Gra var and cha in jer eng

2.1 Linguistic variation and change: Fields and methods 82.2 Morphological and syntactic variation 10 3.1.1 Non-linguistic research relevant to the study 26 3.1.2 Research on Channel Isl

Grammatical Variation and Change in Jersey English

Volume G48

Grammatical Variation and Change in Jersey English

by Anna Rosen

Varieties of English Around the World (VEAW)

A companion monograph series devoted to sociolinguistic research, surveys and annotated text collections The VEAW series is divided into two parts: a text series contains carefully selected specimens of Englishes documenting the coexistence of regional, social, stylistic and diachronic varieties in a particular region; and a general series which contains outstanding studies

in the field, collections of papers devoted to one region or written by one scholar, bibliographies and other reference works

For an overview of all books published in this series, please see

Library of Congress Cataloging-in-Publication Data

Rosen, Anna

Grammatical Variation and Change in Jersey English / Anna Rosen.

p cm (Varieties of English Around the World, issn 0172-7362 ; v G48)

Includes bibliographical references and index.

1 English language Dialects Jersey 2 English language Variation Jersey 3 English

language Jersey Foreign elements 4 English language Jersey 5 Linguistic

change Jersey 6 Languages in contact Jersey 7 Jersey Languages I Title.

the American National Standard for Information Sciences – Permanence

of Paper for Printed Library Materials, ansi z39.48-1984.

2.1 Linguistic variation and change: Fields and methods 8

2.2 Morphological and syntactic variation 10

3.1.1 Non-linguistic research relevant to the study 26

3.1.2 Research on Channel Island French 27

3.1.3 Research on Channel Island English 28

3.2 Socio-historical overview and sociolinguistic situation today 33

4.4 Description and characteristics of the corpus of spoken Jersey English 60 4.4.1 The Jersey Interview Corpus (JIC) 61

4.4.2 The Jersey Archive Corpus (JAC)

and its additional component (JACa) 65 4.5 Analyses and statistical testing 66

4.6 Summary 67

chapter 5

Discourse marker eh 69 5.1 The particle eh 70

5.2 Syntactic contexts and pragmatic functions 72

5.3 Distribution of eh in Jersey English 77

5.3.1 Distribution of eh by age 78

5.3.2 Eh in comparison with other discourse markers 81

5.3.3 Distribution of eh by education and occupation 84

5.3.4 Distribution of eh by gender 86

5.4 Eh – a contact phenomenon? 87

5.5 Comparison with eh in Guernsey English 91

5.6 Comparison with eh in British English 93

5.7 Eh – an identity marker? 96

5.8 Summary and conclusion 101

chapter 6

6.1 I went and buy them: Verb-and-verb constructions 103

6.1.1 Previous findings on verb-and-verb constructions

in standard English 104 6.1.2 Previous findings on FAP in Channel Island English 109

6.1.3 Findings on FAP in Jersey English 111

6.1.3.1 Characteristics of FAP in Jersey English 112 6.1.3.2 Sociolinguistic distribution of FAP in Jersey English 115 6.1.4 FAP – a contact phenomenon? 120

6.1.5 Summary and conclusion 123

6.2 There’s a lot of Jersey cows: Agreement in existential

there-constructions 124

6.2.1 Agreement in existential there-constructions

across English varieties 125

6.2.2 Previous research on agreement in existentials

in Channel Island English 127

6.2.3 Findings on agreement in existential there-constructions

6.2.5 Summary and conclusion 139

6.3 Further observations on the verb phrase 140

6.3.1 Tense and aspect 141

7.2.7 Emphatic that one 161

7.2.8 Demonstrative this here/that here 161

7.2.9 But yes 161

7.3 Prepositional usage 162

7.4 Use of the definite article 165

7.5 Negation 167

7.6 Absence of ’s in the local genitive 169

7.7 She/her and he/him used for inanimate referents 169

7.8 After as a time adverbial 170

7.9 If-deletion in conditional clauses 170

7.10 Analytic vs synthetic comparative forms 171

7.11 As what/than what in comparative sentences 171

7.12 Absence of plural marking 172

7.13 Formation of questions 172

7.14 Adverbs having the same form as adjectives 173

7.15 Like as a focussing device, a discourse marker and a quotative particle 174

8.5 Implications of identity aspects for Jersey English 200

8.6 Summary and conclusion 204

chapter 9

Conclusion 205 References 213

Index 235

This book is based on research carried out at the University of Bamberg and I would like to thank all those who have helped to make it possible First and fore-most, I wish to thank Manfred Krug whose enthusiasm and constant support encouraged me to embark on this project and to see it through He has been a mentor to me since my early student days when he first passed on some valuable advice (over an excellent cup of espresso, I’m sure) For his faith in me, and for his friendship, I will always be grateful My appreciation also goes to Julia Schlüter, who gave generously of her time and advice This work has profited immensely from her incredible eye for detail and from both hers and Manfred Krug’s expertise and feedback

I would also like to extend my gratitude to the entire team of the Chair of English Linguistics at the University of Bamberg It has been a great joy to work

in such an inspiring atmosphere and with such cooperative colleagues I was tunate enough to share an office with Gabriele Knappe, who generously shared her wisdom – linguistic and otherwise Thanks are further due to our student assistants Theresa Schmid, Carolin Leuthäußer, Annkatrin Langguth and Helen Etheridge,

for-my allies in compiling the corpus of spoken Jersey English Theresa transcribed nine archive recordings, Carolin, Annkatrin and Helen checked and proofread parts of the corpus and Carolin also anonymized the interview corpus

Very special thanks go to all participants in my study for their kind tion, good humour and their willingness to share their experience with me in long interviews I am particularly grateful to Sophie, Jennie and Richard Le Sueur, who spared no effort in helping me during my fieldwork in Jersey and who became friends along the way The assistance of the staff of the Jersey Archive, who re-trieved and copied suitable Oral History recordings, was also much appreciated

coopera-It goes without saying that my research would not have been possible without everyone’s help in Jersey

In addition, I gratefully acknowledge a grant from the German Academic Exchange Service (DAAD) that funded my research stay in Jersey, as well as travel funds from the Spanish Ministry for Science and Innovation and the European Regional Development Fund (grant HUM2007-60706/FILO) that allowed me to take part in several international conferences

Heartfelt thanks go to Ole Schützler, Katrin Sell, Christiane Hansen, Johanna Gerwin, Michaela Hilbert, Shane Walshe and Jessica Fischer for providing helpful comments on different parts of earlier drafts and for their general support while I was working on this project I also owe special thanks to Mari Jones and Heinrich Ramisch for insightful comments, expert advice and stimulating discussions at various stages of this study Thanks are further due to Kees Vaes at John Benjamins for editorial help and advice Stephanie Hackert and an anonymous reviewer pro-vided additional constructive feedback on this work It is a pleasure to thank them

here for this and for accepting the manuscript into the Varieties of English Around the World series Needless to say, I alone am responsible for any inaccuracies and

shortcomings which remain

I should also like to take this opportunity to thank my parents, my sister and

my friends As always, I could count on their encouragement and support and I know how lucky I am in having them My deepest thanks, finally, go to my hus-band, Holger Loritz He designed the maps in Chapters 1 and 3 and was unfailingly helpful and patient with advice on statistics and many other matters Most im-portantly, his support throughout the writing process and beyond kept me going

I dedicate this book to him

Bamberg, September 2013

List of maps, figures and tables

Figure 5.1 Distribution of main pragmatic functions across age groups 77

Figure 5.2 Mean distribution of eh by age 79

Figure 5.3 Frequency index of solidarity-building DMs by age

and linguistic background 82 Figure 5.4 Distribution of DMs across age groups in comparison 83

Figure 5.5 Mean distribution of eh by education 84 Figure 5.6 Mean distribution of eh by occupation 85 Figure 5.7 Mean distribution of eh by gender 86 Figure 5.8 Distribution of eh by corpus and age 94 Figure 5.9 Distribution of eh by corpus and main pragmatic functions 95

Figure 6.1 A typology of pseudo-coordinative types 106

Figure 6.2 Frequencies of standard went to/went and/came to/came and

followed by a verb in the JIC and the spoken part of the BNC 118 Figure 6.3 Mean distribution of singular forms of BE in plural existentials

(present tense) by age and gender 132 Figure 6.4 Mean distribution of singular forms of BE in plural existentials

(past tense) by age and gender 133 Figure 6.5 Mean distribution of singular forms of BE in plural existentials

(present vs past tense) by occupational background 134 Figure 6.6 Mean distribution of singular forms of BE in plural existentials

(past tense) in age group 60+ by linguistic background 135

Figure 6.7 Mean distribution of there was in plural existentials by age

(and linguistic background) in the JIC and the spoken part of the BNC 136 Table 3.1 Birthplace of inhabitants according to the 1851 Census 36 Table 3.2 Jersey population by place of birth 39 Table 4.1 Sample design 45 Table 4.2 Analytic classes and operational categories of NS-SEC 50 Table 4.3 Structure of the corpus of spoken Jersey English 60 Table 4.4 Speaker sample of the JIC 61 Table 4.5 Distribution of speakers by social background 64

Table 5.1 Pragmatic functions of eh in JersE 75

Table 5.2 Main pragmatic functions of eh in JersE 76 Table 5.3 Distribution of eh by age, gender and language background of speakers 78 Table 5.4 Distribution of eh by linguistic background of speakers 87 Table 5.5 Distribution of pragmatic functions of eh in four draws

from the spoken part of the BNC 95 Table 6.1 FAP – verb types, forms and frequencies in V1 position 113 Table 6.2 Distribution of FAP by age, gender and linguistic background

of speaker groups 115 Table 6.3 Normalized frequencies for FAP-occurrence in JIC and JAC 115

Table 6.4 Distribution of standard verb-and-verb and FAP structures in JIC

by age, gender and linguistic background of speakers 117 Table 6.5 Distribution of plural existentials in JIC and JAC

by agreement and tense 129 Table 6.6 Distribution of plural vs singular forms of BE according

to present and past tense 129 Table 6.7 Distribution of plural vs singular forms of BE according to polarity 129 Table 6.8 Distribution of plural vs singular forms of BE according

to presence/absence of overt plural marking on the following noun 130 Table 6.9 Distribution of plural vs singular forms of BE according

to presence/absence of intervening material between verb and NP 130 Table 6.10 Distribution of plural vs singular forms of BE according

to presence/absence of extension after the subject NP 130 Table 7.1 Distribution of relative markers in JersE, in more traditional

British English dialects and in eight BNC texts 153

Table 7.2 Distribution of relative which and who by social background in JIC 154

Table 7.3 Distribution of non-standard relatives by age, gender

and linguistic background in JIC 155

Table 7.4 Distribution of relative which and who by age

and linguistic background in JIC 155 Table 7.5 Distribution of pronominal apposition by speaker group in JIC 158 Table 7.6 Distribution of non-standard prepositional usage in JAC

and JACa, JIC and in Ramisch’s (1989) and Barbé’s (1993) data 165 Table 7.7 Distribution of non-standard definite article use by speaker group

in JIC and JAC(a) 166 Table 7.8 Distribution of multiple negation by speaker group in JIC and JAC(a) 168 Table 7.9 Presence of morphosyntactic features in JersE and GuernsE

in comparison 176 Table 8.1 Presence of (non-standard) morphosyntactic features

per individual speaker in JIC 182 Table 8.2 Mean distribution of types of non-standard morphosyntactic features

per speaker group in JIC 187

chapter 1

Introduction

Islands seem to have always held a special interest for dialectologists and guists (e.g Labov 1963; Melchers 1985; Wolfram & Schilling-Estes 1996; Schreier 2003) The assumption has been that speech communities, developing in isola-tion and under special conditions over centuries in well-defined and confined spaces, lend themselves particularly well to in-depth studies of linguistic variation and change and the correlation of such processes with social categories Wolfram and Schilling-Estes (1996: 103) in their study of the dialect of Ocracoke, an is-land located off the North Carolina coast, claim that “[t]he changing status of island communities in a world of ever-widening channels of communication and transportation makes these locales ideal for examining some of the critical issues underlying the socially-situated dynamics of language change and variation.” The present study capitalizes on this idea and places Jersey (a small island between England and France) and the variety of English spoken there at the centre of its investigation As a first systematic study of Jersey English, it will analyse patterns of morphosyntactic variation and change based on a newly created corpus of spoken material and compare them to findings from studies of Guernsey English In ad-dition, the study explores questions about the identity-creating potential of some grammatical Jersey English features and, more generally, about the influence of a local identity on language use in Jersey

sociolin-Jersey is, with 118 km² and 92,500 inhabitants, the largest of the Channel Islands, an archipelago between England and France in the Bay of St Malo, which

is also comprised of Guernsey, Alderney, Sark and Herm and various smaller lands (see Map 1.1) Jersey’s sister island Guernsey is the second-largest island with 62 km² and 62,300 residents Today, the Channel Islands, politically divided into the Bailiwick of Jersey and Guernsey, enjoy a special status as dependencies of the British Crown in Europe granting them largely political autonomy apart from matters of defence and diplomatic representation

is-Overlooked by English dialectologists until the 1980s, the Islands have ditionally been considered a francophone dialect area (cf e.g Viereck 1983: 29) However, English has been spoken on the islands for centuries alongside insular Norman French and, to a much lesser degree, standard French Indeed, English

tra-is by now the dominant and, for the vast majority of tra-islanders, only available guage This continuous (although by now diminishing) contact situation makes

lan-as the actual setting in the Channel Islands goes beyond these languages and also includes contact between various dialects of English As the Islands have long been

an attractive destination for immigration, for reasons that will become clear in a subsequent chapter, the English varieties of incoming speakers have certainly also contributed to shaping a distinct regional variety In an in-depth examination of, partly historical, contact influences, this study will reveal a complicated interplay

of forces at work in this variety

While Viereck (1988: 468) still described the Channel Islands as an “Anglicist’s no-man’s land” in a first article on Channel Island English (CIE), the situation since then has been partly remedied Ramisch (1989) and Barbé (1993, 1995a) deal in their PhD-theses with Guernsey English (GuernsE), both focusing on morpho-syntactic variation, and Jones (2001) addresses the influence of Norman French on Jersey English (JersE) in a subchapter of her monograph on Jersey Norman French

(also called Jèrriais).1 Otherwise, however, no intensive study of JersE exists to date Whereas all three of these studies identify the English spoken on Guernsey and Jersey as distinct regional varieties with a “unique blend of features originating

1 The terms Jèrriais and Jersey Norman French for the Norman dialect of Jersey will be used

in-terchangeably throughout this study The same goes for Guernesiais and Guernsey Norman French.

England

Alderney Guernsey Herm

Sark Jersey

France

N

Map 1.1 Geographical location of the Channel Islands

from different sources” (Ramisch 2004: 214; see also Barbé 1993: 248, 1995a: 1; Jones 2001: 189), the existence of a regional dialect of English in the Channel Islands is generally not recognized on the Islands where public debate usually centres around the obsolescence and preservation of Norman French Yet, more recently, and perhaps due to the plight of insular Norman, islanders’ attitudes and sense of identity seem to be subject to change towards more awareness of the islands’ uniqueness, especially in terms of their Norman heritage After all, traces

of insular Norman French might survive – and this might come as some lation – in the use of Norman features transferred to the English spoken on the islands (cf Lösch 2000: 140–143; Jones 2001: 190, 2010: 54)

conso-As will be shown in the main part of this work, however, there is strong dence to believe that JersE is also moving towards a more standardized and levelled British variety, thereby losing some of its distinctive character which had resulted from contact with Norman French Findings in Ramisch (1989) and Barbé (1993), emerging from data collected some 20 to 25 years prior to the present study, indi-cate for GuernsE that the loss of Norman French and the decline of bilingualism

evi-on the island might also lead to a complete loss of some GuernsE features that originated from contact with Norman French If the same is true for JersE, the present book is timely given that most bilingual speakers of Jersey Norman French and English are aged 60 and over Thus, there is no time to be lost if we want to record and analyse their use of English in order to obtain a fuller picture of earlier developmental phases of JersE

Islanders, though often rejecting the idea of an autonomous English dialect

of the Channel Islands, are nevertheless aware of some special features in their English The following is a typical comment of a JersE speaker recorded during

my fieldwork: “It’s just slightly different things which you pick up on You think,

Oh that’s a Jersey way of saying it.” (JIC17f1954) Jones (2001: 180) rightly points out that it would be interesting to pursue the possibility of a distinct CIE variety being preserved as a mark of local identity In the light of a possibly endangered dialect, the present work therefore aims to furnish further insights into the nature

of the connection between identity and dialect use It will explore whether a sense

of local belonging, positive attitudes towards and a strong identification with Jersey have an influence on the use of distinctly local dialect forms

Jones (2001: 189–190, in a similar vein see also Barbé 1993: 251) also identifies the need for a more systematic study of JersE (as well as the Englishes of Alderney and Sark), which would allow, first, to draw a comparison between GuernsE and other varieties of English spoken in the Islands and to determine the actual extent

of the, so far only suggested, homogeneity of these varieties and, second, to trace the developments and changes in progress in JersE, possibly also towards a decline

of so far preserved Norman French transfer structures

Following up on this desideratum, this study represents a first attempt of a more systematic description of JersE morphosyntax The limitation to the language level of grammar (and to a certain extent also discourse) is grounded in the fact that such a focus permits a direct comparison with the two existing studies of GuernsE and, therefore, will hopefully lead to insights into the general nature of CIE To the same end, the study also includes a diachronic dimension in its use of oral history data recorded at about the same time as Ramisch’s (1989) and Barbé’s (1993) first empirical research Analyses of these data will allow us to perform better comparisons between GuernsE and JersE and help to address the ques-tion of whether and to what extent parallel developments occur in both varieties Furthermore, this study aims to make a contribution towards describing patterns

of variation and change in JersE, thereby drawing on and connecting ideas oped in sociolinguistics, dialectology and contact linguistics

devel-To accomplish these aims, quantitative as well as qualitative insights will be essential Analyses will be based largely on sociolinguistic fieldwork data from

2008 and oral history material from the 1990s, compiled to a corpus of 354,000 words, but also on a range of other data from ethnographic fieldwork, a question-naire and some other written sources I shall argue that the study benefits from

a combination of quantitative and qualitative methods where the interpretation

of corpus data is enriched by detailed information on the social background of individual speakers and on local social categories and speaker networks Patterns

of variation and language stability will then become visible on the level of the speech community and its various speaker groups as well as on the level of the individual This will also help to clarify what combinations of factors determine the use of certain JersE features

The following research questions arise from the situation of JersE as outlined above and will guide the study’s analyses:2

1 What is the relationship between JersE and GuernsE? Can we speak of one or several distinctive CIE varieties? Which morphosyntactic features are charac-teristic of them?

2 How (and to what extent) does JersE reflect the (historical) linguistic contact situation of the Channel Islands, especially with Norman French?

3 In what direction is the variety moving? How (if at all) do more recent opments of JersE relate to phenomena of standardization and levelling?

devel-4 Which social factors influence processes of variation and change in JersE? Do factors of identity and attitudes play a role?

2 Similar questions have – naturally – been put forward in studies of other insular varieties of

English such as Ocracoke English (cf Wolfram & Schilling-Estes 1996: 109).

These questions are not intended to imply that JersE is necessarily a homogeneous variety It is even likely that, due to various linguistic influences and a constant influx of immigrants from diverse backgrounds, the English as spoken on Jersey shows much internal variation There is also no suggestion that the morphosyn-tactic features analysed for this study are all unique to the Channel Islands, though some of them clearly are Rather, it is expected that many morphosyntactic features will either differ quantitatively or qualitatively from elsewhere or be distinctive of CIE in their combination Given the development of corpus linguistics as a method

in the research of English varieties world-wide, claims as to what makes a variety unique can be more reliably substantiated today than at the time of Ramisch’s and Barbé’s initial studies (on the desirability of this, see, e.g., Barbé 1995a: 33) The study therefore adopts an approach of cross-varietal comparison which enables

us to compare frequencies of use of both rarer and more widespread features and their pragmatic contexts in JersE and other English varieties

The study begins with a sketch of the theoretical foundations, on which the subsequent analyses build (Chapter 2), and provides an introduction to the re-search area including both previous (linguistic) research on the Channel Islands and a brief outline of their sociolinguistic setting now and in the past (Chapter 3) This is followed by a detailed description of the study’s data and methods, which are, essentially, sociolinguistic in nature (Chapter 4)

After this background to the present study, the subsequent chapters deal with results and discussions of individual features of JersE taking into account, inter alia, possible sources and their sociolinguistic distribution within the speech community The focus is primarily on distinct or stereotypical JersE features

Three such features are investigated in greater detail: the discourse particle eh

(Chapter 5) and two particularities of the JersE verb phrase, i.e constructions of

the type He went and see them and existential there-constructions, among them those of the type There’s five years I work there (Chapter 6) These three features

have been selected for the following reasons: (i) all of them are clearly tive of JersE, yet illustrate different types of features with varying sociolinguistic distributions and identity-creating potentials, (ii) these features or their formal equivalents have been the subject of extensive research on either GuernsE, other varieties of English world-wide or both and (iii) their frequency of occurrence

distinc-in the data allows for a more detailed approach, which is not the case for all distinctive JersE features One chapter then provides an overview of further non-standard morphosyntactic and discourse features of JersE including a comparison

of JersE and GuernsE (Chapter 7)

In the penultimate chapter, the results of all morphosyntactic features found

in the data serve as the basis for a more general discussion of processes of ation and change currently in progress in JersE (Chapter 8) It is hoped that this

vari-bird’s-eye view of the data will further contribute to the overarching objective of this study, namely to account for patterns of linguistic variation and change in JersE and more generally Finally, Chapter 9 provides succinct summaries of the main empirical results, which are then brought together in a general conclusion that aims to answer the research questions set out in this introduction and suggests avenues for further research on JersE

chapter 2

Theoretical foundations

Positioned at the intersection of dialectology, sociolinguistics and contact tics, this study attempts to discern patterns of variation and change in the insular variety of JersE The Channel Islands present us, or so it seems, with laboratory conditions for investigating theories developed in these research strands There

linguis-is no denying the fact that the Islands’ speech communities can today no longer

be described as isolated – due to easy access to transport and growing immigrant numbers, both partly as a result of the development from a traditional rural agri-cultural community towards an expansion of the finance and service sector But their specific socio-historical setting in a confined space, where speakers adjust to both social and linguistic changes, makes them ideal for a study of language and dialect contact and variation In the spirit of William Labov, who was among the first to recognize the potential of post-war island communities for variation studies (cf Labov 1963), the present study sets out to investigate (mainly morphological and syntactic) variation in the English of Jersey and its neighbouring islands It thereby draws on theoretical frameworks in thriving research fields, which are closely interrelated and not always clearly distinguishable This chapter offers the theoretical backdrop to the present study to give a basic idea of its overall orienta-tion, rather than an in-depth survey of available models or the wide range of work which has been produced in the field of linguistic variation and change The focus will naturally be on how different frameworks and their major concepts and mod-els possibly relate to the situation of CIE More detailed and specific theoretical background will be given and discussed at greater length alongside the analyses

of individual linguistic features of JersE

This chapter starts with a short description of the study’s orientation within relevant fields of linguistic variation and change After discussing some meth-odological issues, I will focus on aspects of the investigation of morphological and syntactic variation more specifically Principles, mechanisms and models of linguistic change including standardization and levelling processes will then briefly

be outlined insofar as they are relevant to the underlying research questions of this study Three possible, and perhaps likely, factors in explaining varying and changing patterns in JersE will subsequently be addressed in further detail, namely linguistic contact, matters of identity and (linguistic) attitudes

2.1 Linguistic variation and change: Fields and methods

Language variation and change have been the focus of interest in linguistics for more than a century Fields and the terms used to describe them abound in this research area with a considerable overlap between concepts and models that are either more dialectological or sociolinguistic in orientation Whereas traditional dialectology (or dialect geography) still forms an important part of variation studies today (cf., for instance, Upton et al 1994; Trudgill 1999; Upton & Widdowson 2006), socio-linguistics as pioneered by William Labov has developed into a discipline of its own since the 1960s, whereby language variation and change has become one of the core branches of sociolinguistic research (cf Fasold & Tagliamonte 2003) While the boundaries of terms and fields are rather blurred (for a discussion see, e.g., Trudgill 1983; Krug 2000: 18–19; Cheshire 2003b; Preston 2004: 140), there is general agree-ment about the reciprocal influence of sociolinguistics (in a wider and narrower, i.e variationist, sense) and dialectology (e.g Trudgill 1983: 21–31) As both disci-plines are inextricably linked, distinguishing between different ‘sociolinguistic’ and

‘dialectological’ studies and research paradigms, which, additionally, can include historical, typological or anthropological approaches, seems more a matter of iden-tifying the objectives and orientations behind them (cf e.g Schneider 1996: 1) This

“growing union of dialectology and sociolinguistics” (Milroy & Gordon 2003: 21)

is also reflected in a number of publications which combine a compilation of a dialect’s phonological, grammatical and/or lexical inventories with an investigation

of (sometimes socially-driven) networks of variation and change in this specific dialect (e.g Sabban 1982; Hundt 1998; Sand 1999)

The present study follows in a similar vein: on the one hand, it draws heavily

on ideas developed by variationist sociolinguists as it makes use of their methods and investigates a few selected JersE features in detail thereby delving deeper into possible correlations of these variables with social or other extra-linguistic factors

On the other hand, it also integrates dialectological approaches in its attempt to present a first systematic description of JersE grammar and to compare this to accounts of GuernsE In the light of this study’s general objectives, therefore, it seems hardly fruitful to limit ways of explanation to a single framework Instead,

I will use multiple and at times competing methods and models to get as close to the locus of variation and change in JersE as possible (for a similar approach see also the introduction to Krug & Schlüter 2013) Further, some classical distinctions between traditional vs modern, rural vs urban dialects and approaches do not apply to the situation of the Channel Islands where – even more so than in other geographical areas – these terms are best represented as a continuum.3

3 Kortmann and Upton (2008: 25) also express their general scepticism about these distinctions.

The present study also puts a special emphasis on methodological questions Sociolinguistics (both in its wider and narrower sense) is a thriving field in linguis-tics and, as Fought (2004: 3) more recently points out, it is “in a process of rapid evolution, in the sense of both uncovering more information about previously documented linguistic patterns and studying new patterns that have only recently come into existence.” This process is closely connected to the continued develop-ment of research methodology A whole variety of methods – from acceptability and attitude surveys to participant observation and fieldworker interviews – has been in use in dialectology and sociolinguistics, most prominent among them is perhaps the sociolinguistic interview as developed by Labov Interview techniques

as well as methodology in general have been much refined and reworked over the last decades, profiting from the experience by sociolinguists and fieldworkers alike, who sometimes came up with ingenious solutions to methodological problems However, many challenges to sociolinguistic and more specifically variationist research persist, such as, for example, fulfilling requirements of sampling, avoid-ing the so-called observer’s paradox (e.g Labov 1972: 209) or identifying social categories and classifications in a speech community Such problems and possible ways of overcoming them will be discussed in Chapter 4

Increased emphasis has more recently been placed on sociolinguistic odology in a range of publications such as, e.g., Milroy & Gordon (2003), Bailey

meth-& Tillery (2004) or Tagliamonte (2006) Bailey and Tillery (2004) focus especially

on how to handle divergent data for linguistic phenomena – something which will also be important for the present study when comparing JersE data with two small-scale studies on GuernsE Presumably, divergent data (and results) often arise from methodological differences in collecting data Bailey and Tillery (2004: 28) rightly point out that “[d]isentangling the effects of our methods from the effects of social and linguistic factors with some certainty is perhaps the most important thing we can do to build upon the solid foundation laid by first-genera-tion sociolinguists.” The present study will attempt to contribute to the resolution

of this pending problem

As regards methodology, mention should also be made of multiple and plementary approaches in variation studies (dialectology included) In their in-

com-troduction to The dynamics of linguistic variation, Nevalainen et al (2008: 4) stress

how much variationist research has benefited from complementary approaches using methods developed both within the realm of sociolinguistics and corpus linguistics.4 Such multi-method approaches allow, for instance, a broader and

4 The question as to whether corpus linguistics should be regarded as providing important

methodology or as a discipline on its own will not be touched on here even though the former seems to be widely accepted today.

more unified study of language variation and change (see, e.g., Nevalainen et al 2008: 1–2) Nevalainen et al (2008: 4), presenting an overview of the history of linguistic (electronic) corpora, also point out that sociolinguists from the earliest hour have made use of corpora – be it text or rather speaker collections Depending

on the forms under investigation, however, corpora might not always be able to provide a large enough or adequate database and are often not suited to gaining sociolinguistic information, whereas a combination of other methods such as ac-ceptability and attitudinal questionnaires or controlled experiments might offer a solution to answer the question at hand (see also Mukherjee 2009: 91) The present study, which looks predominantly at rarer linguistic features and structures, will pursue such a strategy of combining methods (and thereby follow the trend in variationist studies just outlined) insofar as it uses an electronic, searchable corpus compiled from transcribed sociolinguistic interviews as well as results gained from participant observation and a questionnaire survey

In addition, a growing number of corpora compiled with a variationist

back-ground in mind (such as the International Corpus of English for example) help to

lend a new quality to the comparative study of varieties across the world, ing to compare varieties on all kinds of levels, regional, social, stylistic or other (cf Fasold & Tagliamonte 2003) Yet, the existence of so many corpora has also sharpened the problem of what to compare and what to measure a dialect feature against This will, among other things, be discussed in the next section

allow-2.2 Morphological and syntactic variation

Early variationist research focused mainly on phonetic and sometimes on phological variation Therefore, many concepts and methods originally developed

mor-in sociolmor-inguistics and dialectology are better adapted to the purposes of studymor-ing these language levels More recently, however, many researchers working in the variationist framework have tried to redress the balance by investigating a broad range of linguistic phenomena including many at higher linguistic levels such as syntax and discourse For instance, Trudgill and Chambers (1991), Milroy and Milroy (1993), Cheshire and Stein (1997a) and Kortmann et al (2005), amongst others, have done much to remedy the situation so that research into the gram-mar of English dialects, in particular, has been “on the rise” since the 1980s (cf Kortmann et al 2005: v; see also Trudgill & Chambers 1991: 1) While at the be-ginning of the 21st century the study of variation in dialect grammar was still de-scribed as a desideratum in linguistics (cf Krug 2000: 20; Anderwald & Kortmann 2002: 161), it is generally blossoming as a field of research today Yet, this develop-ment is still recent enough to have left unfilled many important research areas (cf

also Leech et al 2009: 7, who focus on grammatical change in standard English)

In what follows I will outline some of the problems linked in particular to research

on syntactic variation, which point to reasons why this area in language variation and change had for so long been neglected and which are taken into account in the analyses in the present study

A first problem area concerns methodological issues although we have to tinguish carefully between morphological variation and syntactic phenomena It

dis-is only in the former domain that the application of sociolingudis-istic methods can sometimes be as straightforward as for phonetic variation As regards syntactic forms, the variable rule, for example, has been much debated (see, e.g., Milroy & Gordon 2003: 169–197 or Smith 2007: 30) There is, for instance, the problem of which forms or variants of a variable can be regarded as functionally equivalent (see also Cheshire 2009): “What to include and what not to include can seem like

a minefield,” Smith (2007: 32) tellingly points out Thus, the analysis of syntactic variation also involves a constant process of counting and recounting forms, as where a form or structure is used can often be as important as where it is not used However, it is sometimes extremely difficult or quite impossible to account for all potential occurrences of syntactic features and to observe what Labov termed the ‘principle of accountability’ (cf Labov 1972: 72) It is also a controversial issue determining how frequently a syntactic or morphosyntactic form should occur in

a database and how many instances per speaker are needed to obtain (statistically) reliable results (cf Smith 2007: 30) It is clear, however, that syntactic investiga-tions require a much larger amount of data than phonetic ones Some estimate that about 40 times as much text is needed (cf Kortmann et al 2005: 5) This problem

is carried to the extreme when rare syntactic forms are under investigation, which can probably only be approached using multiple methods Carruthers (1999), for instance, outlines convincing ways of overcoming obstacles in examining the rare

use of the passé surcomposé in French varieties.

A second problem area concerns the claim that syntactic variation in trast to phonological variation rarely serves the function of distinguishing social groups but is rather subject to internal, cognitive and situational constraints.5 The relative infrequency of some syntactic structures has sometimes been held re-sponsible for this, yet Jenny Cheshire (2003a, 2009) argues strongly against this position While admitting that syntactic variants do serve different purposes and, more importantly, function in other ways than phonological and morphosyntac-tic ones, she suggests that syntactic variants “may distinguish social groups […] They do not simply index membership in these groups; instead they may indicate

con-5 For a summary of this discussion and references to such positions and studies, see Milroy

& Gordon (2003: 197) and Cheshire (2003a: 245, 2009).

deep-seated differences in the way in which different social groups create their cial worlds” (Cheshire 2003a: 260) At the same time, she emphasizes the necessity

so-of taking context and meaning into account when examining syntactic variation (cf Cheshire 2003a: 259) The present study will follow this guideline, which, of course, was also previously stated by others (see, e.g., Holmes 1990), and show how syntactic alongside other grammatical variation can indeed help to indicate the social relations within a speech community

Furthermore, a study in morphological and syntactic dialect variation needs some sort of reference point against which variation can be analysed and measured (cf Hundt 1998: 9) However, it is less clear what this reference point in general and for JersE in particular should be:

To categorise a feature as ‘nonstandard’ is to treat it as if it has a ‘standard’ lent, but whilst we may know what the standard equivalent is for some morpho-logical features […], the general concept of a ‘standard’ in language has never been clearly and satisfactorily defined (Cheshire & Stein 1997b: 1–2)

equiva-Lesley Milroy (2004: 162) also judges that the concept of the standard is ingly underspecified and undertheorized” and, what is more, very problematic in itself, for the term ‘standard’ will always refer to an idealized construct which, with all the uniformity it implies, is unlikely to exist Yet, Stein (1997: 35) argues that

“surpris-“[t]he study of dialect grammar can never be the study of dialect grammar alone, but always the study of dialect grammar in its dialectic with the study of standard grammar” He perceives the difference between dialect and (written) standard as

a continuum With some additions this is the present study’s stance too As JersE

is not formally codified and it is highly questionable whether any homogeneous form of a local JersE standard exists to date, the term ‘standard’ in this study re-fers to standard British English, while bearing in mind that a ‘standard’ cannot be anything but an idealized notion

To complicate matters further, one should also consider cases where ties cannot be clearly placed within the British or American context.6 Hundt (1998: 9–18) in her study on morphosyntactic variation in New Zealand English, for example, follows a dynamic model of English as a pluricentric language with several interacting national varieties Anderwald and Kortmann (2002) on the other hand suggest a typological approach whose established hypotheses and insights into language variation provide a comparative framework against which dialect data can be measured.7 To this end, Kortmann and his associates have

varie-6 On the more complex role of standardization in postcolonial settings, where it is often linked

to matters of politics and identity, see also Inoue (2006)

7 For a similar position see also Sand (2005).

also compiled the Freiburg English Dialect Corpus (FRED), which permits the

study of cross-dialect variation.8 In accordance with this approach, the present study will also analyse and discuss JersE features from a cross-dialectal perspec-tive – simultaneously referring to the (codified) standard of English, as well as, more generally, to English varieties across the world As will be seen, the linguistic sphere of influence for the Channel Islands, apart from its contact with French va-rieties, lies in the South of England and the Islands’ norm is oriented towards the British rather than the American or any other nation’s norm, although historically speaking CIE has without a doubt been influenced by the diverse dialects of im-migrants, mainly from the British Isles.9 This orientation also seems to be reflected

in a process of standardization JersE is currently undergoing Standardization is defined here as the loss of non-standard features in favour of establishing standard British English forms, therefore as a process of rendering the variety of JersE more similar to standard British English Matters of norm-orientation and standardi-zation in JersE and CIE in general will be pursued further in Chapter 8, while methods used to map variation in JersE against other varieties will be described

in Chapter 4

It is also important to note that most of the morphosyntactic and syntactic variation analysed in this study of JersE will not necessarily represent unique fea-tures Many English dialect features, such as multiple negation or the use of singu-

lar forms of BE in plural there-existentials, are widespread in many non-standard

varieties (cf Cheshire et al 1989; Kortmann & Szmrecsanyi 2004) Differences in dialect grammars, therefore, may instead manifest themselves in the combination, frequency and pragmatic use of grammatical features in a dialect area Just as boundaries between dialect areas are blurred and floating rather than clear-cut and fixed (cf e.g Kortmann & Upton 2008: 24–25), the uniqueness of a dialect’s feature

is often reflected in (statistically showing) preferences and tendencies rather than categorical differences (see also Krug 2000: 19; Kortmann & Upton 2008: 24–25) The same can be said for observations of linguistic change where few categorical losses or additions have occurred in the 19th and 20th century: “syntactic change has more often been statistical in nature, with a given construction occurring throughout the period and either becoming more or less common generally or

in particular registers.” (Leech et al 2009: 8) The following section will provide a sketch of those principles and models in linguistic change which will be important

to the discussion of JersE

8 The FRED-project, its aims and approach are described in Kortmann et al (2005).

9 A short overview of Jersey’s immigration history will be given in Section 3.2.

Most importantly, the principle that linguistic variation is not random, but possesses ‘orderly heterogeneity’ – as famously expressed by Weinreich, Labov and Herzog (1968) – and is a prerequisite for linguistic change also underlies the present study Whilst the problem of actuation, i.e who initiates linguistic change and for what reason, remains as puzzling as ever (cf McMahon 1994: 252), (socio)linguists

by now have gained an idea of how linguistic changes spread through language and society, i.e how changes are transmitted and embedded, and developed models of diffusion such as, for example, the wave model or gravity model None of these dif-fusion models will be straightforwardly applicable to the situation in the Channel Islands, however As a vibrant financial centre Jersey attracts employees and clients from all over the world while, at the same time, keeping the rural character of an agricultural and remote society Within the island, more or less densely populated areas do not necessarily correspond to a more urban or rural lifestyle of their resi-dents The situation thus is more complex with a speech community that cannot

be labelled homogeneous and whose speakers do not necessarily share the same way of life or a similar set of norms and values.11 Note that the speech community

of Jersey is here roughly equated with the residents of Jersey, although doubtlessly, not everyone on Jersey speaks a local form of English and even among those born

on the island a homogeneous dialect is unlikely to be found

In contrast to the concept of such a speech community in the Labovian sense, the network theory of the Milroys (cf Milroy 1980; for a concise summary see also Milroy & Gordon 2003: 116–130) might be more useful in representing the sociolinguistic situation in Jersey In a nutshell, this theory claims that social ties among speakers and their relative denseness and multiplexity influence linguistic

10 The most complete overview to date can certainly be found in McMahon (1994).

11 The general situation in Guernsey seems to be very much the same, although the island has

perhaps preserved a more rural, traditional character than Jersey.

change Uniplex and loose ties have been found to favour linguistic change This network approach is particularly useful in settings where the concept of social class is somewhat less clear-cut as for example in smaller and more rural commu-nities without heavy industries such as the Channel Islands (cf Milroy & Gordon 2003: 120) Jersey has undergone dramatic social changes from an almost exclu-sively agricultural to a service and finance economy; this will be shown to be reflected in changing personal network structures as well

Closely related to the network approach – mainly different in method and focus – Eckert’s (2000) research attempts to identify ‘communities of practice’ which are defined by shared practices of their members Her approach is grounded

in ethnographical fieldwork and participant observation, which helps her to lish categories and groups that are not in any way pre-defined She discusses the advantages and disadvantages of such an approach, which allows a very detailed and complex investigation but is less likely to obtain any of the broader results found in large-scale studies (cf Eckert 2009) In the Channel Island context, her approach is certainly valuable for unveiling social networks and bonds that are important for speakers’ choices The insular situation in general seems to be better studied from a network perspective: Wolfram and Schilling-Estes (1996) in their aforementioned study of the Ocracoke dialect found, similarly, that other factors than the ‘usual’ ones correlate with language use on the island: “group affiliation, communication networks, social identity, and social context all come into play in determining the role of the dialect variable in the Ocracoke community” (Wolfram

estab-& Schilling-Estes 1996: 107)

In most cases, it is not possible or practical to carry out real-time research Another important concept for the present study is, therefore, the apparent-time construct, which claims that linguistic differences among different age-groups of a speech community indicate change in progress and mirror actual diachronic devel-opments in the variety when stylistic and social factors are held constant Yet, the underlying assumption that speakers’ linguistic behaviour is rather stable in their lifetime is not unproblematic as Llamas (2007: 73) points out Age and language certainly do form a complex relationship Nevertheless, Bailey (2004) shows that apparent-time studies can serve as an excellent substitute for real-time evidence, although he also stresses the importance of excluding teenagers from apparent-time studies to minimize the risk of age-grading (see also Sankoff 2004) In his eyes the ideal case consists of a combination of apparent-time data and real-time evidence (cf Bailey 2004: 330) The present study will accommodate his findings

by (i) collecting data from adults only and (ii) integrating a real-time component into the study with the help of archive data

Motives for language change can be manifold Linguists have, for example, uncovered contextual, cognitive and psychological, iconic, physiological and social

motivating factors shaping linguistic change Some linguists have argued for the superiority of internal explanations versus external ones, though there is no proof that the former are necessarily more important (cf Jones & Esch 2002: 7) Many have thus criticized the strict dichotomy of external and internal explanations and warned of an ‘either-or’ mentality (cf Jones & Esch 2002: 3; see also Dorian 1993; Wolfram & Schilling-Estes 1996: 143–144) Jones and Esch (2002), discussing the interaction of language-internal, -external and extra-linguistic factors in motivat-ing language change, convincingly point out that

[l]anguage change is a complex phenomenon This may seem like stating the ous, but it has one important implication, namely that explanations for a change will, and should, reflect this complexity A full explanation for a change is therefore unlikely to involve citing one motivating factor but, rather, will invoke a large number of interacting factors, the relationship between which we may only be able to guess at, at least for the time being If we continue to focus only on one of these factors, then we are surely missing an important part of the jigsaw

obvi-(Jones & Esch 2002: 8)

Therefore, even in a former high-contact situation such as in the Channel Islands where explanations of linguistic variation and change seem to be readily available, this study will consider all possible motivating factors and the possibility of inter-acting and/or reinforcing motivations (cf Jones & Esch 2002: 13, 143; Thomason 2003: 688) Yet, linguistic contact with Norman French and other English dialects could certainly emerge as one of the major driving forces behind the processes and patterns to be revealed in this study on JersE Earlier studies on CIE indeed suggest that features occurring in CIE are often contact-induced The following section, therefore, focuses on concepts of contact linguistics relevant to the present study

2.4 Linguistic contact

Theories of linguistic contact, be it dialect contact or contact between languages that are typologically very different, have been systematically exploited in varia-tionist research designs Linguistic contact serves as an umbrella term for a range

of phenomena which affect fields as diverse and interdisciplinary as language quisition, the birth of dialects and languages, discourse, language policy, typol-ogy, language change, etc Hence, important publications in this research area, which also provide overviews of the field, often have a different emphasis (see, e.g., Thomason & Kaufman 1988; Silva-Corvalán 1996; Thomason 2001, 2003; Winford 2003; Matras 2009) In this section, only concepts and findings in contact linguis-tics which are important to the situation of JersE will be briefly listed

ac-JersE – as well as CIE in general – can, from a historical perspective, be acterized as a high-contact variety even though the shift towards English is by now almost complete so that Jersey does not have a significant number of L2, i.e native Jèrriais, speakers any longer According to Mesthrie’s (1992: 2–5) catego-rizations of types of English(es), JersE can be defined as a language-shift variety

char-or as a nativized L2 variety Yet, as the discussion in Kchar-ortmann and Szmrecsanyi (2004: 1184) also suggests, a fuzziness of distinctions between such variety cat-egories certainly remains

One of the most recent and, to date, most noteworthy models for a zation of post-colonial Englishes (PCEs) is put forward by Schneider (2007) His Dynamic Model of the evolution of PCEs, embedded in the framework of contact linguistics, claims that a very similar, though to a certain degree obviously ide-alized, process underlies all historical stages of PCEs in different world regions regardless of their outward dissimilarities (cf Schneider 2007: 21) The model iden-tifies five progressive stages in the development of PCEs from the perspective of both the colonizers and the colonized Above all,

categori-it rests upon the assumption that, in selecting from this pool [a feature pool of possible linguistic choices, AR], speakers keep redefining and expressing their lin-guistic and social identities, constantly aligning themselves with other individuals and thereby accommodating their speech behaviour to those they wish to associate

Although it is questionable whether CIE actually qualifies for the label of colonial English’, given the Channel Islands’ unique history as dependencies of the British Crown in Europe, their early administrative autonomy with regard to local laws and tax regulations and their close affiliation with Britain, it does share the specific type of language-contact situation resulting from the spread of English to new territories Therefore, the Dynamic Model should be a valuable tool in tracing the emergence of CIE and determining its current status

‘post-Following the synopsis of Schneider’s (2007: 56) model, it seems reasonable to propose that CIE has undergone the first three of the five developmental phases, i.e ‘foundation’, ‘exonormative stabilization’ and ‘nativization’ As will be seen, some aspects of phase 4 (‘endonormative stabilization’) can also be found in the Channel Islands, but it is here that the categorization becomes more problematic Whereas the listed linguistic developments of phase 4 clearly do not fit the Channel Island context, since CIE can neither be described as undergoing stabilization nor codification, we might yet be witnessing the emergence of an identity-creating function and a positive attitude towards a local variety of English (for a discussion

of this see especially Chapter 8) There is also some, though not much, evidence of literary creativity in CIE In other words, I shall argue that CIE is at present best

described as a language-shift variety that has completed, in Schneider’s terms, the phase of ‘nativization’, but which borders on the next phase as regards identity construction and linguistic attitudes.

Whereas the future of CIE and JersE in particular seems less certain, a long tory of contact between various dialects of English, a Norman French variety and

his-to some degree also standard French has undoubtedly contributed his-to the shape of JersE (and of course Jersey French) today, but judging to what extent is one of the central research questions of this study and will be systematically investigated.12

Even though Jèrriais is regarded as an obsolescent variety at the beginning of the 21st century, Jones (2002: 143) carefully argues for “a period of convergence prior

to the onset of obsolescence, when a period of relatively stable bilingualism may have resulted in more bi-directional influence.” Convergence is usually said to occur in situations of widespread bilingualism when the languages involved are held to be of equal prestige According to McMahon (1994: 213), it leads to mutual, rather than unidirectional, influence on lexical, but also syntactic and morpho-logical structures, especially when contact has been intense for some time More recent studies have also suggested that linguistic influence in a contact context can principally occur in both directions and to differing degrees, depending on situational and social circumstances (cf e.g Matras 2009: 311–312) Linguistic contact then may indeed explain the existence of and preference for some gram-matical features in JersE.13

Contact has also always been both direct and indirect insofar as bilingual speakers constitute the locus of direct contact while monolingual, i.e English-only, speakers come into indirect contact having acquired passive familiarity with Norman French (see, for example, Chapter 6 and also Ramisch 1989: 63) Additionally, a linguistic account of the Channel Islands has to consider both lan-

guage (English and French varieties) and dialect (various varieties of English)

contact.14 The situation is therefore complex, which makes the question of induced variation and change all the more intriguing

contact-I will use the term ‘transfer’ as a broad description for all forms of influence arising from this specific contact situation being aware, however, of the ongoing

12 For a more detailed description of linguistic contact in the Channel Islands see Jones (2001)

for Jersey, Ramisch (1989: 72–75) for Guernsey and also the relevant sections of Chapter 3.

13 There will be no focus here on the consequences of this prolonged contact for Jersey French

or for matters of code-switching, but relevant studies have been carried out by Jones (2001, 2002, 2005a, 2005b)

14 Trudgill (1986, in particular Chapter 2) provides a good starting point for a discussion of

mechanisms and principles in dialect contact.

debate about terminology in this area.15 Transfer phenomena are often classified according to language levels Further, borrowing or transfer hierarchies have been established to show the likelihood that some forms or structures are transferred before others (cf e.g McMahon 1994: 204–209; Harris & Campbell 1995: 120–150; Matras 2009: 153–154) In the area of morphology and syntax, for instance, less tightly integrated items and structures, like specific word-order patterns, tend to

be transferred and borrowed first (cf Thomason 2001: 69–71) One has to bear

in mind, of course, that such hierarchies do not necessarily apply in all contexts

as many other factors can potentially interfere (cf Thomason 2001: 71; Matras 2009: 154) As mentioned above, morphological and grammatical structures, the focal areas of the present study, are usually transferred after lexical items, but can also be affected in prolonged contact situations as in the Channel Islands Thomason (2003: 691) even states that “if people who are not fluent speakers of A introduce features into A from another language, B, the first interference features (and usually the most common ones overall) will not be lexical, but rather pho-nological and syntactic.”

The kind of transfer she describes seems inextricably linked with transfer in guage acquisition processes and so-called “imperfect group learning” (Thomason 2003: 692) Just as in individual second language acquisition (SLA), transfer can then take the form of substitution, over-/under-differentiation and over-/under-representation of structures, as well as word-for-word translations (see, e.g., Gass

lan-& Selinker 2001) Subsequently, monolingual speakers of the target language – in the case of Jersey this would be monolingual speakers of English – might also bor-row these transfer features In Jersey, this can, in principle, occur either through passive familiarity with Jèrriais, or through the English speech of predominantly Jèrriais speakers Both possibilities are, on a general level, proposed by Thomason (2003) in her article on contact as a source of linguistic change Thus, after all, not only might speakers of Jèrriais have adjusted their speech to what they believed to

be possible structures and patterns in English, but also English-only speakers on Jersey might, in the long run, accommodate to shift-induced transfer phenomena Odlin (2009) describes methods of determining the likelihood of substrate influence in contact varieties involving comparative (statistical) methods as first established in SLA research This basic comparative approach (in this case between English, Jèrriais and local features of JersE) will also be adopted in the present study, though due to the rare occurrence of some forms, more fine-grained statisti-cal evidence is often not available, which, as Odlin (2009: 270) concedes, does not necessarily reduce the likelihood of a transfer explanation

15 For details of this debate see Winford (2003: 16–17) or Matras (2009: 146–147).

If, in addition, the transfer phenomena found in contact varieties resemble, at least partly, those found in individual SLA, then for any such study it should also

be particularly interesting to compare the variety under investigation with opments in those varieties which arise from a contact scenario involving the same languages or, at any rate, regional varieties of them: a number of similar, though not necessarily fully identical, linguistic features should be found in such a compari-son Although from a diachronic viewpoint English and (Norman) French share

devel-“a long love story” (cf Walter 2001), there are today not many regions in the world with intensive contact between varieties of French and English – Canada, Quebec especially, and Louisiana being famous exceptions.16 Studies on Quebec English (see, e.g., Bagola 2005; Fee 2008; Poplack 2008) and Cajun English (see, e.g., Dubois

& Horvath 2000, 2003, 2008; Wolfram & Schilling-Estes 2006: 202–205) will be referred to throughout the present study to provide further evidence for a contact-induced explanation of the emergence of individual JersE features

Surprisingly, it seems that the assumption that Quebec English is a distinctive, French-influenced variety cannot be supported by quantitative evidence: according

to Poplack (2008: 197), “systematic quantitative analysis of a large representative corpus of spontaneous Q[uebe]cE[nglish] speech offers no support for claims that QcE differs from other varieties of Canadian English as a result of its minority status and sustained contact with French.” However, small differences in lexical items can be found and there are indeed various investigations of the discourse

particle eh in Canadian English, whose findings will be compared to those of eh

in JersE and in other varieties of English where eh is also prominent, such as New

Zealand English Unlike Quebec English, Cajun English and JersE seem to invite comparison at a grammatical level, for instance as regards some features of the verb phrase, the use of the definite article or of emphatic pronouns Sociolinguistic patterns identified in Dubois and Horvath (2003), such as correlations between the use of non-standard features and a speaker’s age, level of education and linguistic background, will be compared to the patterns found in the present study as well Recently, contributions in Filppula et al (2009a) have discussed if and how other factors such as vernacular or typological universals and ‘natural’ and general tendencies in language acquisition processes interact in the development of contact varieties or if these factors can be distinguished from each other at all (see, for instance, Thomason 2009; Winford 2009) Sand (2005) also suggests that it might

be promising to investigate typological universals as an explanation behind parallel features in contact varieties – a hypothesis that will be considered for some features

of JersE too (see Chapters 6 and 7)

16 Contact scenarios between French and English, though not exclusively, can also be found

in multilingual communities in a number of African countries

As to the realm of contact linguistics, mention should also be made of modation processes underlying tendencies for varieties to move towards a ‘stand-ard’ of English and to become less distinctive It will be analysed and discussed

accom-in Chapter 8 below if such tendencies can also be found on the grammatical level for the JersE dialect Linguists have often remarked on the decrease of linguistic diversity, on growing standardization and homogenization that followed, among other things, increasing mobility, better education and the spread of mass media

in the second half of the 20th century (cf Williams & Kerswill 1999: 149; Trudgill 2002: 29–30; Kortmann et al 2005: 5)

According to Kerswill (2003: 239), regional dialect levelling and the erosion

of traditional dialects are widespread in the British Isles Levelling is usually fined as “the eradication of socially or locally marked variants (both within and between linguistic systems) in conditions of social or geographical mobility and resultant dialect contact” (Milroy & Gordon 2003: 130; for a similar definition see also Trudgill 1986: 98) This study adopts the definition of levelling as involving the reduction, or even loss, of marked, distinctly local dialect features Levelling processes in JersE, according to this definition, would therefore entail the lessen-ing of differences between JersE and other (British) English varieties, possibly also accompanied by an increase in widespread supralocal features This could also apply to cases where a supralocal feature which once used to differ quantitatively

de-or qualitatively in JersE now tends to occur in JersE in much the same way as where in the British Isles It should be added that this definition of levelling refers

else-to dialect levelling only and should not be confused with levelling as a specific term for the language-internal reduction of variation in morphological paradigms (used this way in Chapter 6) Kerswill (2002, 2003) suggests two possible mechanisms underlying dialect levelling in British English and concludes that both have to be taken into account, namely (i) geographical diffusion by which features diffuse from a densely populated and socio-economically leading centre and (ii) level-ling which implies the loss of marked variants resulting from short- or long-term speech accommodation Levelling seems to be favoured in contexts of high mobil-ity in relatively compact areas with much contact between speakers

Auer (2007) also discusses the role of long-term dialect accommodation as a consequence of migration in linguistic change.17 In a scenario where immigrants

17 Accommodation should be distinguished, as Auer (2007: 110) points out, from koineization,

i.e the process of dialect mixing which leads to the formation of a new dialect (see also, e.g., Tuten 2007: 185–186) The Channel Islands have experienced a long and constant influx of im- migration from all parts of the British Isles and beyond The island-born population, however, has remained relatively stable over time Koineization, therefore, does not seem to explain the present situation in the Channel Islands.

share the standard variety with the receiving region, but also speak their own gional dialect – and this partly resembles the immigration situation in the Channel Islands today – standardization and levelling processes can be a direct consequence (cf Auer 2007: 110–111) According to Auer (2007: 112), network structures or rather their breakdown and reconstruction could be factors influencing these pro-cesses (see also Section 2.3 above) These ideas should prove very useful in a more detailed investigation of general linguistic tendencies in JersE, such as, possibly, levelling, standardization and their influencing factors (see Chapter 8).

re-2.5 Identity and attitudes

Levelling and dialect death as mentioned in the previous section have also been found to be linked to negative attitudes towards regional varieties in many parts

of Britain and elsewhere in Europe:

The connection between dialect death, on the one hand, and the denigration of vernacular varieties, on the other, is clear If we wish to maintain linguistic di-versity and oppose linguistic homogenisation […] we have to consider speakers’ attitudes to their own dialects

(Trudgill 2002: 30; in a similar vein see also Trudgill 1983: 186–200)

Speakers’ attitudes seem to be influenced by ideas of the standard that they believe

in and – sometimes vehemently – defend (cf Milroy 2007) Since Labov’s (1963) study of linguistic variation on Martha’s Vineyard, factors of identity and attitudes have been an integral part of (socio)linguistic research Ideas of covert and overt prestige, i.e speakers’ orientation towards local identity and solidarity factors or towards linguistic forms valued in education and by the middle-class, have been employed by sociolinguists for a long time

Among the first to posit a model of identity in linguistic variation were Le Page and Tabouret-Keller (1985) with their theory of acts of identity.18 In the same year

as their publication, John Edwards’ (1985) monograph on links between language, identity and society appeared Eckert (2000) in her study of language variation in the speech of high school students introduces the term ‘community of practice’ (see also Section 2.3 above) and suggests viewing the linguistic construction of identity as a result of shared (linguistic and stylistic) practices within such ‘com-munities’ where people take part in the same activities and have similar values

To date, a range of textbooks and articles are available that provide overviews of developments in the field (see, for instance, Tabouret-Keller 1997; Joseph 2004; Dyer 2007; Edwards 2009; Llamas & Watt 2010)

18 For a short summary of their ideas see also Sand (1999: 62–67).

What exactly is meant by ‘identity’, however, varies greatly between linguistic studies and is often not further theorized (cf Bucholtz & Hall 2010: 18) In so-ciolinguistics, identity is often construed as a form of social identity defined by group membership, for example in a particular social class Establishing such group memberships then often involves the assumption of more abstract, pre-existing categories By contrast, many scholars today consider identity as a social phenom-enon that, rather than being stable, emerges and forms in social interaction and is constantly shifting (cf., e.g., Bucholtz & Hall 2010: 18) While the present study in part also relies on pre-supposed speaker categories (such as age or socio-economic background) for its analyses and data interpretation, the definition of identity is for the purpose of this study quite restricted as it is largely understood as a speaker’s sense of belonging to Jersey and to the island community This includes identifi-cation with local values and norms and positive attitudes towards the island and island life in general On an island like Jersey, which is steeped in a long history

of cultural and linguistic contact, a local identity can take on many facets as will

be shown in Chapter 8

In principle, identity as a factor in linguistics can work both ways: on the one hand, language crucially contributes to creating a speaker’s identity; on the other hand, identity and attitudes (to language) may shape language in determining speakers’ choices As a rule, language users can draw on different linguistic forms from their sociolinguistic repertoire depending on the situational context, their individual preferences and their wishes to mark their sense of individual or social identity (cf Llamas & Watt 2010: 2, see also Coulmas 2005: 8–11) When refer-ring to speakers’ choices in this study, it is usually implied that these are choices between different variants of a morphosyntactic variable Such variants can be more or less formal, more or less standard, distinctly local or rather widespread features of colloquial English Although such choices can be intentional, they are not necessarily conscious choices, a fact which should be taken into consideration,

as Williams (1992: 66) urges, in any study of the relations between language ation and the negotiation and construction of identity As regards the concept of identity then, the focus of this study is to determine how an individual speaker’s identity influences his or her choices As will be seen, all informants in this study share a distinctly Jersey background so that, seen through a wider lens, their group identity as Jersey islanders seems homogeneous, yet differences are revealed when their individual identity is brought into close focus Both group identity and per-sonal identity will therefore be examined in the analytical chapters

vari-While many have emphasized the intimate relation between place, language, identity and language attitudes (see, e.g., Edwards 1992; Trudgill 1999: 1–2), the role of identity as a factor in language change is, it seems, more controversial Schneider (2007: 30), for example, considers the emergence of post-colonial

English varieties as a “largely identity-driven process of linguistic convergence” Trudgill (2004: 156–157) in his account of new-dialect formation, on the other hand, is much more sceptical about the factor ‘identity’ in linguistic change argu-ing that identity phenomena occur much less frequently than might be expected Whereas the body of research that deals with identity and attitudes has been con-stantly growing for the last three decades, especially in discourse analysis (see, e.g., Duszak 2002 and De Fina et al 2006), not so much research has been car-ried out in accounts of language variation and change, as L Milroy (2004: 161) points out Perhaps Trudgill (2004: 156–157) is right in suggesting the rarity of identity-driven change, however, the lack of evidence might also partly be due to methodological issues Not only are identity and attitudes constantly in flux, they are also mental constructs which we cannot be sure to be capturing correctly (cf Garrett 2007: 116) However, some methodology to overcome this obstacle has been developed, such as the matched-guise technique by Lambert (1967), attitu-dinal questionnaires and other fieldwork methods The present study will try to partly remedy the situation pointed out by Lesley Milroy (2004) by looking into the links between attitudes, identity and language use in Jersey and their implica-tions for JersE Attitudes towards JersE features and towards Jersey in general and the strength of local identity will be elicited from questionnaire and interview data

so as to assess the potential of individual features for marking identity and, more generally, the influence of attitudes and identity on linguistic changes in JersE One chapter of the study is therefore devoted to matters of identity and attitudes

in Jersey within a more general approach to the data (see Chapter 8) Among other things, it will investigate whether a sense of local identity prompts speakers to prefer more traditional JersE forms and, if so, whether this link is straightforward and likely to help to preserve more traditional, local variants – as could be shown

in other studies

This chapter has attempted to set forth the major theoretical foundations of the present study by providing a brief summary of the most important frameworks and models with possible links to the study of variation and change in JersE As will be clear, the many concepts and ideas underlying the present study constitute

a tightly interwoven network These will be further explored and illuminated in the analyses of JersE features, which assess their meaning in relation to more general linguistic and social processes currently in progress on the island

chapter 3

Jersey English in context

Following the sketch of the study’s theoretical foundations, this chapter provides the relevant background of the research area by outlining previous research and highlighting the most important aspects of the history and (socio)linguistic situ-ation of the Channel Islands In line with the present study’s focus, the Island of Jersey will be at the centre of this overview As has been emphasized above, it

is crucial to take the socio-historical context into consideration when ing a study in a language contact setting that is also sociolinguistically inspired (cf Wolfram & Schilling-Estes 1996: 143) In order to interpret findings it will be necessary to consider a range of possible factors such as, for example, linguistic-systemic, cognitive or psychological as well as historical and social considerations The subject areas described in this chapter will, therefore, resurface in discussions

conduct-of the empirical findings (Chapters 5–8)

3.1 Previous research

Linguistic research on the Channel Islands has a twofold focus Whereas the Norman French varieties have by now been relatively well documented and studied, the English variety (or varieties) spoken on the Channel Islands has sparked comparatively little interest in linguistics, especially over the last two decades A more recent trend, however, turns to and engages with lesser-known varieties of English, and in this framework Mari Jones (2010) gives the most up-to-date, albeit short, overview of CIE The present study will follow her lead in its attempt to provide a first comprehensive and systematic description of JersE Whilst no such account of JersE so far exists, there are, however, two PhD-theses about GuernsE, both focusing on morphosyntax: Heinrich Ramisch’s (1989)

The variation of English in Guernsey/Channel Islands and Pauline Barbé’s (1993)

“Exploring Variation in Guernsey English Syntax” As the only in-depth ies of a CIE variety Ramisch (1989) and Barbé (1993) will be summarized in greater detail below Yet before turning to research on CIE, let me briefly list

stud-some general, non-linguistic literature about the Channel Islands and previous research on insular Norman French that this study will also draw on to place the results of data analyses in context.19

3.1.1 Non-linguistic research relevant to the study

There are a number of publications on the Islands’ local history and cultural tage useful for tracing the historical development of the linguistic situation in the Channel Islands as well as for gaining a better understanding of immigration movements to the islands The relevant contents of these publications will be sum-marized in Section 3.2 below and be referred to throughout the main part A va-riety of monographs, each with a specific emphasis, give historical overviews (see, e.g., Uttley 1966; Guillot 1975; Lemprière 1990; Canu 1996) The most comprehen-sive account of the history of Jersey can be found in Syvret and Stevens (1998).20

heri-Of special interest are also works focusing on a particular time-span in Jersey or Guernsey and which allow us to chart demographic and economic changes, such

as Nicolle (1991) on the 17th and 18th centuries in Jersey, Kelleher (1991) on the rural community in 19th century Jersey or Crossan (2007) on economic and de-mographic developments in Guernsey during the 19th century

Up-to-date statistics on the islands’ population, their social and economic fare and numerous other aspects of island life are produced by the Statistics Unit

wel-of the States wel-of Jersey and the Policy and Research Unit wel-of the States wel-of Guernsey.21

These units were also responsible for the analysis and publication of the 2001 sus data, which, alongside many other surveys and statistics, are available online and constitute a valuable source of information for the present study Census data from 1851 onwards can also be accessed (though not electronically) in Jersey’s Lord Coutanche Library

cen-Older accounts of the political, cultural and economic situation in Jersey can be

found, e.g., in Hugo’s (1985) L’Archipel de la Manche for the 19th century, in historical

overviews, such as Dicey (1751), or in travel guides and other reports from the 18th, 19th and early 20th centuries (see, for instance, Stead 1809 or Inglis 1834) These often include some remarks about the languages spoken in the Channel Islands thereby conveying, indirectly, an idea of how the linguistic situation developed

19 Coates (1998) compiled a bibliography of Channel Islands French and the general linguistic

situation in the islands up to the year 1997, which is difficult to get hold of, however, as it is only accessible via the library of the University of Sussex

20 Stevens (1985) represents a short form of this work

21 The States of Jersey and the States of Guernsey are the islands’ governmental bodies.

Additional sources of information on Jersey identity and cultural heritage are provided in some more popularly written books on Jersey culture (see, for exam-ple, L’Amy 1983; Le Dain 1997 or Pittman 2003) and in a few unpublished theses such as the PhD-thesis by Annette Spurway Torode (2001) and the MA-thesis by Daphne Coutanche (1978).

3.1.2 Research on Channel Island French

Linguistic research on the Channel Islands has mostly centred on insular Norman French This line of research is nevertheless important to the inves-tigation of JersE insofar as transfer from Norman French has been established

in the literature as one of the major explanatory factors behind shaping CIE and, consequently, information on the grammar of Jersey Norman French will

be needed and referred to throughout this study Moreover, most publications

on Norman French varieties also trace the history of linguistic development in the Channel Islands offering some comments on the English variety or varieties spoken there A range of publications, Lewis (1895) on Guernsey French and

Le Maistre (1947) on Jersey French probably among the earliest, are available including Hublart (1979) and Tomlinson (1981), two PhD-studies on Jèrriais and Guernesiais respectively, which both have a section on Norman French transfer

to Channel Island English, Brasseur’s (1980) linguistic and ethnographic Norman atlas with lexical and phonetic studies, maps and isoglosses, Lukis (1981) on Guernsey French, Spence (1984, 1993, 1999, 2001) mainly on Jersey French, Price (1984) and Liddicoat (1994), especially on Jersey and Sark French The 20th century also saw the compilation of the first dictionaries of Channel Island

French (cf Spence (1960) A glossary of Jersey-French; Carré (1972) English-Jersey language vocabulary; Le Maistre (1976) Dictionnaire jersiais-français; De Garis (1982) Dictiounnaire Angllais-Guernesiais and, most recently, the Dictionnaithe Angliais-Jèrriais) and the first teaching material and textbooks like Birt’s (1985)

Lé Jèrriais pour tous

More recent linguistic research on Channel Island French includes a thesis by Hellmut Lösch (2000), comprising a very detailed overview of Channel Island French and the Channel Islands in general The author meticulously collects and lists all sorts of facts based on his own observations, but principally on findings from earlier studies Another recently completed PhD-project, which resulted in several published articles (cf Sallabank 2002, 2003, 2005), investigates Guernesiais with regard to language planning, language maintenance and issues of identity Sallabank’s studies will be drawn on especially for a discussion of identity and at-titudes and their role in linguistic changes in Jersey (see Chapter 8)

PhD-Mari Jones has been researching insular Norman French for the last ten years approaching it from various perspectives – from the viewpoint of language pres-ervation and planning (e.g Jones 2000, 2009) to descriptions and discussions

of its linguistic features today and in the past (e.g Jones 2001, 2007a) and the unearthing and investigation of previously undocumented material such as the Martin Manuscript (cf Jones 2007b, 2008) Her work also centres upon questions

of language variation and change influenced by the social setting and context of language contact in the Channel Islands (cf Jones 2002, 2005a, 2005b; Jones & Esch 2002; Jones & Singh 2005) From a broader perspective, Jones and Bulot (2009) explore the status of insular Norman within mainland Norman French Among Jones’ numerous publications on the Channel Islands, two studies are of particular importance to the present study, namely her monograph on Jersey Norman French (Jones 2001)22 and a general overview of CIE (Jones 2010) While her 2001 study, based on questionnaires and recorded conversations with Jersey French speakers, focuses on Jersey French, it also contains a section about cross-linguistic influence

on Jersey (see Section 3.1.3 below)

3.1.3 Research on Channel Island English

As mentioned at the start, linguistic research on the English varieties spoken in the Channel Islands has been infrequent Le Pelley (1975) is one of the first to draw at-tention to a distinct English variety in the Channel Islands In his three-page article

“I am Guernsey – Me!”, published in a local journal, he comments on some of the

typical features of CIE as, for instance, the use of the discourse markers eh and is it?, positional to and emphatic pronouns, Norman French borrowings and some

phonetic features He also points to a difference in the English speech of islanders living in the rural parishes and those living in town Interestingly, he predicted the decline of this distinct spoken dialect of English in Guernsey: “Guernsey-English […] is now in great danger of disappearing under the pressure of the telly It is on its way out.” (Le Pelley 1975: 18)

Hublart (1979) and Tomlinson (1981) both briefly mention the influence

of Jèrriais and Guernesiais on the English varieties of Jersey and Guernsey in their respective PhD-theses on Channel Island French Tomlinson, for instance, states that

22 A short version for the general public was published two years later (cf Jones 2003).