and considerateness styles in homogeneous
conversations
At either end of the `high involvement' - `high considerateness'spectrum are the SP
and CH conversations.A look at comparable sections from each conversation will
make the point adequately: in SP Consuelo suggestsspending money on cleaning
and tidying, including renewingcarpetsand has to defend the idea in the face of
criticism from all the other participants:

45 Con Yeah and of course,I think we needuh, clean

46 and tidy all the areashe will be visit, he will be visiting becausein some of the
47 places in this Uni he can find that it's not a good impression, the first impression
48 you have; it's not very good so I think we have to spenda little mon, a little, little,
49 little money.
50 Ros How much?
51 Con Maybe three or, three or four thousand.
52 Mon It's OK
53 Inc Three or four thousand?
54 Con Yes, becauseyou have to-
55 Inc It's too much
56 Con renew all the carpets,paint, paint again most of the
57 Mon Renew all the carpets,why don't you clean the carpetsinstead of renewing them?
58 Con Becauseuhm, most of the times if you clean it, it is not any difference.
59 All @@@
60 Mon No, You have to take into accounthe is not going to, like-
61 Con But he's using different carpets,I mean they are more, very, very used and they only
62 changea small part it so it's better uh, better impression.
63 Mon Yeah but he's co-
64 Con Only one ca et. Uhm, please,clean.
65 Mon Yeah I know what yuu meanbut he's not going to be walking around every building
66 in the whole university but-
67 Con No but we can-
68 Inc We can prepare
69 Con we can predict in which placeshe will be
70 Mon Yeah but for ample if you have the
71 Con and think about.
72 Mon this big theatre in the De Havilland campus,you could just preparethat building,
73 you know, constantly and clean
74 Con No, no, which, which kind of isit is being only one building?
75 Mon No, no, no. It's not a visit, I mean..
76 Ros But we can speak with him
77 Con Yes and he, he will do a speechand we will having a banquetthere but..
78 Inc Yes, yes
79 Con later he could want see more, more things.
80 Mon Yeah to visit Yeah but you, you.
81 Con Only, only the buildings outside but never er in, on them.
82 All @@@
83 Mon Yeah but it's, what I mean is, the main building where this event is going to take
84 place, he just, if you want to changelike carpets,OK, I agree with that, but you have
85 to changeall the carpetson, in the whole university.
86 Inc in the university
87 Con No, no, no, no, it's so crazy
88 Mon You are not go' g to be able to do that
89 Con I agreewith you, you go-
90 Inc I think it, it's not bad. The other building are clean and the one which he is visiting
91 and he's staying and the banquetand so on, so on, uh, is the, it has to be clean and
92 ready and decoratedand I, You seemy point?
in CH, Xing suggestsspending two thousand pounds for a tea party and has then to
deal with criticism and counter-proposalsfrom the others:

32 Xin Good. I think I will spendabout two thousandpounds uh, to hold a tea party to
33 welcome this VIP and uh, if this visit goes well I mean uh, it finished successfully,
34 the staff and security guardswill get a reward. This sum of money should be two
35 thousandpounds and I will give two thousandpounds to the VIP as cash, so he'll be
36 very happy and I'm going to be very happy.
37 Various @@@
38 Pin But I think uh, uhm, spendtwo thousandpounds uh, is very very ch. uh, expensive
39 and I think it's only uh, one, one thousandpounds or one thousandfive hundred
40 pounds to, to give a tea party, tea party to we me VIP. It's my opinion.
41 Xin Oh, give the person one, one thousand Oh yeah.
42 Lei I suggestedtwo thousand,betweentwo thousa d and three thousandpounds to the
43 tea party becauseit's a very important part for us to welcome them and uhm food
44 and drinks are necessary and it may be we can invite some band to give us some
45 show. It'll be, it'll be nice.
46 Xin Yeah, yeah, yeah.
47 All @@@
48 Lei OK
49 Pin But I think if we spendtwo thousandpounds or uh, three thousandpounds to uhm,
50 welcome them uhm, the VIP uhm, uhm, to give him uhm we are relative luxury in
51 this aspect,I think.
52 Qin It is, you think two thousandis too much?
53 Pin Mmm.
54 Qin But what do you think about three thousandfo furniture are also very much?
55 Pin The too much-
56 Lei But I think the furniture must be to, to repaire or cleaned.The furniture is necessary
57 for the university, you know. The university, essentialof the one university maybe
58 first its equipmentand uh, the maybe the some environment. If the VIPs seethis
59 kind of environment is good, so maybe they will think the, the quality of this
60 university is good. I think it's necessary for uh, a university to put maybe five
61 thousandto ten thousanduhm, pounds this kind of things such as pa..uh, repairing,
62 painting and uhm, cleaning and uh, introduce some new, new furnitures. Yes. That's
63 my opinion.
The SP extract has 36 turns for 471 words (ratio 1:27) while the CII extract has 11
turns for 367 words (ratio 1: 33). While the CH extract is characterised by
impersonal language('the sum of money should be', `spend two thousand pounds is
very expensive') and measuredtones ('I think if we spend two thousand pounds
very
to welcome them, we are relative luxury in this aspect') the SP one is direct ('we
have to spend', `You are not going to be able to do that `) and boisterous ('No, no,
it's so crazy'). The CH extract has 3 instances of `you' for its 367 words
no, no,
(1: 122) while the SP extract has 17 (1: 27).
Most of the remaining homogeneousconversations are similar to the SP one, with
discussion of proposals,in short-ish turns, `you' orientation and little reticence

much
it comes to supporting or undermining other participants. AM, as has been

when

is
noted, rich in backchannelling and collaborative turns, EN and NO conversations
have a large number of turns where one participant directly challenges another,
engaging him or her personally and NI stands out for the familiarity with which
interactantsderide eachother's ideas in quite personalways:
88 Com Your own country, no, I'm not diversing becauseyour own country spendsmore on
89 your president's security than other countries would spend on his security. I hope you
90 know that. W-
91 Jos It depends.
92 Corn It dependson what? But it doesn't mean you are 'ng to have to spendso much, I'm
93 talking of somebodycoming
94 Jos I think, I think ( ................. ) spenda lot of move . You should agreewith me that
95 when a person gets a, when he gets a place-
(NI: 88-95. Seealso NI: 61,103,119,205)
GE seemsto be more measuredand distant, at least in its earlier stages.Sofia's idea
to create publicity material, for example, is approved of by Katia

of using students
Luise without any reference to Sofia herself; similarly, Greta argues against the
and
by taking the idea apart, rather than undermining the proponent, as one of
proposal
the NI participants may have done (GE 78-95). GR is similarly moderate, with little
in the way of direct challenges and mutual support among interactants. Rather than
Philip's suggestion for spending money on a banquet, for example, Gregory

approve
flatly responds by saying that it is `something we should consider in mind' (GR : 40);
the tenor of the whole conversation is on the distant, formal side, without there being
any unfriendliness or antagonism.
IN is to
similar many of the EWL conversationsin that there seemsto be a conflict
styles. Sukvinder, the self-appointed Chair, uses a high

of conversational
involvement style, making great efforts to include the other speakersby calling on
them as `you guys' (IN: 1,5,8,36,45,61,64,72,88,92,97,107) and directing
questions and appeals to them (IN: 15,21,23,27,30,64,81,96-97), the other

participants use a more independent, distant style with impersonal suggestions (IN:
11-12,17-19,24-26) and laconically stated facts (IN: 75,76-77).
4.4. Conclusion
Looking at all the conversations as if participants were sharing a uniform CO-
operative style, it is clear that the urge towards convergence is greater than that
towards divergence. That said, the EWL conversations show a greater tendency
towards convergence than the homogeneous ones, as is made clear by the charts
below (Figs 3 and 4). The data reported in the charts has been obtained by bundling
together all coded instances of convergence signals (unmarked laughter, humour,
collaborative turns, backchannel signals, suggestions which have

accommodation,
been hedged or toned down, explicit agreement and concessions) and divergence
(unhedged points, intensified demands, unmarked requirements, challenging

signals
The fact that participants in CH, GE, GR and IN may have been striving
questions).
for co-operation in a different way, using `high-considerateness' style, does not
to have made any difference to the overall look of the chart. The only striking
appear
of data is the high incidence of convergent behaviour tier AM.

piece
180
160 Q EWL 1
140 QEWL2
120 EWL 3
100 QEWL4
Q EWL 5
80
QEWL7
60
EWL 8
40
Q EWL 9
20
Q EWL 10
0
Fig. 3 Convergence and divergence in EWL conversations
201 Overview
Chapter 8 of all conversations
180
160 DAM
140 DCH
120 DEN
100 OGE
GR
80
DIN
60 ONI
40 DNO
20 SP
0
Fig. 4 Convergence and divergence in homogeneous conversations
5. Explanations
In Chapter 7, different turns in the two conversations under scrutiny were explained
by reference to a number of different speaker factors: sex/gender, cultural
background, interactants' perceptions of the meeting framework (including
connected perceptions of assumed or acquired rights) and participants' perceptions
of each other. The same four categories will each be used in turn to propose
for language behaviour in the remaining conversations.

explanations
5.1. Sex/Gender
Whereas all the EWL conversations except one are among mixed-sex groups, six of'
the homogeneous conversations have participants who are all of the same sex. AM,
EN and NI are among mixed sex participants.
Contrary to the evidence in EWL 6, men do not appear to dominate in the remaining
conversations. In EWL 1, Stavros is the only man and he assumes a

mixed-sex
dominant role. Conversely, in EWL 3 Lina is the only woman, yet she quickly
becomes the dominant force. Derek takes the leadership position in EWI. 2, ill Which
Chapter 9 202 Overview of all COI c satioons

Ping is the only woman, while Ana leads EWL 5, in which Chat is the only man. Ian
seemsto want to lead EN but never succeedsin doing so, while Milne seemshappy
to take a back seatin AM.
There is also little evidence of same-sexcollaboration in mixed-sex conversations.
The only EWL conversationwhich evolves into a collaboration between two men is
EWL 4, where Richard and Kris appear to work together to the exclusion of Pallu
and Bai. In all the other conversations, there appears to be little or no solidarity
among members of the same sex; high levels of convergence cannot, therefore, be
in
explained this way.
5.2. Cultural Background
It was noted in Chapter 7 that turns in EWL 6 may have been influenced by
speakers' perceptions of each other, and of themselves, as culturally situated in the
North, South, East or West constructedcultural areas.These cultural divisions very
the comparative economic dominance of the North over the South

roughly represent
the dominance of the West over the East in terms of establishing the interaction
and
rules for the current intellectual or academic elite. In the remaining EWL
conversationsfourteen participants may be thought of as situated in the North/West
cultural area as follows:
EWL 1 Stavros,Anne
EWL 2 Derek
EWL 3 Lina
EWL 4 Kris
EWL 5 Susy
EWL 7 Betty, Anja, Magda, Greta, Claude
EWL 8 Mary

EWL 9 John, Angela

EWL 10 No representatives
Skeleton profiles of all theseparticipants are given in Appendix C. Mary is included
here because, although geographically from the Southern hemisphere, her
background situatesher in the cultural North and West.
In their respective EWL conversations, these participants tend to use a
style characterised by ìnvolvement' (or `solidarity', `connection',

conversational
àcceptance', ìnterpersonal face'; see Chapter 5, Section 7.2.1) inasmuch as their
turns are more marked for both convergenceand divergence than those of speakers
situated in `South-' or Èast'- constructedcultures. The averageconvergencemarker
score per conversation for these interactants is 29, compared to 19.5 for participants
in the South/Eastcultural area, while the average divergence score is 6.5 compared
to 4.6. The extent to which the use of divergence markers is a feature of North/West-
situated speakers would require further research, given that the tendency is not
totally clear from the present data. In EWL 1, Stavros uses 20 divergence markers
the South/East-situatedGauri uses only 2; the pattern is similar in EWL 3

while
(Lina 15, Ahmed, Ke and David 1 or 2 each), EWL 5 (Susy 14, Chat and Lei 2 and 7
and EWL 8 (Mary 12, Yong, Haluk and Joseph 3 or 4 apiece). EWL 7
respectively)
is among interactants who are all North/West, EWL 10 among interactants who are
South/Eastand in Conversations2 and 9 there is little use of divergence markers

all
But this pattern of divergence markers being used predominantly by

overall.
North/West speakersis somewhatdisturbed by Richard, in EWL 4 and Ana in EWL
5. Richard, culturally situated in the South, has 17 divergence markers (while Kris, a
North/West participant, has only 2) and Ana, situated in the East, has 20 to Susy's 14.

Further evidenceof involvement style used by North/West interactantsmay be found
in their higher levels of `you' orientation. In EWL 1, for example, Anne and Stavros
make frequent use of `you' oriented behaviour (1: 50,75,76,82,84,106,139,147)
in stark contrast to Gauri. Lina, Susy, Betty, Mary and John are similarly `you'
oriented in their respective conversations(EWL 3: 20,25,47,65; EWL 5: 63,172,
258,276,315,347; EWL 7: 32-7,120,257; EWL 8: 4-5,40,47,103,123-4,216,
220; EWL 9: 93,178,217-8).
The `you' factor in the homogeneousconversations was given a little attention in
Section 4.1.3. above: no pattern was found to be evident. It is noticeable, however,
that the group with the lowest `you' count is CH, possibly connected to a
consideratenesstyle, predicted from the literature on the subject.
High Involvement conversationstyle also involves participants' approving of others,
bolstering so-called `positive face'. Again, the presence of the North/West
participants in the EWL conversationsprovides the cultural reason for a fairly high
incidence of explicit agreementand approval. Derek, Susy and Betty use many such
(EWL 2: 19,22,66,70,143; EWL 5: 40,101,161,252,341,364;

approving moves
EWL 7: 73,81,117,132,215) while Lina is almost overwhelming (EWL 3: 23-4,55,
74-5,140-50,264-72). In EWL 4, Kris does not behave in a similar way, which
be explained by the fact that he does not have a leadership role. On the other
might
hand, the explanation might go in the opposite direction: Kris does not emerge as
leaderbecausehe doesnot use the associatedconversationalstyle.
More evidence of high involvement among North/West participants comes in an
of collaborative turns and backchannel signals: speakersseem to enter

examination
into collaborative turns and to provide backchannelling signals in order explicitly to
support other interactants:

220 Lei Uhm, but, uhm, as I think, uhm, becausewe, we really do need, uhm, perfect banquet
221 for the visit-
222 Sus Ummm.
223 Lei Becausewe needto give him a impression of our University and our welcome to him
224 and, uh, to end up with a very happy feeling-
225 Sus Yeah
226 Lei But, uhm, how, how can i, we, we are still, uhm, university and we don't make profit.
227 We are not commercial, you know..
228 Sus Hmmm.
(EW5: 220-228. Seealso EWL 2: 95-106,144-9; EWL 3: 28-34,143-8,245-8,256-9,
262-71; EWL 5: 45-7,54-61,84-91,156-161,200-212,274-9,352-5;
EWL 7: 199-206,311-9; EWL 8: 128-31,154-6; EWL 9: 168-72)
Similarly, there are a couple of instances of accommodation involving these
interactants which seem to show them accommodating to other participants in order
to support them. In EWL 3, for example, Lina echoes David (3: 301-3) and Susy
to be supporting Lei and Chat when she echoes Lei's use of àtmosphere',
seems
Chat's Ìt's worth' (EWL 5: 98-101,160-61). Finally, in EWL 9, John appears

and
to be accommodating to Sammi: his decision to say `Do we need to pay it? seems
here, unless seen in the light of Sammi's previous turn. (EWL 9: 198-201).
odd
The homogeneous conversations do not show any significant patterns in this area:
the distribution of collaborative turns, explicit convergence, backchannelling and
is uneven across the conversations with highs in AM and SP and a

accommodation
low in EN. The two Èast' groups, CH and IN, are at the high middle and low middle
when it comes to these types of turns and the `South' (NI) group
points respectively
has a similar rating to CH (See Fig 4 above).
This might suggest that North/West speakers emphasise high involvement styles
in an international setting. A closer examination of this possibility will be

when
in the following chapter.

conducted
5.3. Perceptions of the meeting generic framework
Whereas participants' sex/gender seems to have little effect on conversational
language, their cultural background, on the contrary, appears, then, to have some

bearing. Equally, the meeting setting in itself, whether perceived of as a committee
meeting or as a negotiation meeting, appearsto have had a determining influence on
participants' conversational behaviour. Further, participants' perceptions of the
meeting setting may themselvesbe partly culturally determined.
5.3.1. Leaders
The cultural construction of the committee meeting suggests a Chair who,
impartially, ensures that all other participants have a fair chance to put their points
and make their case.The negotiation construct, on the other hand, might suggesttwo
sides each trying to gain the upper hand while temporary alliances are
or more
formed in order to see off opposing sides. In the case of a negotiation for the
funds, the classically constructedmeeting might have a manager who is

spending of
responsiblefor expenditure.
ultimately
In the conversationsunder scrutiny, it will be rememberedthat participants were not
specifically instructed to act as a committee or to act as business-style negotiators.
No participant was asked to be Chair or manager in any conversation. Some
conducted along committee lines, while others seem more like

conversationswere
In either case, most of the EWL conversations have a clear leader, as
negotiations.
EWL 6 had Milne and Hao (see Chapter 7, Section 2). Where the leader seeshim or
herself as a committee Chair, or as an ìnclusive' manager, the resulting
style is likely to be strongly convergent. Where the leader perceives

conversational
her or himself to be a more authoritarian manager, the style will naturally be
somewhatdifferent.
In EWL 1, Stavros quickly emerges as the self-appointed Chair and succeedsin
other participants' through

suggestions him, while appearingto take
channelling

overall responsibility for the final in

outcome, particular by making general summary
statements about how the money will be spent (EWL 1:59-68). He uses 28
convergencesignals, more than anyone else in the conversation (Gauri 15, Anne 13
and Comfort 10).
The Chair role in EWL 2 seems to be adopted by Derek who, unlike the other
the proposals put forward.

participants makes seemingly objective evaluations of
Like Stavros, his conversational style is one of high involvement and he also uses
more convergence markers (40) than anyone else (Ping 20, Bayeh 10, Yan 9).
7
EWL also has a clear Chair: Betty assumesthis role shortly after her entry into the
like Stavros and Derek, makes persistent use of a high

conversation and,
involvement conversationalstyle. The convergencemarkers noted in her turns total
60, far outweighing those used by other interactants (Anja 26, Greta 23, Magda 18,
Claude 10).
In EWL 4 Richard appears to take more control than the other three participants but
in a managerial way than as a Chair. In a rather unsubtle way, he uses a high

more
involvement strategy to challenge others head on, calling them to collective
The high incidence of convergence markers he uses (37, compared to

responsibility.
20,13 5) may therefore, once again, be explained in terms of Richard's

and
how negotiation meetings should be conducted and of himself as leader.

perception of
In Conversations 3,5 and 8, the control role is adopted by Lina, Ana and Mary
In all cases, the role is definitely managerial: Lina appears to give

respectively.
herself the right to challenge the others, presumably since she feels the responsibility
for bringing the proceedings to a proper outcome, more than the need for ensuring
has a voice. Ana's role in EWL 5 is similar, while Mary succeeds in having
everyone
through her. As with Stavros, Derek and Richard, Lina and Mary
all turns channelled

use more convergence markers than the other participants (Lina 68, compared to 52,
25 and 16, Mary 26, compared to 15,12 and 7). The very even distribution of
in EWL 5 (Susy 53, Ana 53, Chat 49, Lei 33) will be dealt
convergence markers
with below (3.1.2. ).
In the remaining conversations (9 and 10) John and Shray appear to take on a
leadership role but in much less obvious ways than their counterparts in
Conversations 1-5,7 and 8. John is very low key, occasionally reminding other
of their responsibilities, while Shray, in a similarly low-key way, seems

participants
to be the only participant with the right to change topic and to bring participants'
proposals together. The number of convergencemarkers associatedwith these two
is high comparedto other speakersin the same conversations.EWL 10

speakers not
is, indeed, the least meeting-like of all with one participant, Mala, trying on several
occasionsto establish,unsuccessfully,some sort of order. Perhapsthis might explain
the relatively high number of convergencemarkers attributed to her: 27, comparedto
Shray (12), Fang (13) and Qing (4). Sarraj's score of 22 cannot easily be explained
by referenceto the meeting setting.
As can be seen, Chairs and managers are more likely to be North/West-situated
but there are two very noticeable exceptions in Richard (EWL 4) and
participants,
Ana (EWL 5).
Most of the homogeneous conversations seem to work more like committee
than businessnegotiations and most have no clear Chair figures or leaders

meetings
kind. There is one strong exception to this and a couple of weaker ones. In IN,
of any
Sukvinder takes the role of Chair and dominates the conversation from the beginning
He is responsible for half the instancesof `you' in the whole conversation

onwards.
for half the total number of convergent turns. In AM, Sindy seemsto be keener
and

than others to make sure everyone is heard and appreciated,without really taking on
a thoroughgoing leadership role. Her large share of the backchannelling signals in
the conversation gives her one third of all the convergence turns (there are four
participants in the conversation). In EN, Ian clearly attempts to take the chair,
reminding participants of the general parametersand appealing to speakersto make
concessions.His use of `you' accounts for one third of the total use and nearly a
third of all convergentturns (there are five in

participants the conversation).
Otherwise, speakers in the homogeneous conversations tend to share the
responsibility for making procedural comments: George and Gregory both do so in
GR (GR: 96-97,144), Fang and Lei in CH (CH: 83,157), Greta, Anne and Sofia in
GE (GE: 48,71-72,177), David and Comfort in NI (NI: 175,207-209) and Karen
and Hedda in NO (NO: 29,42,233).
5.3.2. Other, non-leader roles
If the meeting or negotiation setting allows some participants to see themselves as
leaders, and therefore pivotal to the discussion, it will make others see themselves as
directed towards the lynch-pin figure. In some cases, this means

somehow
interactants see themselves in a subordinate position and their language use will
likely be affected accordingly. In others, participants may well decide to challenge
the leader figure, using the meeting setting to behave in ways which they might not
normally consider acceptable.
In the EWL conversationswhich have been identified as including a clear Chair or
manager figure (Conversations 1-5,7 and 8) the non-pivotal speakersall direct most
of their turns towards the central figure or figures.

In EWL 2 and 3, there are clear examples of a leader - subordinate relationship.
After having presented her initial proposal, Ping (EWL 2) does little more that
provide backchannelling signals and signs of explicit agreement against Derek's
responsesto her (EWL 2: 15-39). Whenever she speaksat later stages,it is usually to
agree with, or in any case simply to acknowledge other speakers.Bayeh and Yang
are less subservient but nevertheless address themselves predominantly towards
Derek.
In EWL 3, a pattern soon emerges with Ke and then David making suggestions
which are challenged by Lina, leading to climbdowns in all cases with a great deal of
backchannelling and agreeing on the part of the two men. (EWL 3: 47-64,72-83,
111-123). When Lina puts forward her proposal for refreshments, David and Ke
once again pull out their agreement signals (162-184). Their self-imposed
for the relatively high number of convergence markers

subordinate role may account
for them in the conversation overall (Ke 52, David 25). This pattern is also
noted
apparent in EWL 5, where Chat and Susy often find themselves backchannelling
Ana. The difference is that these sequences do not lead to Chat and Susy
against
accepting Ana's challenge to their proposals (EWL 5: 1-36,44-61). Their
backchannelling signals do account, however, for well over half and over a third of
their respective convergence marker totals (Chat 28/49, Susy 18/53).
1
EWL seemsto be characterised,on the contrary, by a senseof insubordination on
the part of all three non-pivotal participants, who nonetheless do not challenge
Stavros for the Chair position (EWL 1: 50,75,77,106,131-4). Anne eventually
mellows, addressingherself more tentatively towards the Chair (154-5) and Gauri,
too, becomesmore compliant. Comfort seemsto refuse this role and goes out of the
conversationwith a joke (171-2).

In EWL 4, Pallu respondsto Richard, in the first instance, and then to both Richard
and Kris, without subordinating herself to them but also without attempting to take
the pivotal position from them. In EWL 8, the rather heavy-handed,humourlesstone
set by Mary is mirrored in the other participants who do not easily accept her
authority but who neverthelessdo not attempt to take the leadership position from
her.
EWL 7 is perhaps most committee-like of all the conversations. All participants
addressthemselvesto Betty, but do not, seemingly, cast themselvesas subordinates.
Where they mark their turns for convergence,this is, presumably, to do with their
overall willingness to make the meeting work. Their number of marked turns is low
compared to Betty's because she takes such a central role throughout the
conversation.
The relative lack of leaders in the homogeneous conversations is mirrored by a
corresponding lack of subordinate (or insubordinate) turns. In IN where, as has been
noted, there is a definite leader, the other participants do not challenge Sukvinder's
leadership, but neither do they adopt subordinate postures. In all of the other
homogeneous conversations, the behaviour of the participants suggests that, even
where a weak leader seems to want to emerge, the others do not collude in allowing
this to happen. Sindy and Ian (see above) are often side-lined into silence, for
with participants bouncing ideas off each other, rather than channelling
example,
them through a Chair figure. This factor may account for the lower amount of
convergence markers in the homogeneous conversations.

5.4. Interactants' perceptions of each other
A fairly obvious explanation for the greater amount of convergence in the EWL
conversationscomparedto the homogeneous is

ones simply that membersof a group
of people sharing comparatively little, culturally speaking, are likely to make more
effort to be open and responsive towards each other than are people who share
common ground.
Convergent behaviour in the EWL conversations may also be accounted for by
referring to how speakers see each other in terms of comparative maturity,
and language Results

competence. from the preliminary surveys show
responsibility
that fourteen participants were considered by their peers to have a high level of
maturity, while six were thought of as having a high level of

responsibility and
language competence.At the other end of the scale, one participant was considered
to have a low level of maturity and responsibility and seven were thought of as
having relatively poor language skills. The relevant results reported in Chapter 7 are
summarisedbelow in Table 12.
Maturity and res ponsibility Language competence

High Low High Low
EWL 1 Stavros Comfort
Lauri
EWL 2 Derek Ping
EWL 3 Ke Lina Ke
ENVL 4 Pallu Bai
Richard
ENYL 5 Susy Susy Chat
Ana
Lei
EWL 6 Hedda Milne
Hao
EWL 7 Betty
EWL 8 Joseph Yong

Mary
EWL 9 John Angela John Boon
Sammi Sammi Xin
EWL 10 Mala Sarra'
Table 12: Participants perceptions of each others' responsibility, maturity and language
competence

Higher-rated participants tend to use more convergence signals, perhaps to show
solidarity with lower-rated ones. Becausemany of the higher-rated interactants are
also those situated in North/West cultural constructs,their role in collaborative turns,
as well as their accommodationtactics and their use of backchannelling signals has
already been dealt with (See Section 5.2 above). It is no coincidence that thesehigh-
find themselvesfrequently in leader positions, accounting for

rated participants also
further use of convergencesignals (SeeSection 5.3.1. above).
There are, however, some instancesof convergent behaviour which have not been
by cultural background or by reference to participants' perceptions of the

explained
meeting setting. In EWL 7, for example, where all participants are North/West-
situated, Betty standsout as self-appointed Chair and as a relatively prodigious user
turns. As can be seenfrom the above table, she is singled out among
of convergence
all participants in EWL 7, as having a high level of responsibility and maturity. The
way in which she accommodatesto Claude may be the result of this perception
(EWL 7: 217-8).
All the other people who are thought of as having a high level of responsibility and
maturity, as well as the three South and East-situatedparticipants consideredto have
relatively good language competence, show a high level of convergence markers
to
compared other interactants. Richard, Ana, Sammi and Mala all have the highest
number of convergentturns in their respective conversationswhile Josephand Gauri
are in second place to North/West-situated Mary and Stavros respectively in their
conversations.
While thesehigh-rated participants seem to use convergencesignals to include their
weaker co-interactants,there is a high rate of backchannelling signals at the other
end of the language competence scale. In particular, Ke in EWL 3 and Chat in EWL

7 show a very high amount of backchannelling turns (25 and 28 respectively,
compared to 10,8 and 1 for the other three participants in EWL 3 and 18,11 and 4
for the others in EWL 7. While this may be put down to personal idiosyncrasy, it
may also be the casethat, being perceived of as less competent than the others, they
do their best to soundinvolved and at one with the general run of their conversations.
6. Conclusion
When a comparison is drawn between EWL conversations and conversations among
in homogeneous groups, little in the way of lexical difference can be

speakers
What differences there are seem to do with conversational style:

observed.
in EWL conversations are more co-operative than those in homogeneous

participants
the desire to achieve interactional goals is more intense. Speakers

conversations and
do not make great efforts to achieve overt comity in homogeneous

who
nevertheless do so in EWL ones; these speakers are mainly culturally

conversations
in the North/West or are in any case more mature, responsible or

situated
linguistically competent that other participants.
The discourse of EWL conversationsis therefore characterisedby a leaning towards
involvement style rather than its opposite, but largely because speakers who can
this style do so with more vigour in an international context than

most easily use
they do in a homogeneousone.
The following chapter will attempt to examine further the extent to which
bring with them characteristics of their `home' style when they find
participants
in
themselves cross-cultural contexts.

Spoken English as a World Language: international and intranational
settings
Chapter Nine
Results: six case studies
Chapter 9 216 Results: six casestudies

1. Introduction
The previous chapter highlighted general similarities and differences between EWL
and homogeneousconversations. It was found that there was little difference in
vocabulary use and in the strategic use of convergencemarkers while a noticeable
difference emerged in the way speakerssignalled divergence, more strongly in the
homogeneous conversations than in the EWL ones. The chapter concluded with
for more convergentbehaviour in EWL conversations.

some explanations
This chapter will provide a closer study of six individual speakers who appear in
both EWL conversationsand homogeneousones. Through these case studies it may
be possible to confirm the conclusions drawn in Chapter 8 or indeed to add further
detail to the overall picture.
The seventeenspeakerswho appear in both homogeneousand EWL conversations
are given in Table 13 below:
Speaker EWL Conv. Homog Conv
Betty 7 English
John 9 English
Susy 5 English
Ping 2 Chinese
Fang 10 Chinese
Bai 4 Chinese
Lei 5 Chinese
Milne 6 American
Comfort I Nigerian
Joseph 8 Nigerian
Shray 10 Indian
Greta 7 German
Sofia 6 German
Anne 1 German
Stavros 1 Greek Cypriot
Lina 3 Norwegian
Hedda 6 Norwegian
Table 13: Individual speakers participating in both EWL and homogeneous
conversations
From these seventeen, six speakers have been chosen, who seem to be fairly
representativeof particular groups.

Greta and Bai have in common the fact that they learned English in adolescenceand
in an environment where English has becomevery relevant only fairly recently; they
are very different, however, in that Greta is culturally situated in the North/West
while Bai's culture is consideredto be situated in the constructed East. Greta's other
languages are all Indo-European, while Bai's other language is Chinese. Bai was
considered by her peers to have a relatively low language ability, while Greta's peers
did not make any particular comment in this area. Greta may therefore be said to
have much in common with, obviously, the other German participants and also with
the five Norwegian participants. Less obviously, she shares a great deal with the
Spanish participants and, perhaps, the Greek Cypriots. Bai may clearly be said to
the Chinese (and perhaps the Hong Kong Chinese) participants.

represent
Comfort and Shray both learned English in childhood at the sametime as they were
learning another language or other languages. They can thus be said to have learned
to appropriate English to facets of their identity, or to have developed English-
influenced identities, to a greater extent than have Greta or Bai. Each of their
learning environments was one in which English had historical national relevance.
They could both be said therefore, again to a greater extent than Greta or Bai, to have
entered an English-speaking community within their own country. They are very
different from each other, however, in that Shray is culturally situated in the East
Comfort's education is likely to have been far more Western influenced.

while
Comfort is a woman while Shray is a man. Comfort may be said to representthe
Nigerian speakersand Shray the other Indian speakers.

other
Susy and Milne both learned English in early childhood without learning another
language at the same time. Any further language learning they took part in was in
or later. Their identities and community allegiances are therefore more

adolescence

singly oriented than Comfort's and Shray's. The difference between them is
minimal; like Comfort and Shray, there is a sex difference between them; culturally
speaking, they are both situated in the North/West.
For each pair of speakersthree areas will be considered: the extent and type of
participation in both conversations,use of words and conversationalstyles.
2. Greta and Bai
2.1. Participation and Role
In her homogeneous conversation, Greta's participation is very close to the fair level,
where `fair' is the total number of turns in the conversation, divided by the number
of participants. There are five participants and Greta's turns account for 1 in 5.5 of
turns; the fair level would, naturally, be 1 in 5. Bai, on the other hand, is a less
all
active interactant in the Chinese conversation, contributing only one turn in every
8.5: with six speakers participating, the fair level is therefore 1 in 6.
Both Greta and Bai reduce their level of participation in their respective EWL
conversations, Greta by a small amount - she contributes 1 in 6.2 of all turns, while
Bai rather more dramatically, 1 in 12 turns.
Greta's slightly different participation rate in the two conversationsis paralleled by
one of her co-nationals: Sofia has a similarly slightly reduced participation rate in
her EWL conversation, compared to the homogeneous one. The other German
speaker for whom relevant data exists is Anne, who shows the opposite tendency:
she participates more in EWL 1 than she does in the GE (1 turn in 4.9 comparedto 1
in 9.5). Greta's reducedparticipation rate in EWL 7 may be due to the fact that there
is a `native speaker' in her group but, given that the reduction is very small, may be
of no consequenceat all. The increase in Anne's participation rate may have to do

with the fact that, in her EWL conversation, she is one of the two North/West
participants.
Bai ranks low in both her conversations. In the homogeneousconversation, Fang
comes close to her with one turn in 7.8, while Ping and Lei are both slightly aheadof
the `fair' 1 in 6 level with 1 in 4.8 and 1 in 4.25 respectively. The situation in the
showsthat Bai is the exceptionamongthe Chinesespeakersfor

EWL conversations
whom there is relevant data: Ping and Fang both participate at beyond the `fair' level
in their EWL conversations, while Lei's participation is only just below `fair' level.
Bai's 1 in 12 turns is, by contrast, a sparse result. The other participants in her
conversation are either men (Richard and Kris), North/West situated (Kris) or older
than her (Richard 24, Kris 23, Pallu 22 - Bai is 19). Of all the participants in her
EWL conversation,Bai was the only one rated with low languageproficiency by her
peers.
Where Greta and Bai are concerned,then, it would seem that participation levels in
EWL conversationsmay have to do with sex/gender,cultural situation and perceived
languageproficiency. A summaryof relevant data is given in Table 14 below.
Speaker EWL Conversatio n Homogeneous convers ation

Speaker Turns/ ratio of speaker Speaker Turns/ ratio of speaker
Total turns turns to total Total turns turns to total
turns turns
(fair' ratio) ffair' ratio)
Bai 10/124 1:12.4 1:4 12/102 1:8.5 1:6
Greta 44/274 1: 6.2 1:5 19/105 1.5.5 1:5
Sofia 54/263 1:4.8 1:5 25/105 1:4.2 1:5

Anne 16/78 1:4.9 (1:4) 11/105 1:9.5 1:5
Pin 26/87 1:3.3 (1:4) 21/102 1:4.8 1:6

Fan 73/291 1: 3.9 (1:5) 13/102 1:7.8 1:6
Lei 53/267 1: 5 1:4 24/102 1:4.25 1:6
Table 14 Speaker turns for German and Chinese participants

Turning to the question of role, neither Greta nor Bai takes a leading role in their
respective conversations.It was pointed out in Chapter 8 that, with one exception,
the homogeneousconversationswere leaderless,or had only shadowy leaders.In the
German and Chinese conversations,no leader figure seemsto emerge at all: where
Greta and Bai do not occupy leader-like positions, neither are they particularly
submissive or apparently led. Greta makes two major proposals using an inductive
discoursestrategy,as though she feels she must earn the right to put forward ideas:
42 Gre Well, you know I, I think I would like to come in with something that we should also
43 not forget. I think PR is really important so uh, Khofi Annan coming here is just once
44 in a lifetime thing that will happen. I think we should actually have a film crew here
45 and follow, follow him all the time and make proper videos so that we could use that to
46 market the University and actually bring revenueinto the University and that's
47 somethingthat should be professionalsand I would like about five thousandfor that.
71 Gre Yeah I think that's a good idea becausewe have the new campus,we mustn't forget
72 this so uh, rooms and uh, equipmentthere are already fairly new and I think uh, we can
73 take advantageof all thesethings. So I think we can really cut down the money in these
74 areas,seatings,speakersystemand decoration and then maybe actually spendmoney
75 on, on the film crew. BecauseI honestly think if we creategood marketing material
76 that will again bring money into the University. Students. Bums on seats. And that will
77 bring money.
(GE: 71-77)
Bai's albeit deductive approachto suggestingpainting the buildings is nevertheless
weakly expressed:
19 Bai I think uhm, such as painting as the outside of the building uhm, if they come, they
20 will, it is appearedin their mind firstly-
(CH: 19-20)
In her EWL conversation, EWL 7, Greta adopts a submissive role vis-ä-vis the
emergent leader (Betty) but no more so than the other participants who are all
women and all culturally situated in the North/West. Unlike her stance in the
homogeneous conversation, she seems to assume that the meeting situation gives her
certain speaker's rights which she in

uses order to make points using a deductive
strategy (EWL 7: 8-13) and to challenge,openly, other participants:

209 Gre On the other hand, I mean,uh, to be quite honest,I think he could also deliver, uh, a
210 speechat the banquet.I'm not quite sure that we actually would need a whole-
211 somethingthat should be professionalsand I would like about five thousandfor that.
(EWL 7: 209-211 Seealso EWL 7: 214,279)
This may be because,as has been noted, EWL 7 is the most meeting-like of all the
conversations. In her role-adoption in the EWL conversation, she is, however,
similar to Sofia, who, in her EWL conversation,also usesa deductive strategyto put
forward her proposal (EWL 6: 1-4) and to both Sofia and Anne, who confront other
participants head on:

86 Ann But we've got seven thousand pounds budget; why should be only spend two
87 thousand?
(EWL 1:86-87) Seealso EWL 1: 75)
38 Sof: You don't think that the VIP, important person uhm not has security of his
39 own?
(EWL 6: 38-39 Seealso EWL 6: 127,202-204
Bai, on the other hand, adopts a clearly submissive role in EWL 4. She prefaces her
first turn with `No idea', perhaps as an apology for her suggestion and completes it
with an appeal to the other participants (EWL 4: 69-71). She does make a go of
holding out for her original proposal to spend five thousand pounds (EWL 4: 93,97,
99) but uses an inductive strategy to make her point, suggesting that she does not
feel she has the automatic right to state an opinion but needs to earn it through
reasoning (EWL 4: 108-9). When she does give in, she does so without a whimper:
171 Kris Yeah, the, then I probably should go down five hundred more-
172 Bai How about three thousand, three thousandpounds for security?
173 All @@@
174 Bai It's still a lot?
175 Ric Yeah, it's still a lot.
176 Bai Uhm, I think two thousand.
(EWL 4: 171- 176.
The other Chinese speakers do not necessarily share Bai's general approach to role.
Ping, for example, gets in very early in her EWL conversation with a seemingly
Chapter9 222 Results:six casestudies

confrontational question (EWL 2:2), uses a deductive approachto make her proposal
(EWL 2: 15-18) and does not balk at tackling other participants (EWL 2: 134) albeit
not exactly head on. Lei and Fang also assumethe role of those who have the natural
right to propose (EWL 5:67-72; EWL 10: 29-34) and, although not confrontational,
do stick doggedly to their plans, without making easyconcessions:
267 Lei Yes I know. Uhm, I know, uh, we, we can get internal help to reduce the cost
268 but I still want a very high quality of film recording because,uhm, if it is
269 dealing to the, uh, impression of, of the university, I really need the film is very
270 good. If, if I spentfive thousandelsewherefor advise that thing, or anywhere
271 else,uhm, it's not like, uhm, VIP in the film in our university and I don't want
272 to miss it around or can't find (.. ) where he is or hiding ( )
...
EWL 5: 267-272
64 Fan Yeah but I still, I still want to spendthree thousandfor the trans, translation.
(EWL 10: 64)
Bai and Greta seem, therefore, to adopt different roles in the two different
conversations. Bai's abandonment of a deductive strategy and her relative
submissiveness in the EWL conversation may simply be the result of her feeling
somewhat inferior, coupled with her seemingly natural shadowy role. Greta, on the
other hand, seems to need to carve out her role as proposer in her homogeneous
conversationwhile assumingthe immediate right to her ideas in the EWL one.
2.2 Use of words
The twenty most frequently usedwords for Greta and Bai are as follows:
GRETA BAI
Hom Conv E WL Conv Hom Conv F WL Conv
27i 30_i 8_the 6
19_ 15_a 6_for 6i_think
_we
15_that 14 the 4 only h thousand
13_think 12_ycah 3_i 4-a
13_the 12_till 3_and 4 five
13 and 11_think 3_think 4 security
12_ 11 to 3we 3_pounds
_to
11_be 10+ 2_ as 3_the
10 would 9 _and
on 2 is 3- people

8_you 9 that 2_not 3_is

7_a 8 2_much 3-we
8_
-have
7_money 2_painting 2-no
7_uh _would
7-of 2 uh 2 how
6 6 2 they 2 in
-so
6_should -we
6_it 2_erm 2_because
6 but 5 mean 2 thousand 2 it's
6_could 5-but 2-too 2-lot
5_good 5_two 2 films 2_tlu-cc
5_well 5-an 2_will 2
5 will 5 banquet 2 2-two
very
_you
Table 15 Top twenty most frequent words for Greta and Bai
Many of the words are the same in both the homogeneousand EWL conversations
for the two participants, showing that speakershave a general tendency to use a
similar lexicon in both settings. Where there are differences, these are usually easily
explicable.
Bai's frequent use of `for' and ònly' in the Chineseconversationis accountedfor by
a lengthy false start where the two words are twice repeated (CH: 137-139). Greater
use of à' in the EWL conversationis accountedfor by the simple fact that Bai needs
to refer to singular countable nouns in that context. It may also be an indicator of
lack of shared knowledge. The use of `painting' in the homogeneous conversation is
paralleled by `security' in the EWL one: they represent Bai's spending proposal in
each case. The only curiosity seems to be in the use of `pounds' in the EWL
conversation,or rather the non-useof the item in the homogeneousone. Bai seemsto
be following, and perhaps accommodating to, other participants in EWL
Conversation4 by using `pounds' in monetary expressions.The other participants in
the Chinese conversation also make extensive use of `pounds', but Bai choosesnot
to follow them or to accommodate to them. It is possible that her lack of
accommodation in the homogeneousconversation is at one with her more positive
role there, where her perceived weakness in the EWL conversation leads her to
imitate the other participants.

Greta carries over an extensive use of `would' from her homogeneousconversation
to the EWL one but does not parallel this with similar extensive use of `could' and
`should'. It has already been noted that, if anything, she is more tentative among her
national peers than in an international context and her reduced use of the more
tentative modal verbs seemsto confirm this. Against this, she makes frequent use of
`yeah' in her EWL conversation - twelve instances which, combined with two
instancesof `yes' comparesstrikingly with only four instancesof `yeah' (and none
of `yes') in the German conversation. This would seem to indicate greater
disposition towards convergencein the EWL conversation.
She also carries over from her homogeneousconversation to the EWL one frequent
the expression Ì mean' (four instances in GE, five in EWL 5).

use of
Finally, like Bai, she makes fairly frequent use, in the EWL conversation, of the
word connectedto her budget proposal ('banquet') but makes less frequent use of the
equivalent word ('film') in her homogeneous conversation.
The type of vocabulary used by Greta ad by Bai is very similar indeed, as is shown
in Table 16.
EWL HOMOG
Greta Bai Greta Bai
Kl Words (1 to 81.72% 85.29% 86.10% 84.85%
1000):
K2 Words (1001 to 3.74% 2.94% 2.67% 3.03%
2000):
AWL Words 3.30% 3.92% 3.62% 3.54%
(academic):
Off-List Words: 11.23% 7.84% 7.62% 8.59%
Table 16 Greta and Bai's words by type
As can be seen,the use of different types of word is very similar in the two types of
conversation. The only seemingly outstanding figure is Greta's 11.23% of òff-list'
words in her EWL conversation.
Chapter9 225 Results: six casestudies

Examining this òff-list' in comparison to Greta's òff-list' of words from her
homogeneous conversation, little emerges which is worthy of note. In both
conversations, Greta uses òff-list' words in order to talk about specific budgetary
proposals (7 words in each conversation),and to refer to the University or to the VIP
visit in general(3 words in eachconversation). Some òff-list' word use is accounted
for by false starts and hesitations: there are slightly more of these in the EWL
conversation than in the homogeneousone. Greta coins a word èxtravagandic' in
her EWL conversation and uses the word `seating' in a plural form, twice, in the
homogeneousconversation. Both these items are òff-list'. Her use of `yeah' and
òk' is also òff-list' and, has been mentioned, there is more of this in the EWL
conversationthan there is in the homogeneous.
What is left is of some interest: in the EWL conversation, two of the remaining 'off-
list' words are marked for formality: `calibre' and `liaise' while the two remaining
òff-list' words in the homogeneous conversation are, conversely, marked for
informality: `bums' and `dump'.
If this evidence can be consideredweighty enough, Greta's only concessionto the
EWL setting is to abandon informal language and to increase the formality of her
speech, while increasing the amount of hesitation words. At all events, it is a
performance not repeated by either of the other two German interactants in EWL
conversationswho also participate in the German homogeneousconversation. Both
Anne and Sofia use òff-list words only to deal with proposal or university specific
issues, to hesitate or in false starts, to agree using `yeah' and òk' and for one
coinage, in the same mould as `seatings'. Beyond this, Sofia uses two words on the
formal side, `surroundings' and `negotiate' in the homogeneousconversation.

2.3. Conversational style
Consistent with the results reported in the previous two sections, Greta's overall
performance is generally more marked than Bai's. The latter speaker's low level of
participation means that it is difficult to identify trends in her conversational style.
The only possibly relevant fact is that her style is more marked for both convergence
and divergence in her homogeneous conversation than in her EWL one: normalised
for the two conversations are presented in Figure 6 below:

results
0 0
0 13
Bai: Convergence markers in Homogeneous and Bai: Divergence markers in homogeneous and
EWL conversations E\VL conversations
Fig. 6 Bai - conversational style markers
The slightly higher rate of convergence markers in the EWL conversation is
to other speakers' performances but cannot be taken as particularly

comparable
here. The higher level of both sorts of marking in her homogeneous

meaningful
may be the result of her feeling more comfortable in that situation.

conversation
Bai does not appear to be typical of the Chinese speakers. The other three members
homogeneous Chinese group who are also speakers in EWL conversations

of the
have, after normalisation, more marked utterances in the latter speech events than in
the former. Both Ping and Lei use considerably more convergence markers in their
EWL conversation than they do in their homogeneous one, while for hang it is the
reverse. When it comes to divergence makers, the situation is much more systematic,
Chapter 9 227 Results: six case studies

with all three using a more divergent style with their national peers than they do in
the international setting. A summary of Chinese participants' convergence and
divergence marking is given in Table 17 below.
Name Convergence Markers Divergence Markers

EWL Homog Conv EWL Homog Cony
Conv. Cony
Bai 2 2.5 1 2.5
Ping 20 7.4 0 10
Fang 13 33 4 5.6
Lei 33 13 7 20
Table 17 Chinese speakers - conversational style markers
Turning to Greta, she also goes slightly against the general trend in that the number
of convergence markers in her homogeneous conversation is, after nornmalisation,
more than the number in her EWL one. She is closer to the general trend in her use
of divergence markers which are far more frequent in her homogeneous conversation
than in her EWL is

as made clear in Figure 7.
IO HOMOG O HOI
13EWL 13EIP
Greta: Divergence markers in Ilomogeneous and

E\VL conversations
Greta: Convergence markers in homogeneous and
EWL conversations
Fig. 7 Greta - conversational style markers
A possible explanation for her slightly lower than expected use of convergence in
her EWL conversation may be that EWL 7 is among speakers who are all from the
North/West, it is a well-chaired meeting which, as has been stated, runs more along
traditional meeting lines than any of the other EWL conversations, and thus, perhaps,
Chapter 9 228 Results:six casestudies

accords Greta (and the other speakers)the right to express ideas without having to
make many efforts at establishingand maintaining comity.
The other two German speakerswho also participate in EWL conversationsshow a
similar but less dramatic profile. They both have slightly more convergencemarkers
in their homogeneousconversation than in their EWL ones; Sofia, like Greta, uses
far more divergencemarking in her homogeneousconversation than she does in her
EWL one, whereas Anne does the reverse, using many divergence signals in her
EWL conversation. Table 18 provides a summary.

EWL Homog Conv EWL Homog Conv
Conv. Conv
Greta 23 32 4 34
Anne 13 16 7 1.5
Sofia 27 28 8 17
Table 18 German speakers - conversational style markers
2.4 Greta and Bai - Conclusion
These two speakers show personal idiosyncrasies but nevertheless confirm some
patterns. They both participate slightly less in their EWL conversationsthan they do
in their homogeneous ones; no particular pattern emerges by comparing their
performance to others from their homogeneousgroups: of the other two Germans
under consideration, one participates less and the other more in her EWL
conversation; the other three Chinese speakers all participate fully in both
conversations.
While Greta seemsto switch strategy from one conversation to the other, Bai uses a
similar strategy in both.

Both Greta and Bai use a very similar lexicon in their homogeneousconversationto
the one they use in their EWL one. Interestingly, Greta uses more tentative
expressionsbut fewer expressionsof agreementin her homogeneousconversation.
When it comes to conversational style, nearly all German and Chinese participants
make substantially greater use of discourse marking in their homogeneous
conversationsthan they do in their EWL ones and all save one use noticeably more
divergencemarking in their homogeneousconversationsthan they do in their EWL
ones.
3 Comfort and Shray
Unlike for Bai and Greta, there is little or no robust comparative information
available for Comfort and Shray: among Comfort's co-nationals, there is only one,
Joseph, who is a participant in both the homogeneous, Nigerian conversation and an
EWL one. The other Indian participants in EWL conversations did not take part in
the homogeneous Indian conversation.
3.1. Participation and Role
Comfort is a more prolific participant in her homogeneousconversation (NI) than
she is in her EWL one. In the former, she contributes a total of 56 turns out of the
171 for the whole conversation, a rate of one turn in three, where the fair proportion
would be one in four. She clearly feels at home among her co-nationals. In EWL 1,
on the other hand, her turns number 11 out of a conversation total of 78, making the
ratio one in seven, with the fair proportion, again, being one turn in four. Joseph

makes a similar contribution to Comfort's in NI but, unlike Comfort, contributes at
the fair level in his EWL conversation.
Shray's participation in both conversations is very close to the fair proportion: he
contributes 62 of the 291 turns in his EWL conversation, a ratio of one to four point
six, where the fair ratio would be one to five, and 49 of the 171 turns in his
homogeneousconversation, or one turn in three point four, where the fair ratio
would be one in three. Table 19 provides a summary for Comfort, Josephand Shray.
Speaker EWL Conversation Homogeneous conversation

turns turns
ffair' ratio) 'fair' ratio)
Comfort 11/78 1:7 1:4 56/171 1:3 1:4
Joseph 26/116 1:4.4 4 49/171 1:3.4 4
Shra 62/291 1:4.6 1:5 17:59 1.3.4 1:3
Table 19 Speaker turns for Comfort, Shray and Joseph
Comfort's relatively low participation rate in her EWL conversation may be put
down to several different factors: she is, with Gauri, culturally situated in the South,
where two of the other participants are North/West (Stavros and Anne). She is also a
woman in a conversation dominated by a man (Stavros). Gauri, however, also
situated in the `South', as well as the Èast', also a woman, has a participation rate
much closer to the fair one. The culture and gender explanations do not hold
particularly well. What may be more relevant is Comfort's perceived languagelevel:
where Stavros and Gauri are both accordedhigh levels of languageskill and maturity
by their peers in the EWL conversation,Comfort is consideredto have a lower level
of language ability than anyone else. If this perception somehow made itself
apparentin the conversation (and there is nothing very evident to that effect) then it
might account for Comfort's relative reticence.

Turning to the question of role, it has already been noted (Chapter 8, Section 5.3.1)
that EWL Conversation 1 is clearly led by Stavros who establishesa certain tone to
the meeting which Comfort seemsleast willing to accommodateto. She enters the
conversation in a very forthright way, giving the impression that she has the perfect
right to make her point, using a deductive discourse strategy. Further, she carries
over the somewhatrhetorical style she uses in her homogeneousconversation,where
she also introduces her proposal with a deductive strategy, making considerableuse
of emphatic repetition. If sheusesthis strategy and this style designedly to imply she
has an upper hand, it does not seem to work and she seems forced to leave the
conversation for a while (EWL 1:26). When she re-enters (EWL 1:124) she uses the
sameapproachand, again, doesnot make her mark.
Comparison with Joseph shows some similar tendencies: like Comfort, Joseph
carries over a markedly confrontational role from the Nigerian conversation to his
EWL one. He enters the EWL conversation, for example, by explicitly ridiculing
another interactant's idea:
6 Jos I think spending such money on security would be ridiculous.

(EWL 8: 6)
Unlike Comfort, however, he uses a more inductive discourse strategy in both
conversationsto present his own ideas and unlike Comfort, he holds his own in both
the EWL conversationand the homogeneousone. His gender may have something to
do with it, as might the perception of his language competenceand maturity, both
rated highly by the other participants in EWL 8.
Shray, in contrast to Comfort, is a dominant force in his EWL conversation,which, it
will be remembered, is unique among the EWL conversations in that all participants

are South/Eastsituated. He is one of two men in the conversation, but has a lower
perceived languageability than the other man, Sarraj.
His domination, however, stems from the fact that he succeedsin channelling other
participants' proposalsthrough himself and trying to knock them down. He does not
presenthis own proposal properly until near the end of the conversation,having built
up the right to do so through his previous argumentation. This strategy is a little
similar to his inductive approach in the Indian conversation where he is careful to
presentreasonsbefore declaring his spendingwishes.
Other Indian participants in EWL conversations also seem to achieve successby
constructing the right to their ideas though the use of inductive strategy. In EWL 1,
Gauri never seeks to wrest the leadership role from Stavros, but nevertheless
succeeds in achieving her budgetary aim and continuing to participate fully in the
proceedings; Sarraj, though less dominant than Shray in EWL 10, nevertheless
succeeds in making his point and holding on to his monetary request, having made
the point through an inductive process.
3.2. Use of Words
Comfort and Shray's twenty most frequent words in their two conversationsare
given in Table 20 below.
COMFORT SHRAY
Hom Conv EWL Conv Hom Conv EWL Conv
50you 24_to 15_the 23 uh
35_to 17 he 11 security 22- i
24_we 13_be 10 to 22_you
20_i _i
12_ 8 that 16 to
18_need 11_is _
7so 14 can
17_that 7 like 6-person 14 acid
16 it 7 the 6_will 14_ycah
16 know 7-thousand 6_is 13_a
16_the 6 just 6i 12 the
14 don't 6i and 5 our 12_that

14_are 6 comfortable 5_think 12_for

13_so 6_for 5_vip 12-is
11_for 5-people 5_we 11 it
10_be 5conference 5_what 11think
10_have 5- has 4_on 1U_not
10_and 5 needs 4_be 10 thousand
10_is 5- going 4_for 9_how
9_much 5 well 4_percent 9 it's
4- like -
5 it 4_do -
9-be
9just 4_have 4_university 8-don't
Table 20 Top twenty most frequent words for Comfort and Shray
Similarly to Greta, most of Comfort's high-frequency words in her EWL
conversation are the same as most of those in her homogeneous one. Shray is similar
to Bai in that he carries over fewer of his words from the one setting to the other. In
their choices of words used in EWL conversations,the two participants in question
point up some interesting issues.
Comfort uses `you' fifty times in her homogeneousconversation but only twice in
her EWL one. Even after turn-basednormalisation, there are still only 10.1 uses of
the word. It may be that Comfort, feeling more at ease with her co-nationals, decides
not to be confrontational with her EWL co-participants. This squares with the
general pattern of greater convergencein the EWL conversations. Conversely, she
uses `he' far more frequently in the EWL conversation than she does in the
homogeneousone.
Some of the frequent words in her EWL conversation can be accounted for by
Comfort's use of rhetorical repetition, already referred to. She repeats`comfortable',
`conference' `needs' and `going' severaltimes in the turns in which they appear(e.g.
EWL 1:14-21,124).
Like many other speakers,she tends to use the word `thousand' more frequently in
the EWL conversationthan she doesin the homogeneousone.
Shray uses the word `security' with great frequency in his homogeneous
conversation where, indeed, it is central to his proposal for spending money.

Interestingly, it is also his proposal in his EWL conversation but, in that context, he
makes comparatively sparseuse of the word. This seemsto be becausehe usesmore
of his turns in his EWL conversationto argue against other participants than he does
to argue for his own proposal. Many of the frequent words unique to his EWL
conversation are indices of a more interactive approach, words like `yeah', `how',
`don't' and `you'. The fact that his homogeneous conversation is much more
managed (by Sukvinder) than his EWL conversation may well account for the
greater amount of interaction markers in the latter. It may further be accountedfor by
the generalidea emerging from the data of more convergencein EWL conversations
than in homogeneousones.
Yet again, like other participants, he makes greater use of the word `thousand' in his
EWL conversation.
As far as word types are concerned, it is noticeable that Comfort uses almost
precisely the same amount of words per range in her EWL conversation as she does
in her homogeneousone, while Shray shows some substantial differences. Table 21
provides a summary
EWL HOMOG
Comfort Shray Comfort Shray
KI Words (1 to 1000): 88.25% 84.34% 89.05% 85.35%
K2 Words (1001 to 2000): 4.87% 2.01% 4.83% 1.10%
AWL Words academic : 2.01% 2.16% 1.77% 6.59%
Off-List Words: 4.87% 11.49% 4.36% 6.96%
Table 21 Comfort and Shray's words by type
As can be seen from the table, Shray uses more K2 words in his EWL conversation
than he does in his homogeneousone and more AWL and òff-list' words in the
latter than in the former.
His greater use of K2 words in the EWL conversation can be accounted for, again,
by the fact that he tackles the other participants about their proposals, where in his

homogeneousconversation, he insists mainly on his own proposal. There are no
proposal or event-specific words among the K2 group in his homogeneous
conversation;thesewords count for all the K2 words in his EWL conversation.
The same explanation accounts for the greater number of AWL words in his
homogeneousconversation: `security' is counted as an AWL word, where cameras,
frame, translation, for example, are `K2' words. The greater number of òff-list'
words in Shray's homogeneousconversationcould be accountedfor in the sameway
as Greta's slightly higher number in her homogeneousconversation. Shray's òff
list' words in his EWL conversation is swelled by a large number of false starts,
hesitations and the agreementwords `yeah' and 'OK'. Other òff-list' words are,
again, connected with proposals other than Shray's.
3.3 Conversational Style
Comfort and Shray seem to have opposite approaches to conversational style
inasmuch as Comfort marks her EWL conversation for both convergence and
divergence more than she marks her homogeneous conversation whereas Shray
marks his homogeneousconversation more than he does his EWL one, as is made
clear in Figures 8 and 9 below.

Q HOM
0
13EWL 0
Comfort: Convergence markers in Homogeneous Comfort: Divergence markers in Homogeneous and

and t\\ L conversations EWW'Lconversations
Figure 8 Comfort - conversational style markers
Q HOM JO HOM
DEWL MEWL
Shray: Convergence markers in Homogeneous and Shray: Divergence markers in Homogeneous and E\ 't.
EWVL conversations
conversations
Fig. 9 Shray - conversational style markers
It is clear from Figure 8 that Comfort makes use of many convergence markers in
both her EWL and her homogeneous conversations and, consistent with the majority
of participants uses more convergence marking when in the international setting than
she does when with her co-nationals. Where Comfort seems to differ from most
other participants is that she also uses many divergence markers in her EWL
conversation, more than she does in the homogeneous setting. The only other
Nigerian speaker who is a participant in both types of conversation is Joseph, who

conforms to the mainstreampattern with more convergencemarkers than divergence
in his EWL conversation and vice versa in his homogeneous.Table 22 makes the
difference clear betweenComfort and Joseph.

Conv. Conv
Comfort 10 5.1 11 5.6
Joseph 15 10 3 8.5
Table 22 A comparison of Comfort's and Joseph's use of convergence and

divergence markers
There is no obvious explanation as to why Comfort decides to give greater marking
to her EWL it
conversation: seemsthat she feels she needsto make extra efforts both
to establish comity (as with all other participants) as well as to make her point clear
despite interactional goals. Her perceived relatively low language competencemay
be a factor here. What is clear is that she is apparentlythe least successfulparticipant
in her EWL conversation.
Shray's opposite profile posesa different puzzle. While his greater use of divergence
when among his national peersis consistentwith other participants, his greateruse of
convergence markers in the same situation is not. There do not appear to be any co-
ordinate factors to explain this and, unfortunately, no other member of the Indian
homogeneousgroup is a participant in an EWL conversation. A rough comparison
may be drawn, however, by looking at the other Indian speakersin either EWL or
homogeneousconversations.
After normalisation, the number of convergencemarkers used by Gauri, Pallu and
Sarraj in EWL conversationsis consistently higher than the number used by Shray,
while the number of divergencemarkers used is roughly the same for all four people.
Shray does, then, seem to fall out of the norm, where his co-nationals are concerned.

The convergence and divergence markers used by Shray in his homogeneous
conversation are similar to those of his co-nationals. Any inferences regarding a
national tendency to use a more or less equal number of convergencemarkers as
divergence ones would, obviously, require some wide-ranging research in its own
right. For the moment, it seems sufficient to say that Shray's performance in his
homogeneous conversation seems normal, where his interaction in his EWL
conversationis deviant. Table 23 provides a summary.

Conv. Conv
Shray 9 28.9 5 36.34
Gauri 24 5.5
Pallu 16.5 4.1
Sarra 20 8
Sukvinder 29 14.5
Pravin 34.8 29
Table 23 Indian or Hindi speakers - conversational style markers
3.4. Comfort and Shray - Conclusion
Despite similar English language backgrounds, Comfort and Shray are quite
different in their approachesto their two conversations.
Comfort seemsquite at easein her homogeneousconversation where she addresses
others head on and achievesa high level of participation. In her EWL conversation,
however, she is not very successful,seeming to be unwilling, or perhaps unable to
participate on the same footing as the other participants. Her perceived low language
level has been offered as a possible explanation for this but, at all events, she seems
to be trying too hard to achieve both interactional and transactional goals and, as a
result, becomes a background figure in the conversation. Interestingly, she does not
make any real changes to her overall approach when she shifts from the
homogeneousconversationto the EWL one.

Shray, on the other hand, seems almost the opposite: it is true that he is a fair
participant in both conversationsbut really seemsto come into his own in his EWL
conversation where he takes a leadership role and is generally more interactive.
Despite not carrying over his more convergent approach from his homogeneous
conversationto his EWL one, he is neverthelessa successfulinteractant in the latter.
4. Betty and Milne
Betty and Milne both speakin a homogeneousand an EWL conversation, but while
Betty may be comparedto two of her English co-nationals (Susy and John) who also
participate in both types of conversation, the other American participants do not
figure in any of the EWL conversations. The extent to which Milne is typical,
therefore, will be difficult to evaluate.
4.1 Participation and role
In their homogeneousconversations,both Betty and Milne participate with a number
of turns equal to the fair norm. Betty has 75 turns out of the conversation total of
409, giving a ratio of 1 to 5.4. With five interactants in the conversation, she is
therefore very close to the fair norm of one to five. Milne contributes 59 of the total
of 285 turns in the American conversation, a ratio of 1:4.8. There are four
participants in this homogeneousconversation so, like Betty, Milne is close to the
fair ratio of one to four.
When it comesto the EWL conversations,the situation is completely different. Betty
takes twice as many turns as might be considered fair: she contributes 104 out of the
total of 274 turns in the conversation, a ratio of 1:2.6, where the fair ratio is one to

five. As has been noted (See Chapter 8, Section 5.3.1) she takes a leadership role in
her EWL conversation and, as such, channels all other participants' turns through
her. The reason why the other participants allow her this position may have to do
with the perception among her peers of her high level of language competenceand
maturity. Of the other two English speakers who participate in both types of
conversation, John also increases his rate of participation in the EWL setting
compared to the homogeneousone while Susy does not; both of them are only just
over the fair participation ratio, as is made clear in Table 24 below. John, like Betty,
emergesas a leader in his EWL conversation,is one of only two North/West situated
speakersand is accorded a high level of language competenceand maturity by his
peers. Susy is not a leader in her EWL conversation, is the only North/West situated
participant and is accordedthe samehigh level of languagecompetenceand maturity
as two out of the other three participants. Betty's very high participation rate may
therefore be put down to purely personal,individual reasons.
Speaker EWL Conversation Homogeneous conversation

turns turns
ffair' ratio) 'fair' ratio
Betty 104/274 1:2.6 1:5 75/409 1:5.4 1:5
Milne 30/263 1:8.7 1:5 59/285 1:4.8 1:4
Susy 67/267 1:3.9 (1:4) 86/409 1:4.7 (1:5)

John 40/152 1:3.8 (1:5) 63/409 1:6.4 1:5
Table 24 Speaker turns for Milne and English participants
In direct contrast with Betty, Milne is a relatively infrequent contributor to his EWL
conversation: with only 30 turns out of 263, he contributes only one turn for every
8.7, where the fair ratio is 1 to 5. One obvious reason for this low level of
participation is the fact that he entersthe conversation relatively late on. This may be
because, as was noted in Chapter 7, he sees himself as a negotiating expert and tries

to assesswhat is going on before putting his spoke in. He is considered by the others
to have a high level of languagecompetenceand he may therefore be reticent so as
to give others more of a chance to speak. It is unfortunate that there are no other
American speakersinvolved in EWL conversations: Milne's performance may or
may not be typical.
While Betty is self-appointed leader in her EWL conversation, she is not a leader in
her homogeneousone, where Ian seemsto take that role, albeit in a weak way. Betty
does carry over, from her homogeneousconversationto her EWL one, her somewhat
inductive approachto making her own proposal:
53 Bet That should come under the University budget anyway. We, we're discussing
54 a budget for our visitor and what I'd like to seewas a bi, would be to have
55 somekind of uhm, acknowledgementof the event in terms of the history of
56 the University. Perhapswe could film the event and have, so that it was,
57 would be a record for, you know, part of the showcase idea; a record for
58 future visitors and students.We could put it on our website, show them how
esteemedwe are
32 Bet Mm, yeah. Well, yes, I mean th, that, that's, all these proposals are very nice
33 and very grand, but if you look around you at the university at the moment,
34 it's looking pretty shabby and, you know, it's no good having a wonderful
35 banquet filming when, you know, he's going to be walking past huts with the
36 paint peeling off and dirty carpets with chewing gum on them. Uhm, you
37 know, look over there @@@
38 Anj @@@
39 Bet at that wall, for example.So I think there has to be some money set aside for,
40 repairs, painting, uhm, new furniture, so that, you know, what we're filming
41 and what is being seenwhen they are coming for the banquet and the
42 conference isn't really shame-making. You know, it's no use getting, raising
43 if
our profile what people see in this profile and on the filmed record is, you
44 know, peeling walls and broken chairs. So, you know, I would like some of
45 this budget set asidejust for basic maintenance,which then we will still have
46 and, you know, for whoever else visits and for us and our students.
Despite casting herself in the leader role she does not, then, automatically assume
the right to put forward her ideas with a deductive strategy. A more obvious
concessionto her different role in her EWL conversation is her abandonmentof the

divergent conversational style: she seemsto see herself as very conciliatory in her
EWL conversation,where she is confrontational in her homogeneousone.
John is a little similar to Betty in that he doesnot createa leadershiprole for himself
in his homogeneousconversationwhile he seemsto do so in his EWL one. His role
change is, however, less pronounced than Betty's: like Betty, he transfers from the
homogeneous conversation to the EWL one his chosen discourse strategy for
presenting his own proposal, which is somewhat deductive in both cases, but unlike
Betty he seems to be fairly easy-going and conciliatory in both conversations and he
noticeably does not want, or allow other participants' turns to be channelled through
him. Susy adopts, in part, a similar role change to Betty's: she does not take a
leadership role in her EWL conversation but she does become far more obviously
supportive of other participants, with a large number of backchannelling signals and
far more agreement expressed than in her homogeneous conversation.
Milne doesnot take a leadershiprole in the leaderlessAmerican conversation and his
the èxpert negotiator' role in his EWL conversation has already been
adoption of
to Chapter 7). His role in his homogeneous conversation is, however,

referred
remarkably similar to the one adopted in his EWL conversation: he comes across as
more of a seasoned negotiator in the EWL conversation because the other
do not, as he does, evaluate each other's proposals in the same objective

participants
way. Milne proceeds in a very similar fashion in his homogeneous conversation
his role is less prominent because the other participants exhibit similar
where
behaviour.

4.2 Use of words
Betty and Milne, like most participants, have a similar word-frequency count in both
their conversations. Of Milne's twenty most frequent words in his homogeneous
conversation, nine are also most frequent in his EWL one; of Betty's top twenty
words from her homogeneousconversation,twelve are among the top twenty in her
EWL one. The top twenty words in all four conversations are given in Table 25
below.
BE TTY MILNE
Homog EWL Homog EWL
52 the 52-YOU 26 to "10 the
49i 41 to -
24 we 20_to
39_to 41_i 19_a 17_and
34 we 40 the 16 it 17-we
30_it 36_wc -
16 the 13-thousand
25_of 33_and 13_you 12_it
24_you 32_for 12_so 10_that
23_a 30_a 11-be 8_a
23_and 30_ycs 11_yeah 8_havc
21_that 27 that 9-need 6-two
18 think 25_so 9-but 5_presentation
15_in 25_know 8_of 5-be
15_have 23_of 8 that Si
15_be 21_no 8_in _
5_need
13_university 21_be 7_because 5_can
11_my 21_it 7_get 5_what
11_what 20yeah 7_i 4_bottom
11_is 19-well 7_know 4_do
11_for 17_mcan 7_would 4
10 don't 15 Toing 7 it's _maybe
4 on
Table 25 Top twenty most frequent words for Betty and Milne
The most striking set of words appearingin Betty's EWL top twenty but absentfrom
the equivalent homogeneous list are `yes', `yeah' and `no' with 30,20 and 21
occurrences each. Betty uses these words once, four times and four times
respectively in her homogeneousconversation. Even after normalisation, the figures
are still low at 1.38,5.54 and 5.54. This disparity is at one, however, with Betty's

change of role from one conversation to the other and also at one with the general
result whereby EWL conversations are generally more convergent than
homogeneousones. Incidentally, where Betty uses `no' in her EWL conversation, it
is mostly to agree with other participants (EWL 7: 42,73,81,117,125,129,132,
160,215,242,347). Her greater use of `so' in her EWL conversation also fits with
her changedrole since, as leader, she frequently uses `so' in order to structure the
bringing other participants' contributions into a meaingful whole:

conversation,
112 Cla then, it should be re-negotiatedwith Building and Estatesso that their current,
113 ongoing programme of rebuil, refurbishment and the set-asidebudget they
114 must have-
115 Bet So, so
116 Cla can be brought in.
117 Bet Yeah. No. That's a good point.
(EWL 7: 115) Seealso EWL 7: 120,121,212,249,281,294,297,301,321,322,
331,339,349,350,356,360,371).
Similarly the comparatively frequent use of Ì mean' and `you know' in her EWL
conversation seems to be part of her general, conciliatory approach there, with all
save one instances of `mean' being accounted for as a softening device and `you
know' being used extensively as a convergence marker:
32 Bet Mm, yeah. Well, yes, I mean th, that, that's, all theseproposals are very nice
(EWL 7: 32. Seealso EWL 7: 54,84,88,91,94,150,177,178,183,198,243,254,
258,261)
Taking the eleven very frequent words used by Milne in his EWL conversation but
in his homogeneousone, at least four are actually also fairly frequent in the
not
latter, although not appearing in the top twenty ('have', `can', `what', 'do'). Of the
two have to do with Milne's budget proposal ('two', `presentation') while

remaining,
one ('bottom') is frequent partly becauseof some repetition over the misunderstood
expression `bottom line' (6: 205-208). The relatively high frequency of `maybe' in
his EWL conversation can be explained by the general tendency to emphasise
in this casewith a hedging word and the comparatively frequent use of

convergence,

`thousand' in other EWL conversations has already been remarked on. Milne's
frequent use of ànd' in his EWL conversationmust remain an unexplained curiosity,
while the comparatively low frequency of `yeah'/`yes' is at variance with the
tendencytowards co-operativeness.
When it comes to type of word, Betty is fairly consistent across the two
conversations, with only slight or even negligible variations in the given fields.
Milne is also fairly consistent except that he uses a noticeably higher percentageof
K2 words in his homogeneousconversation.Table 26 provides a summary.
EWL HOMOG
Betty Milne Betty Milne
K1 Words(1 to 1000): 85.43% 90.81% 86.04% 85.37%
K2 Words (1001to 2000): 3.18% 2.43% 2.78% 5.85%
AWL Words(academic): 2.42% 1.89% 3.30% 2.19%
Off-ListWords: 8.98% 4.86% 7.89% 6.58%
Table 26 Betty and Milne's words by type
When the actual words are examined it is clear, however, that Milne is not using a
much richer vocabulary in his homogenousconversationthan he deploys in his EWL
one. Twenty-two of the K2 words he uses in the American conversation are specific
to the budget proposals (`cameras' 5, `clean/ing' 6, `cook' 2, `film/ing' 6, `wine' 3).
A further four may be paralleled by two K2 words in his EWL conversation
(òriginally', èxactly', `split', `worried' - compared to àttend', òpposed'). The
remaining three K2 words in Milne's homogeneous conversation are more
interesting. They are `bit', `stuff and `lot', all three of which suggest a level of
informality which is not presentin his EWL words.

4.3 Conversational style
It has already been noted that Betty changes her role in the two conversations and
that this entails, at least in part, a change of conversational style. In this respect, she
fits the general pattern in the majority of conversations: she marks her turns for
convergence more in her EWL conversation than she does in her homogeneous one
and does the opposite in marking turns for divergence, as Figure 10 makes clear.
Fig. 10 Betty - conversational style markers
Milne falls slightly outside the general pattern in that he seems to mark his
conversational turns for both convergence and divergence evenly across both
conversations as Figure 11 makes clear.
Figure 11 Milne
- conversational style markers

Betty's extensive use of convergencemarkers in her EWL conversation has already
been referred to in the context of a discussion of her leadership position. Her
keennessis, however, matched by another of the English speakers: Susy uses a
similar number of markers in her EWL conversation. The other English speaker,
John, is a much less enthusiasticuser of convergencemarking, perhaps for gender
reasons.John also marks his turns for divergence to a lesser extent than his two
female co-nationals, but all three follow the trend of marking their homogeneous
conversation turns more for divergence than they do their EWL conversationsand
vice versa with convergencemarking. The complete figures are given in Table 27.

Conv. Conv
Betty 55 34.64 5 23.48
Sus 52 13.2 14 16.3
John 4 4.4 2 10.7
Table 27 English speakers - conversational style markers
Milne's conversationalstyle seemsunchangedin the two types of conversation and,
as has been noted, it is impossible to examine the extent to which this is a trait he
might sharewith his co-nationals.The existent data does,however, show that his use
of marking in his homogeneousconversation is fairly consistent with that of his
peers. In the American conversation, all participants use a (normalised) number of
divergence markers between 4.8 and 7 while three out of four of them use a
(normalised) number of convergence markers between 12.5 and 17. Sindy is
exceptional, with a very high convergence marker rating, largely because of her
persistent, almost tic-like use of backchanneling words. Full figures are reported in
Table 28.


Conv. Conv
Milne 10 12.5 6 6
Sindy 43 4.8
Dolores 14 7
Candice 17 6.3
Table 28: American speakers - conversational style markers
All that can be said without further data is that Milne's consistencyacrossboth types
of conversation is in line with his similar consistency when it comes to role. In the
light of the general result, reported in Chapter 8, showing that the American group
differed from the others in maintaining a high level of convergence in its
conversation, compared to its corresponding level of divergence, it would be
extremely interesting to find out the extent to which Milne's behaviour is normal
and, therefore, the extent to which EWL conversationsare influenced by American
conversationalstyle.
4.4 Betty and Milne - Conclusion
Betty and Milne present somewhat different profiles in that Betty conforms to and
exceedsby far the norm by which EWL conversationsshow more convergencethan
homogeneous ones. Perhaps more strikingly, Betty's role-change from one
conversationto the is
other quite dramatic but in this respect it seemsclear that she is
not particularly typical. Milne, on the other hand, seemsto make very few changesat
all from one conversation to the other, with the noticeable exception of his
participation rate, falling considerably in his EWL conversation. It may seem
paradoxical to ascribe his relatively few turns in EWL 6 to his perceived high
languagecompetencewhen the reverse was suggestedin the case of Comfort. Lack
of familiarity with international settings may be another factor, which might also

explain Milne's tendency not to make changesto his style and role and not to enter
the EWL conversationuntil comparatively late on.
5 Conclusion
This study of six individual participants has shown that within general tendencies
towards differences between intercultural and intracultural conversations, there are
also many irregularities: some speakerscarry over their vocabulary, style and role
from one type of conversationto the other with almost no changewhile others make
marked, almost dramatic The

changes. use of `culture' (in inverted commas because
of the inevitable stereotyping associatedwith North/West and South/East cultural
constructs)and speakergender to explain tendenciesmay be of some interest, while
more individual traits, such as the perceived level of maturity or language
competence seem to be more relevant.
In terms of general tendencies,the six case studies do show that, where participants
do make changes,the changeshave them being more interactive, more attentive to
each other and more convergent in their EWL conversations than they are in their
homogeneousones.

Chapter Ten
Conclusion
Chapter 10 251 Conclusion

1. Summary
This thesis has set out to characteriseEnglish as it is spoken across cultural and
national boundaries.The aim was to add to the ongoing body of researchconcerning
itself with language in a changed world, where the traditional boundaries between
English spokenby and taught to native speakers,and English as a secondor foreign
languagehave been breaking down. At the outset, the notion of àn' English, spoken
world-wide, was problematised and, at a theoretical level at least, found to be less
than one hundred percent tenable. It was decided that the objective of the thesis
could not be that of identifying an international or intercultural variety of English
and the focus therefore shifted to the question of how English, irrespective of variety,
is used in international settings. The two labels often associatedwith such use, EIL
and ELF, were examined and found to be slightly wanting and, as a result, the EWL
label was chosen to represent English used by any speakers whatsoever, whether so-
called native speakers,non-native speakersor anyone in-between, in extra-national
settings. In order to typify English used in these settings, it was decided to compare
international conversationswith conversationsamong co-nationals.
Previous research in the areas of EIL and ELF suggested that there may be
asymmetries in international conversationswhere native speakers are present and
that, conversely, a good deal of co-operative languagemight be expectedwhere non-
native speakerstogether constructedcommunities via their use of English. The thesis
therefore took on the specific aim of testing these hypothesesalongside the general
aim of characterisingEnglish as a World Language.
The plans were then laid for carrying out data gathering and analysis: it was decided
to collect togethera relatively large pool of speakersfrom a wide rangeof English-
speaking backgrounds and to try to find out, in the first instance, how they saw
Chapter10 252 Conclusion

themselves and others within the native speaker/non-native speaker framework.
Speakerswere then placed into international groups and further tests were carried out
to find out how group members thought of each other, again in terms of their
language use. A range of recording, transcribing and analysis options were then
pondered before a decision was made to record each group in a simulation which
was later transcribed and analysed using an eclectic approach with a bias towards
discourse analysis, pragmatics, interactional sociolinguistics and critical discourse
analysis. The resulting analysed conversations were compared to similar
conversations, the result of identical simulations, carried out among speakers in
national groups. Wherever possible, speakers who had taken part in the international
conversations were also recorded speaking among their co-nationals.
The results have been reported in three separate chapters of this thesis: one
international conversation was examined in very close comparison with one
intranational one, all the conversationswere then examined more superficially and
finally six individual speakers, in three pairs, had their use of language scrutinised in
an international and an intranational setting.
The results seem to confirm that speakersin international settings tend towards the
construction of comity, attempt to co-operatewith eachother and try to find common
ground by making extensive use of convergence strategies. This is not to say that
speakers, even the same speakers, do not use convergence strategies when
conversing with their co-nationals; what the is

results show that there is a substantial
reduction in divergence strategies when English is being used internationally,
compared to intranational use. The presence of native speakers in some of the
conversations does not seem to have made a great deal of difference to the results.

2. Discussion
A first reaction to theseresults might be that they presenta most predictable picture:
it would seemvery obvious that a group of people who do not know each other and
who come from different cultural backgroundsare likely to be more sensitive to each
other, more tentative and less direct than people who are among co-nationals, with
whom they share a great deal, culturally and linguistically, and with whom they
therefore feel at easeenoughto be more direct and more open.
That said, the results are still somewhat uneven and further research would be
required in order to make a stronger statement. It was noted in Chapter 9, for
example, that the change in language behaviour in one participant, from his
intranational conversation to his international one, might or might not be
representative of his national group. Another individual actually used more divergent
behaviour in her international conversation than she did in her intranational one,
going against the general pattern emerging from the other participants. If the results
confirm a common-sense perception, there is all the more reason to carry out further
research to guard against the danger of the `common sense' comfort zone.
What might be less obvious is the apparentlack of observableasymmetry associated
with native speakers confronting non-native speakers. There are plenty of
asymmetriesin all the conversationsbut little in the way of general explanations. At
times it seemsthat participants situated in the culturally-constructed North and West
are dominant participants, but there are instancesin which the opposite is true. Men
sometimes take a dominant role, but, again, there are instances where other
participants seem subordinateto a woman. In the international conversationswhere
native speakersare present,no particular pattern of asymmetry emerges.

What does emerge has as much to do with anthropology and sociology as it does
with the study of language: people accommodateto each other, using `Foreigner
Talk' at times (if that term is chosento describetheir behaviour), they seemto make
use of slightly different identities in different circumstances,behaving, typically, as a
national among co-nationals but modifying that behaviour when in international
settings.
A further less obvious facet of the results lies in the choice of linguistic forms with
which participants clothe their attempts at creating convergence.
Participants hedge their suggestions and proposals using `maybe' and `perhaps',
using `could' instead of `can', `would' instead of `will'; they tone them down using
à little bit' and `just' or by being vague; they agree with each other using the
performative verb àgree', but mostly by saying `yeah' and, similarly, they make
little use of the performative `concede', preferring to make concessions in
appropriately coherent statements to that effect. They use question tags, `you' and
`we' to appeal to each other in an atmosphere of community-building and pepper
their conversations with laughter, much of which seems slightly forced. They make
limited, but noticeableuse of colloquial language,vague languageand evenjokes in
order to keep up an atmosphereof co-operation. And underlying everything else,
they use forms of English which, in the main, seemto be perfectly comprehensibleto
other group members,irrespective of nationality.
A number of questions seemsto be suggestedby all of this. The first question is
about whether or not the striving for comity and the tendency towards convergence
is a human universal. Since it is anthropological or sociological in nature, and not a
language-studyquestion, it falls well beyond the scopeof this thesis and has possibly
beenansweredin the relevantliterature.

Following closely this general behavioural question, is the further question which
asks about the universality or otherwise of how, linguistically, the striving for comity
manifests itself: are the various linguistic devices used to achieve comity universal
and do users of English, therefore, translate them more or less from their other
languages?
A connectedquestion has to do with the position of monolingual speakersof English
who, by definition, cannot translate forms from another language but who may or
may not make a selection from their monolingual repertoire, according to their
perceptionsof how English might work when used internationally.
A final question needs to be asked concerning the underlying comprehensibility of
the languageforms used in EWL.
Different answers to these questions will imply different general views of the
position of English as a World Language and will entail different approaches to
teaching English for World communicationpurposes.
2.1 Anthropological and sociological universals

If the striving for comity and convergence is a natural, universal one, then many of
the worries concerning English and linguistic or cultural imperialism are, from an
EWL is only a positive force, facilitating a

albeit naive point of view, unfounded:
There are many riders to this ingenuous position, of course: in order

universal urge.
to be positive and beneficial, English has to be available to anyone and everyone and
has to be acquired without detriment to other languages and associated cultures,
be not be artificially held back from development. It must not facilitate

which must
the empowerment of some, at the expense of others (Kandiah 1998: 82-3).

Brutt-Griffler's view is highly optimistic in this respect,while Phillipson represents
the pessimistic side of the coin (Brutt-Griffler 2002, Phillipson 1992).
The pedagogic and educational entailments are straightforward: policies should
provide for the maximum spread of English to all citizens and should, at the same
time, seekto protect and nurture other languagesand cultures.
2.2 Universal manifestations
The secondquestion has to do with the extent to which users of English may borrow
strategies,procedures and language forms from other languages and cultures. The
in
results this thesis show common tendenciesto use convergent strategies, which
make use of a range of procedures including hedging and downtoning. Linguistic
items such as `maybe' and `perhaps' are used to hedge, while items such as à little
bit' and `just' are used to tone down utterances.
While the strategy itself may represent a universal, the procedures and language
forms may not: procedures for saving interlocutors' face, acquiring the right to give
an opinion, accepting compromise without losing face are, it seems different in
different cultures (Scollon and Scollon 1995). Language items may or may not be
approximately translated.
In this respect, English is certainly not neutral. It representsa way of dealing with
people which is not necessarilysharedby other cultures and languagesand, as such,
may well wash back into other languages and cultures. Cameron has noted, for
example, how English-basedforms of communication are often considered the best
way to personal advancement in Japanesesociety, in what she calls the `global
ideologyof communication'(Cameron2002).

PAGE
NUMBERS
CUT OFF
IN
ORIGINAL
This does not necessarilyhave a great deal to do with the native-speakingcountries
and their quasi-imperialistic reach. It is, however, interesting to note how, in the
conversationsin this research,the American homogeneousconversationwas, among
all the intranational ones,the most `convergent', leading to possible speculationthat
the international conversationsfollow American lines. It does, however, signal the
globalising `discourse of English' which implies à cultural politics', a different
representation of the self, even where those at the so-called periphery have
rearticulatedEnglish for their own purposes(Pennycook 1994:34 and 2002).
This view, which the results in this thesis could so easily support, entails an
approach to teaching English which would place it in a critical framework,
presumably at one with the culture within which it was being taught, and which
would allow learners and users to maintain a distance from the communication
systemsin play and to be wary of those systemsuncritically affecting other cultures
and languages.
2.3 The native speaker question
The results show that native speakersare similar to non-native speakersin that their
language behaviour in international settings is different from that in intranational
ones. The notion of Foreigner Talk has been found to be inadequateto account for
this while the idea of accommodation is more satisfactory. In seeking to
to,
accommodate and then to convergewith their international
interlocutors,
native
draw
speakers on their language in
resources ways which they feel
presumably are
appropriate.It might be said therefore, that they participate in the processof creating
a `third' culture,just do
as non-native speakers (c.f. Meierkord 2002:119).
('hantnr in 7CR ! `nhrhiainn

The results also show that native speakersare not the necessarily most influential
language users when it comes to creating models to which other speakers may
This
accommodate. the
strengthens point that EWL, the `discourseof English', is not
necessarilya native-speaker-drivenmachine and that the `globalising periphery' (c.f.
Pennycook2002) may have some influence over its functioning.
The Nuffield enquiry, referred to in Chapter 2, Section 4.1, suggestedthat British
pupils be taught how to use English internationally. The results here might show
in
ways which this could be approached.On the other hand, the fact that the native
appearedto select appropriately from their language repertoire in order to

speakers
make their international conversations run smoothly might suggest that no such
instruction would in fact be necessary.
Z.4 International English revisited'
All the participants in the reported conversations communicated with each other
difficulties. There were very few instancesof clarification and

without any apparent
no obvious instances of misunderstanding. Participants in the international
conversationscame from 24 different national backgrounds, several of which were
by
represented people speaking different languages.
The-relative-easewith which everyonecommunicatedpoints at the very least to a
common core which interactants drew on, judiciously avoiding anything which
might have too

seemed local.
While it is therefore incontrovertible that a new community and culture is potentially
founded each time a fresh international group of English-users meets and talks, it
also appears that the notion of a standard, or standardising international English,
Chant,., in ', co f nnrlýýýinn

dismissed in Chapter 2, may be a practical reality even though it cannot be a
theoretical one.
As far as teaching English is concerned,it therefore seemsto matter little what is
taught, since the evidence here suggeststhat, irrespective of the variety of English
learned, interactants were able to draw on their resources successfully. If there is
anything to be gained, it is in the areaof how English is taught, bringing up again the
critical questions referred to above and the necessity for rehearsing strategies and
procedureslikely to be useful in international settings.

References
261 References
Abbs, B. & Freebairn,I. (1979) Building Strategies Harlow, Longman

Abe, Megumi (2004) `Maintaining or constructingnew identities in World Englishes',paper given at
the IA WEannual conference,Syracuse,July 2004
Ahulu, S. 1997 `GeneralEnglish: A considerationof the nature of English as an international medium'
English Today 13.1.(49) 17-23
Amin, N. (1999) Minority women teachers of ESL: negotiating white English in Braine, G. (ed)
1999Non-native educators in English Language Teaching Mahwah, NJ. LawrenceErlbaum
Associates
Ammon, Ulrich (1994). International Languages. in Asher, RE. (GeneralEd.) The Encvclovoedia of
LanauaQeand Linguistics. Vol 4. Oxford, Pergamon.1725-1730
Annamalai, E. (1998) Nativity of Language in Rajendra Singh (ed) The Native Speaker. New Delhi,
SagePublications
Appadurai, Arjun. (1996). Modernity at large. Cultural dimensions of globalization. Minneapolis: U
of Minnesota Press.
Aravamudan, S (1999) Colonialism and agency 1688-1804 Durham, Duke University Press
Aston, M (1968) The Fifteenth Century: The Prospects of Europe New York, Harcourt Brace
Atkinson, J., and Heritage, J. (1984). Transcription notation. In J. Atkinson and J. Heritage (eds.).
Structures of Social Action: Studies in Conversational Analysis (pp. ix-xvi). Cambridge, C.U. P
Austin, J (1962) How to do things with words Oxford, ClarendonPress
Bamgbose,A. (1998) `Tom betweenthe norms: innovation in world Englishes' World Englishes 17/1
Barotchi M. (1994): Lingua Franca. in Asher, RE. (GeneralEd.) The Encyclopoedia of Language
and Linguistics, Vol 4. Oxford, Pergamon.2211
Bartsch R. (1998) Norms of Language London, Longman
Bateman, J and Paris, C (1991) Constraining the development of lexicogrammatical resources
during text generation: towards a computational instantiation of register theory in Ventola, E
(ed. ) Functional and systemic linguistics: approaches and uses Berlin, Mouton de Gruyter
(Trends in Linguistics Studies and Monographs 55), 81-106
Beebe,L. M. (1994) 'Field not dataon power and the power of field note data' Paper presentedat
TESOLconvention,Baltimore1994
Beneke, J (1991) Englisch as lingua franca oder als Medium interkultureller Kommunikation? In
Grebing, R. (ed) Grenzenloses Sprachenlernen. Festschrift für Reinhold Freudenstein. Berlin,
Cornelsen and Oxford University Press 54-66
Bex, Tony. (1993) `Standardsof English in Europe' Multilingua 12,3: 249-64
Bialystok, E. (1990) Communication Strategies: a Psychological Analysis of Second Language
Use Oxford, Blackwell
Biber, D., Conrad, S. and Reppen,R. (1998) Corpus Linguistics Cambridge,C.U.P.
Blank, P. (1996) Broken English London, Routledge
Bloomfield, L. (1933) Language New York: Henry Holt and Company
Bodtsch, G. & Villain-Gandossi, C. (eds.) (2001) Stfriotvpes dans les relations nord-sud Paris,
CNRS editions.
Bolton, K. (2003) Chinese Englishes Cambridge,C.U.P.
Braine, G. (ed) 1999 Non-native educators in English Language Teaching Mahwah, NJ. Lawrence
Erlbaum Associates
Brazil, D., Coulthard, M., & Johns,K. (1980) Discourse Intonation and Language Teaching,
London, Longman,
Brutt-Grifller, J (2002) World English: a study of its development Clevedon: Multilingual Matters
Bryan, B. (1994) English in its place in Hayhoe, M. and Parker, S. (eds.) (1994) Who Owns English
Milton Keynes, Open University Press.
Burton, D. (1978) `Towards an analysisof casualconversation' Nottingham Linguistics Circular 7(2):
131-59
Butler, S. (1997) Corpus of English in Southeast Asia: implications for a regional dictionary In
Bautista, M. L. S. (ed) English is an Asian language: the Philippine context Manila, Macquarie
Library, 103-124.
Cameron. D. (2002) Globalization and the teaching of 'communication skills' in Block, D. &
Cameron,D. (2002) Globalization and language teaching London, Routledge
Campbell, D., Ekniyom, P., Haque, A. and Smith, L. (1983) English in International Settings:
Problems and their Causes in Smith, L. (ed) Readings in English as an International
-
262 References
Language- Oxford, Pergamon1983

Canagarajah, A. S. (1999a) Interrogating the "Native Speaker Fallacy": Non-linguistic roots, non-
pedaeoaical results in Braine, G. (ed) 1999 Non-native educators in English Laneua2e
Teaching Mahwah, NJ. Lawrence Erlbaum Associates
Canagarajah,A. S. (2005) `Globalisation of English and changing pedagogicalpriorities' Paper given
at IATEFL Conference, Cardiff 2005
Candlin, C.N. (1987) Beyond Description to Explanation in Cross-cultural discourse in Smith, L.
(ed.) Discourse Across Cultures - Strategies in World Enelishes Hemel Hempstead, Prentice
Hall International,
Carter, R. (1999) Standard grammars. spoken grammars: Some educational implications in Be; T
and Watts, R. J. (eds) Standard English The Widening Debate London, Routledge
Carter, R. and McCarthy, M. J. (2006) The Cambridge Grammar of English Cambridge, C.U.P.
Cathcart,N., Carletta,J. and Klein, E. (2003) À shallow model of backchannelcontinuers in spoken
dialogue'. Proceedingsof the 10th Conferenceof the European Chapter of the Associationfor
ComputationalLinguistics (EACLIO), Budapest,April 2003.51-58
Chafe, W. L., Du Bois, J.W. and Thompson, S.A. (1991) Towards a new corpus of spoken American
English in Aijmer, K. and Altenberg, B. (eds.) English Corpus Linguistics London, Longman
Cheshire, J (1999) Spoken Standard English in Bex, T and Watts, R. J. (eds) Standard English The
Widening Debate London, Routledge
Chevillet, F (1991) Les variet6s de l'analais Paris, Nathan
Chevillet, F. (1993) Ènglish or Englishes?' English Today 9 (4) 29-33
Chiba, Reiko., Matsuura,Hiroko., Yamamota, Asako (1995) `Japaneseattitudes toward English
accents' World Englishes 14,77-86
Chomsky,N. (1965) Aspects of the theory of syntax Cambridge MA, MIT Press
Clyne, M. G. (1981) "Second Generation' Foreigner Talk in Australia' International Journal of the
Sociology of Language 28:69-80
Cobb, T. M. (1997) From Concord to Lexicon: Development and Test of a Corpus-Based Lexical
Tutor, unpublished PhD Thesis, Department of Educational Technology, Concordia University,
Montreal, Quebec, Canada
Connor, U. L. (1999) Learning to write academic prose in a second language: a literacy biography
in Braine, G. (ed) 1999 Non-native educators in English Language Teaching Mahwah, NJ.
Lawrence Erlbaum Associates
Coppieters,R. (1986) 'CompetenceDifferences betweenNative and Fluent Non-native Speakers'
Language 63(3): 544-73
Corder, S. P. (1973) Introducing Applied Linguistics. Harmondsworth, Penguin
Coulmas,F. (1981) Introduction: The Concept of Native Sneaker. in Coulmas,F. (ed.) A
festschrift for native speaker The Hague, Mouton
Coulthard, R. M. and Brazil D. (1979) Exchange structure: Discourse analysis monographs no. S
Birmingham: The University of Birmingham, English LanguageResearch.
Coupland,N. Coupland,J., Giles, H. and Henwood, K. (1988) Àccommodating the elderly: Invoking
and extending a theory' Language in Society 1,7
Crystal, D (1994) Which English - or English Which in Hayhoe, M. and Parker, S. (eds.) Who
Owns English Milton Keynes, Open University Press.
Crystal, D (1997) English as a Global Language Cambridge,C.U.P
Crystal, D (2001) The Future of Enalishes in Burns, A and Coffin, C (eds.) Analysing English in a
global context London, Routledge
D'Souza, J., (1988) First response Ito Singh, R Introductionll in Singh, It, D'Souza, J., Mohanan,
K.P. and Prabhu,N. S. (1998) On 'New/Non-native' Enelishes: A Quartet in Rajendra Singh
(ed) The Native Speaker. New Delhi, SagePublications
Dalby, A. (2002) Language in Dana-erLondon, Allen Lane
Dalby, D. (1998) The Linguasphere: from person to planet Hebron, Wales, LinguaspherePress
Dalton-Puffer, C. and Kaltenbock, G and Smit, U. (1997) 'Learner Attitudes and L2 pronunciation in
Austria' World Englishes 16/1: 115-28
Davies, A. (1989) 'Is International English an Interlanguage?' TESOL Quarterly. 23/3 p447-67
Davies, A. (1991) The Native Speaker in Applied Linguistics Edinburgh, Edinburgh University
Press
Davis, H. (1999) Typography. Lexicography and the Development of the Idea of `Standard
English' in Bex, T and Watts, R.J. (eds) Standard English The Widening Debate London,
Routledge
Deen, J (1997) Dealing with problems in intercultural communication Tilburg, Tilburg University
263 References
Press
Derwing, T. M., Rossiter, M. J. and Ehrensbreger-Dow, M (2002) "They Speaked and wrote real good':
judgements of non-native and native grammar' LanguageAwareness11/02
Dirven, R. and Putz, M. (1994) Intercultural Communication in Pürschel, H., Bartsch, E., Franklin,
P., Schmitz, U. and Vandermeeren, S. (eds) Intercultural Communication. Proceedings of the
17a'. International L. A. U. D. symposium Duisberg 23-27 March 1992 Frankfurt-am-Main, Peter
Lang
Duranti, A. (2005, first published 1989) Ethnography of speaking: toward a linguistics of the praxis
in Kiesling, S.F. and Paulston, C. B. Intercultural discourse and communication Malden MA,
Blackwell
Dürmüller, U. `The Internationalisation of English and its effects on the ELF/EL2 curriculum' Invited
conferencepaper at 'The Condition of the subject' University of London, July 17-19 2003
Eckersley, C.E. (1959) Essential English for foreign students London, Penguin
Eggins, S. and Slade D. (1997) Analysing Casual Conversation London, Cassell
Fairclough, N. (1995a) Critical Discourse Analysis London, Longman
Fairclough, N. (1995b ) Media Discourse London, Edward Arnold
Fairclough, N. (1989) Language and Power. London, Longman
Fairclough, N. (1992) Critical Language Awareness London, Longman
Fant, L. M. (1992) Analyzing negotiation talk - authentic data vs. role play in Grindsted, A and
Wagner, J. (eds.) Communication for specific purposes. Fachsprachliche Kommunikation
Tübingen: Narr. 164-175
Ferguson,C. (1975) 'Towards a characterizationof English foreigner talk' Anthropological
Linguistics 17: 1-14.
Firth, A. (1990) `Lingua Franca'Negotiations: Towards an Interactional Approach' World Englishes 9,
3 (1990) pp. 269-280
Firth, A. (1996) `The Discursive Accomplishment of'normality': On ConversationAnalysis and
'Lingua Franca" Journal of Pragmatics. 26: 237-259.
Fishman,J.A (1969) `National languagesand languagesof wider communication in the developing
nations' Anthropolgicallinguistics 11,111-135
Fishman, J.A. (1972) Language and nationalism Rowley, Mass. Newbury House
Fishman,J.A., Conrad, A. and Rubal-Lopez, A. (eds) (1996) Post-imperial English Berlin and New
York: Mouton de Gruyter
Freed,B. (1981) `Foreignertalk, baby talk, native talk' International Journal of the Sociology of
Language 28: 19-39
Fries, C. C. (1952) The structure of English New York, Harcourt Brace
Gairns, R. & Redman, S. (1996) True to Life Cambridge, C. U. P.
Gallois, C., Franklyn-Stokes, A., Giles, H and Coupland, N. (1988) Communication accommodation
in intercultural encounters. in Kim, Y. Y. and Gudykunst, W. B. (eds.) Theories in Intercultural
Communication Newbury Park, Sage. 157-185
Garfinkel, H. (1967). Studies in Ethnomethodologv. Englewood Cliffs, NJ, Prentice-Hall.
Gass,S. and Varonis, E. (1991) Miscommunication in Nonnative Speaker Discourse in Coupland,
N., Giles, H and Wiemann, J. (1991) Miscommunication and Problematic Talk Newbury Park,
SagePublications
Giles, H. (1970) Èvaluative reactionsto accents' Educational Review, 22,211-227
Giles, H. (1973 )'Accent mobility: A model and some data' Anthropological linguistics 15,87-105
Giles, H. and Powesland,P.F. (1975) SpeechS les and Social Evaluation London, Academic Press
Giles, H and Ryan, E.B. (1982) Prolegomena for developing a social psychological theory of
language attitudes in Ryan, E.B., Giles, H. (eds.) Attitudes towards Language Variation
London, Edward Arnold
Giles, H. and Smith, P.M. (1979) Accommodation theory: Optimal levels of convergence in Giles,
H. and St.Clair (eds.) Language and social psychology. Oxford, Blackwell, 45-65
Gnutzmann,C. (1998) 'English as a global language:what doesit mean?' Neusprachliche
Mittteilungen, 51(3) 130-137
Goffman, E. (1967) Interaction Ritual. GardenCity, N. Y.: Doubleday (Anchor Books)
Goffman, E. (1971). Relations in Public. New York: Basic Books.
Goffman, E. (1981) Forms of talk Philadelphia: University of Pennsylvania.
Golato, A. (2003) 'Studying Compliment Responses:A comparisonof DCTs and recordings of
naturally occurring talk' Applied Linguistics 24/1: 90-121
Görlach, Manfred (1988). 'English as a World Language: The State of the Art' English World-Wide,
9: 1.
264 References
Graddol, D. (1996) Global English, Global Culture? in Goodman, S and Graddol, D. (eds.)
Redesigning English London, Routledge
Graddol, D. (1997) The Future of English Manchester,The British Council
Graddol, D. (2004) 'The Future of English Next - Envisioning the future world of English language
learning' Paper given at Going Global Conference,British Council, Edinburgh, 2004
Gramkow Andersen,Karsten (1993). Lingua Franca Discourse: An Investigation of the Use of
English in an International Business Context. Unpublished M. A. Thesis.Aalborg University,
Denmark.
Green,G. (1992) The Universality of Gricean Accounts of politeness: You gotta have wa
Unpublished manuscript,University of Illinois
Grice, H. (1975) Logic and conversation William JamesLectures, Harvard University. Reprinted in
Cole, P. and Morgan J. (eds) Syntax and semantics Vol 3: Speech acts New York, Academic
Press,3-58
Gumperz, J. (1982a) Interethnic Communication in Gumperz, J. (ed) Discourse Strategies
Cambridge, C. U. P
Gumperz,J. (1982b) Language and social identity Cambridge,C.U.P.
Gumperz, J. (1990) The Conversational Analysis of Interethnic Communication in Scarcella, R.,
Andersen, E. and Krashen, S. (eds) Developing Communicative Competence in a Second
Language New York, Newbury House
Gumperz,J. (1991) Cross Talk Programmefrom BBC Education and Training, London, Mosaic
BBC Education,
Haegeman,P. (2002) Foreigner Talk in lingua franca businesstelephone calls in Knapp, K and
Meierkord, C. (eds.) Lingua Franca Communication Frankfurt, Peter Lang
Hall, E.T. (1976) Beyond Culture New York, Doubleday
Halliday, M. A. K. (1973) Explorations in the functions of language London, Edward Arnold
Halliday, M. A. K. (1975) Learning how to mean: explorations in the development of language
London, Edward Arnold
Halliday, M. A. K. (1978) Language as a social semiotic London, Edward Arnold
Halliday, M. A. K. (1994) An introduction to functional grammar London, Edward Arnold
Halliday, M. A. K. and Hasan,R. (1985) Language. context and text: aspectsof language in it social-
semiotic perspective Geelong,Deaking University Press
Hannam, S. (2005) 'Accent prejudice in ELT' Paper given at IATEFL annual conference,Cardiff,
2005.
Higgins, C. (2002) `"Ownership" of English in the Outer Circle: an alternative to the NS-NNS
dichotomy' TESOLQuarterly 37/4.615-644
Hockett, C.F. (1958) A Course in modern linguistics New York, Macmillan
Holmes, J. and Meyerhofl M. (eds) The Handbook of language and gender Oxford, Blackwell
Honey, J(1997)Language is Power London, Faber and Faber
Honey, John (1991) The Concept of "Standard English" in First and Second Languages in:
Tickoo, Makhan L., (Ed). Languages & Standards: Issues. Attitudes. Case Studies. Anthology
Series 26. CS: Southeast Asian Ministers of Education Organization (Singapore). Regional
Language Centre.
Hope, J. (2000) Rats, bats, sparrows and dogs: biology, linguistics and the nature of Standard
English in Wright, L. The development of standard English 1300-1800 Cambridge, C. U. P.
House, J (1999) Misunderstanding in intercultural communication: interactions in English as a
lingua franca and the myth of mutual intelligibility- in Gnutzmann,K. Teaching and learning
English as a global language Tübingen: Stauffenburg.73-89
House, J. (2002a) Ènglish as a lingua franca: A threat to multilingualism? ' Paper given in the
Workshopon Language and Globalisation, BAAL Conference,Cardiff 2002
House, J. (2002b) Developing araamatic competence in English as a lingua franca in in Knapp, K
and Meierkord, C. (eds) Lingua Franca Communication Frankfurt, Peter Lang
Hüllen, W (1982) `Teachinga foreign languageas'lingua franca" Grazer Linguistische Studien 16,
83-88
Hultfors, P (1987) Reactions to non-native English: Stockholm Studies in English LXXI,
Stockholm, Almqvist and Wiksell International
Hymes, D (1972b) Models of the interaction of language and social life in Gumperz, J. and Hymes,
D. (eds.) Directions in sociolinguistics: the ethnography of communication New York, Holt,
Rinehart and Winston, 35-71
Hymes, D (1974) The ethnography of sneaking in Blount, B. (ed.) Language. culture and society
265 References
Cambridge,Mass. Winthrop, 189-223

Hymes, D. (2005, first published 1986) Models of the interaction of language and social life:
toward a descriptive theory in Kiesling, S.F. and Paulston, C. B. Intercultural discourse and
communication Malden MA, Blackwell
Hyrkstedt, I and Kalaja, P (1998) Àttitudes toward English and its functions in Finland: A discourse-
analytic study' World Englishes, 17/3
Janicki, J. (1985) The Foreigner's Language: a Sociolinguistic Perspective Oxford, PergamonPress
Janicki, J. (1986) Accommodation in native-speaker-foreigner interaction in House, J. and Blum-
Kulka, S. (eds.) Interlinaual and Intercultural communication 169-78, Tübingen, Gunter Narr
Verlag
Janney,R.W. and Arndt, H. (1994) Interpersonal dimensions of intercultural communication in
Pürschel,H., Bartsch, E., Franklin, P., Schmitz, U. and Vandermeeren,S. (eds) Intercultural
Communication. Proceedings of the 17th.International L. A. U.D. symposium Duisbera 23-27
March 1992Frankfurt-am-Main: Peter Lang
Jenkins, J. (2000) The Phonolo2v of English as an International Language Oxford, O. U. P.
Jespersen, O. (1933) Essentials of English Grammar London, Allen and Unwin
Johnson, R. K. (1990) International English: Towards an Acceptable. Teachable Target Variety
World Englishes 9,3: 3 01-15
Joos,M. (1967) The Five Clocks New York, Harcourt, Brace and World
Jucker,A. H., Smith, S.W. and Ludge, T. (2003) Ìnteractive aspectsof vaguenessin conversation'
Journal of Pragmatics 35 (2003) 1737-1769
Kachru, B. B. (1985) Standards, codification and sociolinguistic realism: the English lantusee in
the outer circle in Quirk, R. and Widdowson, H. (eds) English in the World. Cambridge, C. U. P.
Kachru, B. B. (1992). Models for Non-Native Englishes. In: Kachru, B. B. (ed. ). The Other Tonaue:
English across Cultures. Urbana, Univ. of Illinois Press.
Kachru, B. B. (1994) `Crossculturalspeechact researchand the classroom' Pragmatics and Language
Learning 5,39-51
Kachru, B. B. & Nelson, C. (2001) World EnQlishes in Burns, A. & Coffin, C. Analysing English in
a global context London and New York, Routledge
Kalaja, P (1997) `Language attitudes reconsidered: from cognitive representation to discursive
reconstruction' Paper given at the American Association for Applied Linguistics Annual
Conference, Orlando, FL March 1997
Kalin, R. and Rayko, D (1980) The social significance of screechin the job interview in St.Clair,
RN., and Giles, H. (eds) The Social and psychological contexts of language Hillsdale, New
Jersey,Erlbaum
Kandiah, T. (1998) Epiphanies of the Deathless Native User's Manifold Avatars: A Post- Colonial
Perspective on the Native Speaker in Rajendra Singh (ed) The Native Sneaker. New Delhi,
Sage Publications
Kasper, G. (2000) Data collection in pragmatics research in Spencery-Oatey,H. (ed) Culturally
speaking: managing rapport through talk across cultures London and New York, Continuum
316-41
Kasper, G. (2005, first published 1997) Linguistic Etiquette in Kiesling, S.F. and Paulston, C. B.
(eds.) Intercultural discourse and communication Malden MA, Blackwell
Kilgarriff, A. `Putting Frequenciesin the Dictionary' International Journal of Lexicography 10 (2)
1997.Pp 135--155.
Knapp, K. (1987). English as an International Lingua Franca and the Teaching of Intercultural
Communication, In: Lörscher, Wolfgang and Schulze,Rainer (eds.). Perspectiveson Language
in performance. Studies in Linguistics. Literary Criticism, and Foreign Language Teaching
to honour Werner Hüllen on the Occasion of his Sixtieth Birthday. Tübingen,Narr. 1022-
1039.
Knapp, K. (2002) The fading out of the non-native speaker. A casestudy of unco-operative lingua
franca communication in Knapp, K and Meierkord, C. (eds) Lingua Franca Communication
Frankfurt, PeterLang
Knapp, K and Meierkord, C. (2002) Approaching lingua franca communication in Knapp, K and
Meierkord, C. (eds) Lingua Franca Communication Frankfurt, Peter Lang
Kramsch, C (1998) The Privilege of the intercultural speaker in Bryan, M. and Fleming, M. (eds.)
Foreign Language Learning in Intercultural Perspective Cambridge, C.U. P
Labov, W. (1972) `The study of language in its social context' Studium Generale 23: 30-87
Lakof, R. (2003) Language. gender. and politics: nutting "women" and "power" in the same
,
sentence in Holmes, J. and Meyerhot M. (eds) The Handbook of language and gender Oxford,
,
266 References
Blackwell
Lasagabaster,D. and Sierra,J.M. (2002) Ùniversity students'perceptionsof native and non-native
speakerteachersof English' LanguageAwareness,11,2
Leech, G. (1991) The state of the art in corpus linguistics in Aijmer, K. and Altenberg, B. (eds.)
English corpus linguistics New York, Pearson
Leech, G. & Svartvik, J. (1994) A Communicative grammar of the English language London,
Longman
Lehrer, A. (1989) 'Rememberingand representingprose: quoted speechas a data source' Discourse
Processes12: 105-25
Lester, R (1978) ELT Documents. English as an International Language London, British Council
Lesznyäk, A. (2002) From chaos to the smallest common denominator. Tonic management in
English lingua franca communication in Knapp, K and Meierkord, C. (eds) Lingua Franca
Communication Frankfurt, PeterLang
Lesznyäk, A. (2004) Communication in English as an International Lingua Franca Norderstedt
Books on demand,
Leung, C., Harris, R. and Rampton, B. (1997) The Idealised Native Speaker: Reified Ethnicities and
Classroom Realities: Contemporary Issues in TESOL Centre for Applied Linguistic Research,
London, Thames Valley University
Lippi-Green (1997) English with an Accent London, Routledge
Llamzon, T. A. (1983) Essential features of new varieties of English in Noss, R. B. (ed. ) Varieties of
English in South East Asia Singapore, RELC. 92-109
Long, M. (1981) Prior foreigner-talk experience and the negotiation of conversation with non-
native speakers Annals of the New York Academy of Science
Lorenz, G. (1998) Overstatement in advanced learners' writing: stylistic aspects of adjective
intensification in Granger, S (ed) Learner English on Computer Harlow, U.K. Addison Wesley
Longman
Lowenberg, P. (2000) AssessingEnglish proficiency in the global context: the significance of non-
native norms in Ho, W.K. and Ward, C (eds.) Language in the global context Singapore,RELC
Marton, W. and Preston, D. R. (1975) 'British and American English for Polish University Students;
researchreport and projections' Glottodidactica, 8: 27-43 1975
Matthiesen, C. M. I. M. and Bateman, J.(1991) Text generation and systemic linguistics:
experiences from English and Javanese London, Pinter
Mauranen, A. (2003) `The corpusof English as Lingua Frnacain Academic Settings' TESOL Quarterly
37,3.513-527
Mazzon, G. (2000) The ideology of the standard and the development of Extraterritorial Englishes
in Wright, L. The development of standard English 1300-1800 Cambridge, C. U. P.
McArthur, T. (1996) English in the World and in Europe in Hartmann, R. The English Language in
Eur e Intellect, Exeter
McArthur, T. (1998) The English Languages Cambridge, C.U.P
McArthur, T. (2003) 'World English, Euro-English, Nordic English' English Today 73, Vol 19.1 54-58
McCarthy, M. (2002) Good listenershin made plain in Reppen, R., Fitzmaurice, S.M. and Biber, D.
(2002) Using corpora to explore linguistic variation Amsterdam, John Benjamins
McKay, B. (1995) Attitudes among ESL students toward native and non-native sneaking
instructors University of Illinois, Urbana-Champaign Term Paper in Sociolinguistics
McKay, S.L. (2002) Teaching English as an international language Oxford, OUR
McNamara, T. (1996) Measuring Second Language Performance London, Longman
McWhorter, J. (2001) The Power of Babel London, Heinemann
Medgyes, P. (1992) `Native or non-native:who's worth more?' ELT Journal 46.4
Medgyes, P. (1999) Language Training: a neglected area in teacher education in Braine, G. (ed)
1999 Non-native educators in English Language Teaching Mahwah, NJ. Lawrence Erlbaum
Associates
Meeuwis, Michael (1994): `Nonnative-nonnative intercultural communication: An analysis of
instruction sessions for foreign engineers in a Belgian company' Multilingua 13/1-2: 59-82
Meierkord, C. (1996). Englisch als Medium der Interkulturellen Kommunikation:
Untersuchungen zum Non-Native-/ Non-Native-Speaker-Diskurs. Frankfurt a.M., Lang.
Meierkord, C. (1998) Lingua franca English: Characteristics of successful non-native-/non-native
sneaker discourse Erfurt Electronic Studiesin English (EESE), 1998.
http://webdoc.sub.gwdg.de edoeiateeseleese. html
Meierkord, C. (2000) Interpreting successfullingua franca interaction, An analysis of non-native-
/non-native small talk conversations in English in Fetzer, A. and Pittner, K. (eds) Conversation
267 References
Analysis: New Developments Linguistics Online, 5. SpecialIssue. http://www. lingiiistik-

online.com
Meierkord, C. (2002)Language stripped bare' or 'linguistic masala'? Culture in lingua franca
conversation in Knapp, K and Meierkord, C. (eds) Lingua Franca Communication Frankfurt,
PeterLang
Meierkord, C. (2004) `Local Englishes in Global Contexts' Paper given at IA WE annual conference,
Syracuse,July 2004 and forthcoming in Gnutzmann,C& Intemann, F (eds) The Globalisation of
English and the English language classroom Clevedon, Multilingual Matters.
Mey, J. (1981) Right or Wrong. my native speaker in Coulmas,F. (ed.) A festschrift for native
aker The Hague: Mouton
s eý
Mey, J. (1985) Whose Language? A Studv in Linguistic Pragmatics. Amsterdam,John Benjamins
Milroy, J (1999) The Consequencesof Standardisation in Descriptive Linguistics in Bex, T. and
Watts, R J., (eds) Standard English. The Widening Debate London, Routledge
Modiano, M. (1999a) 'International English in the Global Village' English Today 15.2(58),22-28
Modiano, M. (1999b) `StandardEnglish(es) and Educational Practicesfor the World's Lingua Franca'
English Today 15.4(60), 3-13
Modiano, M. (2001) 'A new variety of English' in Jenkins, J, Modiano, M and Seidlhofer, B. Èuro-
English' English Today, 17:4,13-19
Mufwene, S.S. (1998) Native Speaker. Proficient Speaker and Norms in Rajendra Singh (ed) The
Native Speaker, New Delhi, SagePublications
Mugglestone,L (2003) Talking Proper Oxford, O.U.P.
Nair-Venugopal, S. (2003) Ìntelligibility in English: Of what relevancetoday to intercultural
communication? ' Language and Intercultural Communication 3: 1, Clevedon,Multilingual Matters.
Nayar, P. B. (1997). ÈSLJEFLdichotomy today: languagepolitics or pragmatics?' TESOL Quarterly,
31 (1), 9-37.
Nero, S.J. (2002) Ènglishes,attitudes,education' English Today, 69/18.1
Norell, P. (1991) Native-Speaker Reactions to Swedish Pronunciation Errors in English:
Recoenition, Intelligibility and Attitude Stockholm Studies in English. LXXIX Almqvist and
Wiksell International, Stockholm
Norton, B. (1997) `Language,Identity and the Ownership of English' TESOL Quarterly 31(3) 409-430
Nuffield LanguagesEnquiry (2000) Languages: the next generation London, The Nuffield
Foundation
Partington, A. (1998) Patterns and Meanings Amsterdam,John Benjamins
Partridge,A. C. (1969) Tudor to Augustan English London, Andre Deutsch
Pennycook,A (1994) The Cultural Politics of English as an International Language London,
Longman
Pennycook,A. (2002) `Turning English inside out' Indian Journal of Applied Linguistics 28 (2) 25-43
Phillipson, R. (1992) Linguistic Imperialism Oxford, O.U.P.
Phillipson, R. (2003) `The Future of English (LanguageTeaching)' Paper given at TheFuture of
English and English Language TeachingConference,London, International House, November
2003
Platt, J., Weber, H. and Ho, M. L. (1984) The New Enalishes London, Routledge and Kegan Paul
Poos,D. and Simpson,R. (2002) Cross-disciplinary comparisons of hedging in Reppen,R.,
Fitzmaurice, S.M. and Biber, D. (2002) Using corpora to explore linguistic variation
Amsterdam, John Benjamins
Prabhu,N. S. (1998) Third response Ito Singh. R Introduction]) in Singh, R (ed) The Nativ
S ep aker, New Delhi, SagePublications
Preisler, B (1999) Functions and Forms of English in a European EFL Country in Bex, T. and
Watts, RJ. (eds) Standard English The Widening Debate, London, Routledge
Preisler, B. (1995). `StandardEnglish in the World' Multilingua 14:4.341-362
Promodrou,L. (2005) You see,it's sort of tricky for the L2 user: the puzzle of idiomaticity in
English as a lingua franca Unpublished PhD dissertation,University of Nottingham, UK
Promodrou,L. (Forthcoming a) `Towards a description of English as a lingua franca' ELT Journal
Promodrou,L. (Forthcoming b) Àn idiomatic puzzle: control and creativity in ELF' ELT Journal
Qiong, H.X. (2004) 'Why China English should standalongsideBritish, American, and other `world
Englishes" English Today 78. vol 20, n2.
Quirk, R. (1985) The English Language in a Global Context, in Quirk, R, Widdowson, H. G. (eds)
English in the World, Cambridge, C. U. P.
Quirk, R. (1990) What is Standard English? in Quirk, R and Stein, G. English in Use London,
Longman
268 References
Quirk, R (1991a) The Question of Standard in the International Use of English. In: Tickoo Makhan
L. (Ed.) Languages & Standards: Issues, Attitudes, Case Studies. Antholoav Series 26, CS:
SoutheastAsian Ministers of Education Organization (Singapore).Regional LanguageCentre.
Quirk, R. (1991b) Language Varieties and Standard Lan2ua2e. In: Tickoo, Makhan L. (Ed.)
Languages & Standards: Issues, Attitudes, Case Studies Antholozv Series 26. CS: Southeast
Asian Ministers of Education Organization (Singapore).Regional LanguageCentre.
Quirk, R., Greenbaum,S., Leech, G. and Svartvik, J. (1972) A Grammar of Contemporary English
London, Longman
Rajagopalan,K. (2004) `The concept of 'World English' and its implications for ELT' ELTJournal
58/2, April 2004
Rampton,M. B.H. (1990) `Displacing the `native speaker': expertise,affiliation and inheritance' ELT
Journal, 44/2.
Renouf R (1987) Corpus Development in Sinclair, J. (ed) Looking Up London Collins
Richards,L (2000) Using Nvivo in Qualitative Research Doncaster(Victoria, Australia) QSR
International
Rubin, D.L. (1992) `Nonlanguagefactors affecting undergraduates'judgementsof nonnative English-
speaking teaching assistants' Research in Higher Education 33(4): 511-531
Ryan, E.B., Carranza,M., and Moffie, R.W. (1977) 'Reactionstoward varying degreesof accentedness
in the speechof SpanishEnglish bilinguals' Language and Speech,20 267-273
Sacks,H. (1974) An analysis of the course of a joke's telling in conversation in Bauman, R. and
Sherzer,J. (eds.) Explorations in the ethnography of speaking Cambridge,C.U.P. 337-53
Samarin,William J. (1987). Lingua Franca. In: Ammon, U., Dittmar, N., Mattheier, K. J. (eds.).
Sociolinguistics Vol. 1. Berlin: Walter de Gruyter. 371-374
Samimy, K. K., & Brutt-Grifller, J.(1999) To Be a Native or Non-native Speaker: Perceptions of
"non-native" students in a graduate TESOL programme" in Braine, G. (ed) Non-native
in English Language Teaching Mahwah, NJ. Lawrence Erlbaum Associates
educators
Schwartz, J. (1980) The negotiation for meaning: Repair in conversations between second
language learners of English In Larsen-Freeman, D. (ed. ) Discourse Analysis in Second
Language Acquisition Research Rowley, M. A.: 138-153.
Scollon, R and Scollon, S (1983) Face in interethnic communication In Richards, J.C. and Schmidt,
RW. (eds) Language and Communication London, Longman
Scollon, R. and Scollon, S. (1995) Intercultural Communication. A Discourse Approach Oxford,
Blackwell
Searle,J. (1969) SpeechActs: an essayin the philosophy of language Cambridge C.U. P.
Searle,J. (1976) À classification of illocutionary acts' Language in society 5: 1-23
Seidihofer, B (2000) Lingua Franca English: Description and Pedagogy
htty: //wwweng helsinki fi/doe/ESSE5-2000/Seidlhoferhtm
Seidlhofer, B (2001) 'Closing a conceptualgap: the casefor a description of English as a lingua franca'
International Journal of Applied Linguistics, 11/2 133-158
Seidlhofer, B (2005) Vienna Oxford International Corpus of English
htttp://www. univie.ac.at/Analistiklvoice/ accessedApril 2005
Seidlhofer, B. (2002) Basic Questions in Knapp, K and Meierkord, C. (eds) Lingua Franca
Communication Frankfurt, Peter Lang
Seidlhofer, B. (2002) Personalcorrespondencedated September9th
Shaw, W. D. (1981) Asian Student Attitudes Towards English in Smith, L. E. (ed) Readings in
English as an International Language Oxford, Pergamon Press
Sifakis, N. C. (2004) 'Teaching EIL-Teaching International or Intercultural English?What Teachers
Should Know' System32,237-250
Silva, R.S. (2000) `Pragmatics,bilingualism, and the native speaker'Language and Communication
20(2), 161 - 178
Sinclair, J. and Coulthard, M. (1975) Towards an analysis of discourse Oxford, O.U.P.
Singh, R., D'Souza, J., Mohanan, K. P. and Prabhu,N. S. (1998) On 'New/Non-native' Englishes: A
Quartet in Rajendra Singh (ed) The Native Speaker. New Delhi, SagePublications
Singh, R, Lele J. & Martohardjono, G. (2005, first published 1988) Communication in a
multilingual society: some missed opportunities in Kiesling, S.F. and Paulston,C. B. (eds.)
Intercultural discourse and communication Malden MA, Blackwell
Singh, U.N. (1998) Series Editor's Introduction in Singh, R (ed) The Native Speaker. Sage
Publications,New Delhi
Smith, L. (1976) English as an International Auxiliary Language In Smith, L. (ed) Readings in
English as an International Language Oxford, Pergamon 1983
269 References
Smith, L (1978) Some distinctive features of ETIL vs. ESOL in English Language Education in
Smith, L (ed) (1983) Readings in English as an International Language Oxford, Pergamon
Press
Smith, L (1979) English for cross-cultural communication: the question of intelli2ibility in Smith,
L (ed) (1983) Readings in English as an International Language Oxford, PergamonPress
Smith, L (1981) English as an International Language. No Room for linguistic chauvinism in
Smith, L (ed) (1983) Readings in English as an International Language Oxford, Pergamon
Press
Smith, L. (1984) `TeachingEnglish as an International Language' Studium Linguistik 15.52-59
Smith, L. (1987a) Discourse Strategies and Cross-cultural Communication in Smith, L. (ed.)
Discourse Across Cultures - Strategies in World Enalishes Hemel Hempstead, Prentice Hall
International
Smith, L. (1991) Standards in World Enalishes. In: Tickoo, Makhan L. (Ed.) Languages
-and
Standards: Issues. Attitudes. Case Studies. Antholowv Series 26. CS: SoutheastAsian Ministers
of Education Organization (Singapore).Regional Language Centre
Smith, L. (1992) Spread of English and Issues of Intelligibility in Kachru, B (ed. ) The Other
Tongue, second edition Urbana and Chicago, University of Illinois.
SPE 1919-1948 Tracts 1-66 of the Society for Pure English, Oxford O.U. P
Spitzberg, B. H. & Cupach, W. R. Interpersonal communicative competence Beverly Hills, Sage.
Stokoe, E. H. (2005) Ànalysing gender and language' Journal of Sociolinguistics 9/1,118-133
Strevens, P. (1983) What is 'Standard English'9 in Smith, L. (ed) Readings in English as an
International Language Oxford, Pergamon
Swan,M. & Walter, C. (1993) The New Cambridge English Course 4 Cambridge, C.U.P.
Takashi, T and Beebe L (1987) `The development of pragmatic competenceby Japaneselearners of
English' JALTJournal8.131-155
Tannen, D (1984) Conversational style: analyzing talk among friends Norwood, NJ. Nablex
Tannen,D. (1989) Talking voices: repetition. dialogue and imagery in conversational discourse
Cambridge,C.U.P.
Tannen,D. (1990) You just don't understand: men and women in conversation New York,
Ballantine Books
Tao, H. (1998) 'An Interactional Account of The Generic 'You' ExpressionsIn Conversational
English' Paper presentedat the Annual Meeting of the International PragmaticsAssociation.
Reims, France.July 19-24,1998.
Tarone, E. (1977) Conscious communication strategies in interlanguage in Brown, H. D., Yorio,
C.A. and Crymes, R.C. (eds) On TESOL 1977 Washington,D. C. TESOL
Tarone,E. (1980) 'Communication strategies,foreigner talk, and repair in interlanguage'Language
learning, 30,417-3 1.
Tarone,E. and Yule, G. (1987) Communication Strategies in East-West Interactions in Smith, L.
(ed.) Discourse Across Cultures - Strategies in World Enalishes. Hemel Hempstead,Prentice
Hall International
Terrell, T. D. (1990) Foreigner talk as comprehensible input. In J. E. Alatis (Ed.), GURT 190.
Linguistics, language teaching and language acquisition: The interdependence of theory.
practice and research. Washington, DC: Georgetown University Press, 193-206
Thomas,J (1999) Voices from the periphery: Non-native teachers and issuesof credibility in
Braine, G. (ed) 1999Non-native educators in English Language Teaching Mahwah, NJ.
Lawrence Erlbaum Associates
Thomas,J. (1983) 'Cross-cultural pragmatic failure' Applied Linguistics 4.2
Timmis, I. (2002)'Native-speaker norms and International English' ELT Journal 56/3.
Toolan, M (1997) `RecenteringEnglish: New English and Global' English Today 13.4:3-10
Trudgill P and Hanna J (1994) International English London Arnold
Trudgill, P (1999) Standard English: What it Isn't in Bex, T and Watts, R. J. (eds) Standard English
The Widening Debate. London, Routledge
Trudgill, P (1995) Sociolinguistics: an introduction to language and society London, Penguin
Valdman, A. (1981) `Sociolinguistic aspectsof Foreigner Talk' International Journal of the Sociology
of Language 28: 41-52
Varonis, E.M. and S.Gass (1985) 'Miscommunication in native/non-native conversation' Language in
Society 14: 327-43
Ventola, E., and Mauranen, A. (1991) Non native writing and native revising of scientific Articles in
E. Ventola (ed. ) Functional and systemic linguistics Berlin, Mouton De Gruyter 457-492
Viereck, W. (1996) English in Europe: its nativisation and use as a lingua franca with special
270 References
reference to German-speaking countries in Hartmann, R. (ed) English Language in Europe.

Intellect, Exeter
von Helmolt, Katharina (1997): Kommunikation in internationalen Arbeitsgruppen. Eine
Fallstudie über divergierende Konventionen der Modalitätskonstituierung. München,
ludicium
Wagner, J. and Firth, A. (1997) Communication Strategies at Work in Kasper, G. and Kellerman, E.
(eds) Communication Strategies London, Longman
Warschauer,M (2000) `The ChangingGlobal Economy and the Future of English Teaching' TESOL
Quarterly, 34/3
Widdowson, H (1994) `The Ownership of English' TESOLQuarterly 28,2 377-89
Widdowson, H. G. (1995a) `Discourseanalysis: a critical view' Language andLiterature 4 (3): 157-
172.
Widdowson, H. G. (1995b) `Review: Norman Fairclough: Discourse and social change, 1992' Applied
Linguistics 16 (4): 510-516.
Widdowson, H. G. (1996) 'Reply to Fairclough: discourse and interpretation: conjectures and
refutations' Language and Literature 5 (1): 57-69.
Widdowson, H. G. (1997) `The use of grammar,the grammar of use' Functions of Language 4 (2): 145-
168.
Widdowson, H. G. (1998) `Review article: the theory and practice of critical discourseanalysis'
Applied Linguistics 19 (1): 136-151.
Widdowson, H. G. (2000) Òn the limitations of linguistics applied' Applied Linguistics 21(1): 3-25.
Widdowson, H. G. (2001a) Ìnterpretationsand correlations: a reply to Stubbs' Applied Linguistics 22
(4): 531-538.
Widdowson, H. G. (2001b) `Scoring points by critical analysis: a reaction to Beaugrande' Applied
Linguistics 22 (2): 266-272.
Wittgenstein, L. (1953/2002). Philosophical Investigations, Oxford: Blackwell
Wolff, H (1959) Ìntelligibility and inter-ethnic attitudes' Anthropoloical Linguistics 1/3: 34-41
Wong, I. (1982) Native-speaker English for the third world today? in Pride, J. (1982) New
Englishes. Rowley, MA: Newbury House
Würm, S.A. (1971) Pidgins. creoles and lingue franche In Sebeok, T. A. et at. (eds) Current Trends
in Linguistics Vol 8 The Hague, Mouton, 999-1021
Yano, Y (2004) Ìs Codification of English as an International LanguagePossible?A challengeto
diversification of World Englishes' Paper given at IA WE Conference,Syracuse,July 2004
Yngve, V. (1981) The Struggle for a theory of Native Sneaker in Coulmas, F. (ed. ) A festschrift for
native speaker The Hague, Mouton
Young, R. (1988) `Variation and the interlanguage hypothesis' Studies in Second Language
Acquisition 10:281-302
Yuan, Y. (2001) Àn Enquiry into empirical pragmaticsdata-gatheringmethods:written DCTs and oral
DCTs, field notes, and natural conversations'Journal of Pragmatics 33,271-292
Yule, G. (1990) Ìnteractive conflict resolution in English' World Englishes 9/1: 53-62
Zima, Petr (1977). `Lingua Franca: Is Grammatical Reduction and Interferencefrom Other Languages
Universal?' Linguistica Generalia, 2.137-145.
Zuengler, J (1991) Accommodation in native-nonnative interactions: Going beyond the" what" to
the "why" in second-languageresearch in Giles, H. N. Coupland and J.Coupland (eds).
Contexts of Accommodation. Developments in Applied Sociolinguistics. Cambridge. C.U. P
271 References

Vol1 PDF

Uploaded by

Copyright:

Available Formats

Vol1 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Vol1 PDF

Uploaded by

Copyright:

Available Formats

Roberts, Paul (2005) Spoken English as a world

language: international and intranational settings. PhD

Access from the University of Nottingham repository:

Copyright and reuse:

For more information, please contact [email protected]

SPOKEN ENGLISH AS A WORLD LANGUAGE: INTERNATIONAL AND

Thesis submitted to the University of Nottingham

English used in homogeneous,intranational settings. After a brief introduction, the

relevant literature is reviewed in two chapters: firstly the concept of an international

variety of English is challenged and, following this, there is consideration of current

thinking under the headingsof English as an International Language and English as

concerning the way in which EWL might be characterised,with particular attention to

attitudes among different sorts of speaker.

ChapterFive introducesmethodologies(1) for finding data-providing participants and

(2) assessingtheir language-relatedattitudes relevant to the research questions. It

applied to ascertainparticipants' language-relatedattitudes. Following this, the results

may be characterisedby greater convergenceamong irrespective

or not they are native speakers.Chapter 9 examinesthe whole suite of conversations

homogeneousconversationstend to be as divergent as they are convergent, where in

they try their bestto maintainan atmosphereof comity. Chapter

10 completesthe results sectionby comparingthe performanceof six speakersin

particular, who each participate in an EWL conversationand in a homogeneousone.

They are found to draw on convergencestrategiesfor their EWL conversationswhile

being more direct and divergent in their homogeneousones. Chapter 11 attempts to

this thesis from my supervisor,ProfessorMichael McCarthy.

I would also like to acknowledgethe following:

the large number of studentsand staff at the University of Hertfordshirewho

the School of Combined Studies at the University of Hertfordshire which made my

work possible by providing resources,most importantly time resources,allowing the

the Ethics Committee in the Humanities Faculty at the University of Hertfordshire

who allowed me to go aheadand gatherdata from human subjects;

made useful suggestions during the early stages of writing.

1. ChangingEnglish, Changing view of English 2

Chapter 2: Is English as a World Language a Variety of English?

Chapter 3: English as a ..... Language

3.2.4 ELF and the conceptof `common core' 56

Chapter 4: Hypotheses concerning EWL interactions

Chapter 5: Gathering and analysing data

6. Analysis of the conversationdata: word counting and annotation 115

Chapter 6: Results: preliminary discussionsand questionnaires

Chapter 7: Results: A comparison between and international and an

3.1 Summaryand further explanation 163

Chapter 8: Results: Overview of all conversations

Chapter 9: Results: six casestudies

2.3 Conversationalstyle 227

Chapter 10: Conclusion

1. Changing English, Changing view of English

1.1 The Growing number of people speaking English

`Global English' (Preisler 1995, Graddol 1996, Toolan 1997, Brutt-Griffier

2002:176, Bolton 2003:4,).

standardEnglishes or `local' vernaculars(Graddol 1996). Instead,it will focus on the

latter result of change,the apparentcoming together of English users and whatever it

is thatthey share,ratherthanwhatthey do not.

suggestionand prediction made in 1832 by JacobGrimm, that `the English language

all portions of the globe' (Grimm 1832,cited in Pennycook, 1994:70). In connecting

politically, but also culturally)' (Fishman et. al. 4),

English as `the languageof international high-flyers' (Toolan 1997) and McArthur,

somewhatreluctantly perhaps,concludesthat `what is sharedworldwide, in fact, is

international elite' (McArthur 1996:14). Pennycook, too, considers that those