Standards and Norms in The English Language Contributions To The Sociology of Language

Standards and Norms in the English Language
≥
Contributions to the Sociology of Language
95
Editor
Joshua A. Fishman
Mouton de Gruyter
Berlin · New York
Standards and Norms
in the English Language
edited by
Miriam A. Locher
Jürg Strässler
Mouton de Gruyter
Berlin · New York
Mouton de Gruyter (formerly Mouton, The Hague)
is a Division of Walter de Gruyter GmbH & Co. KG, Berlin.
앪
앝 Printed on acid-free paper which falls within the guidelines
of the ANSI to ensure permanence and durability.
Library of Congress Cataloging-in-Publication Data
Standards and norms in the English language / edited by Miriam A.

Locher ; Jürg Strässler.
p. cm. ⫺ (Contributions to the sociology of language ; 95)
Includes bibliographical references and indexes.
ISBN 978-3-11-020398-1 (alk. paper)
1. English language ⫺ Standardization. 2. English language ⫺
Variation. 3. English language ⫺ Dialects. 4. English language ⫺
Study and teaching ⫺ Foreign speakers. 5. Sociolinguistics.
I. Locher, Miriam A., 1972⫺ II. Strässler, Jürg.
PE1074.7.S75 2008
428⫺dc22
2008012721
ISBN 978-3-11-020398-1
ISSN 1861-0676
Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche
Nationalbibliografie; detailed bibliographic data are available in the Internet
at http://dnb.d-nb.de.
” Copyright 2008 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin
All rights reserved, including those of translation into foreign languages. No part of this
book may be reproduced in any form or by any means, electronic or mechanical, including
photocopy, recording or any information storage and retrieval system, without permission
in writing from the publisher.
Photo courtesy of Leo Jones and Annemarie Watts
Printed in Germany.
Acknowledgments
The editors would like to express their gratidude to the authors of this col-
lection, who collaborated in an exemplary manner in order to complete this
work in time for Richard Watts’ retirement. In addition, we would like to
thank Anke Beck, Joshua Fishman and the Mouton team, among them Re-
becca Walter, Marcia Schwartz, Monika Wendland and Wolfgang Kon-
witschny for making this publication possible.
Dedication
This collection of papers is inspired by the work of Richard J. Watts,
whose wide research interests are reflected in the topics covered in this
collection. The contributors to this edited book have answered the call to
work on the the theme of ‘norms’ and ‘standards’ in relation to the English
language. Many other colleagues who worked together with Richard over
the years or who appreciate his contribution to the filed of linguistics have
added their names to the tabula gratulatoria. All of us would thus like to
dedicate this work to Richard as a token of our appreciation of his aca-
demic work, which inspired many linguists in the more than three decades
of his publishing activity, his collegiality, his friendship, support, encour-
agement, humour and the many folk song events that he shared with us. We
wish him all the best for his retirement and have no doubt that there will be
many more exciting research findings to be reported from his current re-
search projects that we can look forward to in the upcoming years.
Miriam Locher 1 March 2008 Jürg Strässler

University of Berne University of Berne
University of Zurich
Tabula Gratulatoria
David J. Allerton Jenny Cook-Gumperz

Franz Andres Morrissey Jonathan Culpeper
Anita Auer Silvia Day Tognola
Annalies Aurich Julie Diamond
Anne-Françoise Baer-Boesch Yvonne Dröschel
Francesca Bargiela-Chiappini Mercedes Durham
Jachin Baumgartner Urs Dürmüller
Lorenz Becher Konrad Ehlich
Anke Beck Balz Engler
Heike Behrens Andreas Fischer
Jörg Berger Jacek Fisiak
Raphael Berthele Susan Fitzmaurice
Tony Bex Udo Fries
Roland Bielmeier Dariusz Galasinski
Lukas Bleichenbacher Robert Gassmann
Michel and Claudia Bodmer Kellie Gonçalves
Brook Bolander Scott Gordon
Marc Bonhomme Sage Lambert Graham
Derek Bousfield Fritz Gysin
Margaret Bridges Eric Häberli
Christian Casparis Peter Halter
Jenny Cheshire Jean Hannah
Sarah Chevalier Sandra Harris
Chris Christie Monica Häuptli Gawron
Michael Clyne John Haviland
x Tabula gratulatoria
Monica Heller Eva Maurer

Elke Hentschel Sara Mills
Ernest Hess-Lüttich Bruno Moretti
Martin and Marianne Heusser Louise Mullany
Simon Hicks Heather Murray
Sebastian Hoffmann Silvia Nadig
Marcel and Christine Holliger Walter and Annebeth Naef-
Hinderling
Janet Holmes
Nicole Nyffenegger
Beat Homberger
Adrian Pablé
Juliane House
Gabriela Peikert-Gerster
Hartwig Isernhagen
Henri Petter
Adam Jaworski
Ingrid Piller
Andreas Jucker
Carol G. Preston
Barbara Keller Tanner
Dennis R. Preston
Matt Kimmich
Patricia Pullin
Danièle Klapproth
Ursula Reist Pritzker
Ursula Kluwick
Virginia Richter
Wolfgang Konwitschny
Gabriele Rippl
Felicia Kreiselmaier
Lukas Rosenberger
Silvia Kübler
Michael Rutman
Wilfried Kürschner
Martin Sager
Andreas Langlotz
Srikant Sarangi
Adrian Leemann
Peter Schaerer
Miriam Locher
Iris Schaller-Schwaner
Margaret Mace-Tessler
Annemarie Scheidegger
Didier Maillat
Daniel Schreier
Jennifer Mand
Tabula gratulatoria xi
Marcia Schwartz Beatrice Türler

Philipp Schweighauser Vivian Uebersax-Ridley
Jürg Schwyter Gustav Ungerer
Werner Senn Susi and Walter Vaterlaus
Pieter Seuren Mercedes Viejobueno
Birgit Sievert Li Wei
Kazimierz A. Sroka Katie Wales
Adrian Stähli Rebecca Walter
Brigitte Statzer-Zimmerli Heinrich Weber
Daniel Stotz Urs Weidmann
Felix Strässler Margun Welskopf
Jürg Strässler Monika Wendland
Julia Straub Iwar Werlen
Ingrid Tieken-Boon van Ostade Judith Wieser
Andy Torr Dewi Williams
Gunnel Tottie Siegfried Wyler
Peter Trudgill Jürg Zortea
Cornelia Tschichold
Bibliography of Richard J. Watts’ work
1. Monographs
Watts, Richard J.
1976 Lokative Präpositionen im Deutschen, Englischen und Zürichdeut-
schen: Eine generativ-transformationelle Analyse. Bern: Francke
Verlag.
1981 The Pragmalinguistic Analysis of Narrative Texts: Narrative Co-
operation in Charles Dickens' “Hard Times". Tübingen: Gunter Narr
Verlag.
1991 Power in Family Discourse. Berlin: Mouton de Gruyter.
2003 Politeness. Cambridge: Cambridge University Press.
2. Edited books
Watts, Richard J. and Urs Weidmann (eds.)

1984 Modes of Interpretation: Essays Presented to Ernst Leisi on the
Occasion of his 65th Birthday. Tübingen: Gunter Narr Verlag.
Watts, Richard J. and Franz Andres (eds.)
1990 Zweisprachig durch die Schule/Le bilinguisme à travers l’école.
Berne: Haupt.
Watts, Richard J., Sachiko Ide and Konrad Ehlich (eds.)
1992 Politeness in Language: Studies in its History, Theory and Practice.
Berlin: Mouton de Gruyter.
2005 Politeness in Language: Studies in its History, Theory and Practice
(2nd revised and expanded edition). Berlin: Mouton de Gruyter.
Watts, Richard J. and Jerzy Smolicz (eds.)
1997 Cultural Democracy and Ethnic Pluralism: Multicultural and Multi-
lingual Policies in Education. Frankfurt a M.: Lang.
Watts, Richard J. and Tony Bex (eds.)
1999 Standard English: The Widening Debate. London: Routledge.
Watts, Richard J. and Heather Murray (eds.)
2001 Die fünfte Landessprache? Englisch in der Schweiz. Zürich: vdf.
Hochschulverlag
Watts, Richard J. and Peter Trudgill (eds.)
2002 Alternative Histories of English. London and New York: Routledge.
xiv Bibliography
3. Articles
3.1. Single authored
Watts, Richard J.
1977–78 Grammar unit by unit. In English, of Course!: A Two-year Begin-
ners' Course to Intermediate Level, Volumes 1 & 2. Zürich: Sabe
AG.
1981 Swiss German in English language teaching: A plea for the dialect.
Bulletin CILA 34: 71–80.
1983 The conjunction that: A semantically empty particle? Studia Anglica
Posnaniensia 15 (13–37).
1984a An analysis of epistemic possibility and probability. English Studies
65 (2): 129–140.
1984b Interaktion im Fremdsprachenunterricht: Eine Kritik der funktionel-
len Methode. In Modes of Interpretation. Essays Presented to Ernst
Leisi on the Occasion of His 65th Birthday, R. J. Watts and Urs
Weidman (eds.), 93–99. Tübingen: Gunter Narr.
1984c Laying the formalist ghost: An answer to the charge of scientism.
Studia Anglica Posnaniensia 16: 15–24.
1984d On infinitival complement clauses. Studia Anglica Posnaniensia 16:
45–69.
1984e Spinning a yarn: The linguistic skills of story-telling. In Modes of
Interpretation. Essays Presented to Ernst Leisi on the Occasion of
His 65th Birthday, R. J. Watts and Urs Weidman (eds.), 119–127.
Tübingen: Gunter Narr.
1985a Generated or degenerate? Two forms of linguistic competence. In
Linguistics across Historical and Geographical Boundaries, Dieter
Kastovsky and Aleksander Szwedek (eds.), (Vol. 1), 157–173.
1985b Narration as role–complementary interaction: An ethnomethodologi-
cal approach to the study of literary narratives. Studia Anglica Pos-
naniensia 17: 157–164.
1986a Relevance in conversational moves: A reappraisal of well. Studia
Anglica Posnaniensia 19: 37–59.
1986b Complementation and meaning. Studia Anglica Posnaniensia 18:
101–129.
1986c Sharing a text: A co-operative aspect of verbal interaction. In The
Structure of Texts. Swiss Papers in English Language and Literature
3, Udo Fries (ed.), 37–46. Tübingen: Gunter Narr Verlag.
1987 Relevance in conversational moves: A reappraisal of well. Studia
Bibliography xv
1988a Language as a marker of socio-cultural identity in Switzerland. In

Culture, Philosophy and the Future: Essays in Honour of Sutan
Takdir Alisjahbana on his 80th Birthday, Deliar Noer (ed.), 59–83.
Jakarta: P.T. Dian Rakyat.
1988b Language, dialect and national identity in Switzerland. Multilingua 7
(3): 313–334.
1988c A relevance-theoretic approach to commentary pragmatic markers:
The case of actually, really and basically. Acta Linguistica Hunga-
rica 38 (1–4): 235–260.
1989a Taking the pitcher to the well: Native speakers' perception of their
use of discourse markers in conversation. Journal of Pragmatics 13:
203–237.
1989b Comic strips and theories of communication. Word and Image 5 (2):
173–180.
1989c The perceptual salience of discourse markers in conversation. In
Sprechen mit Partikeln, Harald Weydt (ed.), 601–620. Berlin: Walter
de Gruyter.
1989d Relevance and relational work: Linguistic politeness as politic behav-
ior. Multilingua 8 (2–3): 131–166.
1990a Cross–cultural problems in the perception of literature. In Literary
Pragmatics, Roger Sell (ed.), 26–43. London: Routledge.
1990b The role of early grammar writers in creating a linguistic tradition. In
Sprachtheorie und Theories der Sprachwissenschaft: Geschichte und
Perspektiven. Festschrift für Rudolf Engler zum 60. Geburtstag, Ri-
carda Liver, Iwar Werlen and Peter Wunderli (eds.), 299–315. Tü-
bingen: Gunter Narr.
1991 Linguistic minorities and language conflict in Europe: Learning from
the Swiss experience. In A Language Policy for the EC, Florian
Coulmas (ed.), 75–101. Berlin: Walter de Gruyter.
1992a Acquiring status in conversation: ‘Male’ and ‘female’ discourse
strategies. Journal of Pragmatics 18: 467–503.
1992b Linguistic politeness and politic verbal behaviour: Reconsidering
claims for universality. In Politeness in Language: Studies in its His-
tory, Theory and Practice, Richard J. Watts, Sachiko Ide and Konrad
Ehlich (eds.), 43–69. Berlin: Mouton de Gruyter.
1994a Male vs. female discourse stategies: Tabling conversational topics. In
Texte und Diskurse: Methoden und Forschungsergebnisse der Funk-
tionalen Pragmatik, Gisela Brünner and Gabriele Graefen (eds.),
218–237. Opladen: Westdeutscher Verlag.
1994b Planning in-service training courses: Institutional constraints and
non-native EFL teachers’ perceptions. International Journal of Ap-
plied Linguistics 4: 19–56.
xvi Bibliography
1995a Justifying grammars: A socio-pragmatic foray into the discourse

community of early English grammarians. In Historical Pragmatics,
Andreas Jucker (ed.), 145–186. Amsterdam: Benjamins.
1995b Needs and wants: The semantic and pragmatic interpretation of Eng-
lish privative verbs. Pragmalinguistica 3–4: 173–185.
1996 Schweizerische Identität und der schweizerische Beitrag an Europa:
aus der Sicht eines Ausländers. In Schweizer Eigenart – eigenartige
Schweiz, Wolf Linder, Prisca Lanfranchi and Ewald R. Weibel (eds.),
129–142. Bern: Haupt.
1997a Silence and the acquisition of status in verbal interaction. In Silence:
Interdisciplinary Perspectives, Adam Jaworski (ed.), 88–115. Berlin:
Mouton de Gruyter.
1997b Language policies and education in Switzerland. In Cultural Democ-
racy and Ethnic Pluralism: Multicultural and Multilingual Policies
in Education, Richard J. Watts and Jerzy Smolicz (eds.), 271–302.
Frankfurt a M.: Lang.
1997c Relevance theory and verbal interruptions: Assessing discourse
status. Multilingua 16 (2–3): 153–186.
1999a The ideology of dialect in Switzerland. In Language Ideological
Debates, Jan Blommaert (ed.), 67–103. Berlin: Mouton de Gruyter.
1999b Language and politeness in early eighteenth century Britain. Prag-
matics 9 (1): 5–20.
1999c The social construction of Standard English: Grammar writers as a
‘discourse community’. In Standard English: The Widening Debate,
Richard J. Watts and Tony Bex (eds.), 40–68. London: Routledge.
2000 Mythical strands in the ideology of prescriptivism. In The Develop-
ment of Standard English 1300–1800: Theories, Descriptions, Con-
flicts, Laura Wright (ed.), 29–48. Cambridge: Cambridge University
Press.
2001 Discourse theory and language planning: A critical reading of lan-
guage planning reports in Switzerland. In Sociolinguistics and Social
Theory, Nikolas Coupland, Christopher Candlin and Srikant Sarangi
(eds.), London: Pearson Educational.
2002 From polite language to educated language: The re–emergence of an
ideology. In Alternative Histories of English, Richard J. Watts and
Peter Trudgill (eds.), 155–172. London: Routledge.
2003 Why fuude is not ‘food’ and tschëgge is not ‘check’: A new look at
the actuation problem. In Social Dialectology. Festschrift for Peter
Trudgill, David Britain and Jenny Cheshire (eds.), 115–129. Amster-
dam: Benjamins.
2005 Linguistic politeness research. Quo vadis? In Politeness in Lan-
guage: Studies in its History, Theory and Practice, Richard J. Watts,
Bibliography xvii
Sachiko Ide and Konrad Ehlich (eds.), (2nd revised and expanded
edition), xi–xlvii. Berlin: Mouton.
in press Grammar writers in eighteenth-century Britain: A community of
practice or a discourse community? In Grammars, Grammarians and
Grammar Writing, Tieken-Boon van Ostade, Ingrid (ed.), Berlin:
Mouton de Gruyter.
in press Rudeness, conceptual blending theory and relational work. Journal
of Politeness Research.
3.2. Collaborations
Watts, Richard J. and Beat Homberger

1972 Proposals concerning the teaching of the English progessive aspect.
In Proceedings of the Third International Congress of Applied Lin-
guistics, vol. 2, Copenhagen/Heidelberg: Julius Groos Verlag.
Watts, Richard J. and Siegfried Wyler
1977 Erhebungen zum Englischunterricht an schweizerischen Schulen.
Ergebnisse einer Umfrage. Bulletin CILA 25: 9–23.
Watts, Richard J., Sachiko Ide and Konrad Ehlich
1992 Introduction. In Politeness in Language: Studies in its History, The-
ory and Practice, Richard J. Watts, Sachiko Ide and Konrad Ehlich
(eds.), 1–17. Berlin: Mouton de Gruyter.
Watts, Richard J. and Franz Andres
1993 English as a lingua franca in Switzerland: Myth or reality? Bulletin
CILA 58: 109–127.
Watts, Richard J. and Johanna Ziberi
1994 Cross-cultural miscommunication in a Swiss day-care center. Multi-
lingua 13 (1–2): 13–33.
Locher, Miriam A. and Richard J. Watts
2005 Politeness theory and relational work. Journal of Politeness Re-
search 1 (1): 9–33.
2008 Relational work and impoliteness: Negotiating norms of linguistic
behaviour. In Impoliteness in Language, Derek Bousfield and
Miriam A. Locher (eds.), 77–99. Berlin: Mouton de Gruyter.
Preface
Opening doors, crossing borders
Monica Heller
This book is inspired by the work of Richard Watts (Dick, to people whose
linguistic repertoire protects them from the inappropriate connotations of
the German: as Anke Beck once said to me, “I cannot possibly think of him
as ‘Dick’, he is anything but!”). Characteristically, it is focused on the
ideas, not the man, and yet Watts himself I am sure would acknowledge the
importance of (yes, yes, socially-constrained) life trajectories in shaping
the construction of knowledge. So I will use my privileged position as
preface-writer to at least aim at the balance that is at the core of Watts’
thinking between agency and structure, and which lies in the ability of
people to creatively call linguistic resources into play in the interactional
moment, in ways that are also both structured and structuring.
In many ways the defining features of Watts’ work are contained in his
interest in, and ability to, open doors (to new ideas, to new social actors, to
new spaces) and to cross borders. He himself is of course the consummate
border-crosser, having left his native UK for Switzerland, married a Swiss
psychologist, Anne-Marie, brought up a Swiss son, Chris, and in the ulti-
mate identity move in the Swiss context, appropriated Swiss German as his
own (no one, but no one, who arrives in Switzerland as an adult is sup-
posed to be able to do that). And we won’t even begin to talk about what
he does with Australian English. These border-crossing moves are, I am
convinced, experiences which high-lighted for him the importance of lan-
guage in the construction of social identity and of relations of power.
His initial interest in features of English, which could have resulted in a
career devoted to formal synchronic analysis of language, resulted instead
in a keen awareness that English is not just any old language, it is con-
nected to the wielding of power in specific, and specifically important,
ways – ways which also shift over time as political and social conditions
connected to the use of English also change. The same position flowed out
of his work on the teaching of English (EFL and formal analysis being the
dominant mode of applied linguistics), and to his insistence that teaching
English in particular, as well as other languages, is a deeply political act.
This is of course a matter of public debate worldwide, and a particularly
sharp one in Europe in general and in Switzerland in particular. Watts
xx Preface
found himself in the heat of the action, and decided that both the heat and
the action merited serious scholarly attention. This has not led to an aban-
donment of analysis of the formal features of language; rather, he argues
compellingly that these are part and parcel of the game.
As a result, he inspires a body of work which attends to how speakers
and writers mobilize features of communicative systems in ways which –
whatever else they do – are embedded in the construction of social posi-
tions, that is, of how individuals are connected to always-shifting social
categories, themselves defined in relation to the production and distribu-
tion of the resources (both material and symbolic) which count in our eve-
ryday lives, in the moment and over time. The system features of language
are, in his approach, part of the practice features, and the practice is tied to
how local jockeyings for position are linked to the development and repro-
duction of social institutions over time. While this sociolinguistic episte-
mology is by now fairly widely shared, it is rare to find the kind of careful
empirical work we see here, which works meticulously away at the meth-
odological problem the theoretical position poses, that is, how to describe
those linkages, how to discover how they actually work. This strong em-
piricism also makes Watts’ work compelling across a wide range of epis-
temological positions in linguistics, sociolinguistics and applied linguistics,
synchronic as well as diachronic. This may be a happy by-product of the
continental European structuring of the discipline by language areas rather
than subspecialty.
A less happy by-product, and one that poses a problem for all border
crossers, is the academic structure’s insistence that if you work on one
language you shouldn’t be working on any other languages, which of
course makes work on multilingualism a little difficult. Of course, the
academy mirrors the still-dominant ideology that multilingualism can only
work, indeed can only be allowed to exist, if it is a set of multiple mono-
lingualisms, in which the borders between languages are clear and fixed.
Life in a place like Switzerland, not to mention Watts’ own border-
crossing, provides daily counterexamples, both as regards the kinds of
variation we usually think of as involving different “languages” (English,
French, German, Italian, Romantsch) and the kinds we put into a pile
called “dialect” or something else (notably in this case, involving the so-
called Swiss German dialects, the fading patois of la Romandie, and the
varieties of Romantsch and the problems of its standardization). Watts has
consistently drawn attention to the vital importance of putting these issues
at the centre, not the periphery, of the study of language; that is, of seeing
these not as problems to be cleaned up, but as part and parcel of the ways
Preface xxi
linguistic features are bound up in the construction of social categories,

and hence identities, and of relations of power which those categories help
regulate.
At first glance it may seem that the attention (and tension?) in Watts’
work to the fields of politeness and language standardization / multilin-
gualism involves two somewhat separate domains. But in my view there is
a strong connection between them, a connection which revolves around the
concepts of norms and regulation. In both cases, Watts asks us to attend to
questions about what kinds of values, and what kinds of interests, lie be-
hind our ideas about what it means to use language in ways that are norma-
tively valued, or at least acceptable. He asks us to think about why those
specific forms and practices take the shape they do, given the historical
conditions which position social actors in specific ways with regard to
access to power. He urges us to understand that language practices are all
about shaping (that is, defining and regulating) social relations, not in and
of themselves, but as they become significant with respect to the social
organization of power in specific historical conditions. They are about
struggles between those who benefit from existing linguistic regimes, and
who therefore seek to reproduce them, and those who lose out, and seek
change, or at least some means of resistance.
In this respect, Watts’ work is humanizing. Language is not about some-
thing outside of us, it is part of how we are in the world. It also constitutes
a call for reflexivity, since the way we do linguistics needs to be consistent
with this understanding of what language is. Watts has shown many of us
what it means to do linguistics in the world.
Table of contents
Acknowledgments.................................................................................. v
Dedication .............................................................................................. vii
Tabula Gratulatoria ................................................................................ ix
Bibliography of Richard J. Watts’ work ................................................ xiii
Preface.................................................................................................... xix
Chapter 1
Introduction: Standards and norms
Miriam A. Locher and Jürg Strässler .................................................... 1
Part I. English over time and space
Chapter 2
Swiss English, German English and American English:
In grammatical alliance against traditional British English?
D. J. Allerton.......................................................................................... 23
Chapter 3
Regional variation in English in the new millennium:
Looking to the future
Katie Wales ............................................................................................ 47
Chapter 4
The role of dialect contact in the formation of Englishes
Peter Trudgill......................................................................................... 69
Chapter 5
Non-standardisation
Daniel Schreier ...................................................................................... 85
Chapter 6
From ‘standard’ to ‘nonstandard’ grammar. New England in the days
of Salem Witchcraft and the Civil War
Adrian Pablé .......................................................................................... 105
xxiv Table of contents
Chapter 7
The rise of prescriptive grammars on English in the 18th century
Miriam A. Locher................................................................................... 127
Chapter 8
Lest the situation deteriorates –
A study of lest as trigger of the inflectional subjunctive
Anita Auer .............................................................................................. 149
Chapter 9
The BBC Advisory Committee on Spoken English or
How (not) to construct a ‘standard’ pronunciation
Jürg Schwyter......................................................................................... 175
Chapter 10
Liverpool to Louisiana in one lyrical line:
Style choice in British rock, pop and folk singing
Franz Andres Morrissey ........................................................................ 195
Part II. English usage in non-native contexts
Chapter 11
‘Standard’ English, discourse grammars and English language
teaching
Tony Bex................................................................................................. 221
Chapter 12
Towards a new English as a Foreign Language curriculum for
Continental Europe
Urs Dürmüller........................................................................................ 239
Chapter 13
Language learning and medium of instruction:
Understanding contemporary discourses and practices in Swiss
schools and classrooms
Daniel Stotz ............................................................................................ 255
Table of contents xxv
Chapter 14
Can academic writing style be taught?
Jürg Strässler ......................................................................................... 281
Chapter 15
Linguascaping Switzerland: Language ideologies in tourism
Adam Jaworski and Ingrid Piller........................................................... 301
Chapter 16
The rules of “Denglish”
Elke Hentschel........................................................................................ 323
Part III. Issues on politeness and impoliteness
Chapter 17
(Im)politeness in English as Lingua Franca discourse
Juliane House......................................................................................... 351
Chapter 18
How to be impolite: Rating offensive strategies
Mercedes Viejobueno, Carol G. Preston and Dennis R. Preston.......... 367
Contributors ........................................................................................... 393

Contact information ............................................................................... 399
Subject index.......................................................................................... 403
Author index .......................................................................................... 407
Chapter 1
Introduction: Standards and norms
Miriam A. Locher and Jürg Strässler
1. Norms and standards: The theme of the collection
This collection of papers is inspired by the work of Richard Watts, whose

wide range of research interests is reflected in the topics covered here. An
important theme in his work can be summarised as an academic concern
for norms and standards in the English language. As these two terms may
evoke different connotations and associations, we should first establish
how the terms are to be understood in this collection.
We will start our discussion by looking at the entry for ‘standard’ in the
Oxford English Dictionary. It contains many different subheadings; the one
most closely connected to the theme of this collection reads as follows:
B. 3.e. Applied to that variety of a spoken or written language of a country
or other linguistic area which is generally considered the most correct and
acceptable form, as Standard English, American, etc. (Oxford English Dic-
tionary 1989, online)
This definition is closely linked to the following:
A. 10. a. An authoritative or recognized exemplar of correctness, perfection,
or some definite degree of any quality.
b. A rule, principle, or means of judgement or estimation; a criterion, meas-
ure. (Oxford English Dictionary 1989, online)
These definitions are said to be figurative derivations of the definition
given under the entry A. 9.a:
A. 9. a. The authorized exemplar of a unit of measure or weight; e.g. a meas-
uring rod of unit length; a vessel of unit capacity, or a mass of metal of unit
weight, preserved in the custody of public officers as a permanent evidence
of the legally prescribed magnitude of the unit.
(Oxford English Dictionary 1989, online)
With respect to this definition, it is obvious that a standard measure must
be authorized by some institution and that there is no room for any varia-
tion. Irrespective of what system of measurement a country declares as its
2 Miriam A. Locher and Jürg Strässler
standard, a metre is always defined as the length of the path travelled by

light in vacuum during a time interval of 1/299,792,458 of a second and the
British imperial yard was defined in 1878 as the distance at a specified
temperature between two lines engraved on gold studs sunk in a certain
bronze bar. Milroy and Milroy (1985: 23) used the coinage as a similar
example for standardisation in matters outside language, stating that “the
coinage is strictly standardised so that there can be no variation in the val-
ues assigned to the counters in the system ... to ensure reliability and hence
confidence.”
In the figurative derivations given above, the authoritative component is
already weakened and no longer a necessary prerequisite for a standard,
and in definition B.3.e. there is no longer any mention of authority. Stan-
dard is defined as the variety “which is generally considered the most cor-
rect and acceptable form” (emphasis added). This definition, however,
implies that beside the standard form its opposite must exist as well, which
opens again the door to a discussion of variation.
The entry for ‘norm’ in the Oxford English Dictionary is closely linked
to the notion of ‘standard’:
1. a. That which is a model or a pattern; a type, a standard.
b. A standard or pattern of social behaviour that is accepted in or ex-
pected of a group. Usu. in pl.
c. A value used as a reference standard for purposes of comparison.
(Oxford English Dictionary 1989, online)
While the definitions of ‘norm’ in the subentries 1.a. and 1.b. treat the term
as synonymous to the concept of ‘standard’ described in meaning A.10.
above, meaning 1.c. is arguably more restricted and matches the descrip-
tion for ‘standard’ given in 9.a. above. In this reading, the actual bench-
mark is highlighted: It refers to precise, reliable values declared by an au-
thority for special purposes, mainly in technology and the natural sciences.
Hierarchically, the concept of norm appears to be entailed in the concept
‘standard’. Practically, however, many laypeople as well as linguists use
the two terms interchangeably.
While the concepts of ‘standard’ or ‘norm’ have a high currency in lin-
guistic research in general, the way they have been treated in the different
fields of linguistics is rather diverse. Most generally, we can say that what
a standard or a norm constitutes in a specific context has to be defined as a
point of reference. As the topics discussed in this collection cover a wide
field in which the notions are of relevance, we also see different conceptu-
alisations at work. The contributing researchers take up the challenge in
Introduction: Standards and norms 3
different ways and discuss original data and new research questions con-
nected by their focus on standards and norms.
This collection is organized into three parts, each of which covers an
important research field for the study of norms and standards: (1) English
over time and space, (2) English usage in non-native contexts, and (3) is-
sues on politeness and impoliteness. While these areas of linguistic inves-
tigation are by no means comprehensive with respect to the study of norms
and standards, the choice was determined by the fact that Richard Watts
has contributed significantly to each of these three fields of enquiry.
2. English over time and space
The first part of this collection covers topics on the English language over
time and space with the notions of norms and standards in mind. The nine
chapters loosely form three thematic subgroups: (1) standard and non-
standard features in English varieties and dialects, (2) research on English
standardisation processes and (3) issues of standards and norms in oral
production. Before we summarise the content of the chapters, we will brie-
fly introduce these fields of study.
Nowadays, according to Trudgill (1998: 38), there is a clear consensus
among sociolinguists that Standard English is a dialect, i.e. “one variety of
English among many. It is a sub-variety of English.” This, however, has
not always been the case. In the first edition of the Oxford English Dic-
tionary (1933), the term ‘dialect’ is defined as
one of the subordinate forms or varieties of a language arising from local
peculiarities of vocabulary, pronunciation and idiom ... A variety of speech
differing from the standard or ‘literary’ language.
This definition implies that the standard language is a hyperform of all the
respective dialects. Furthermore it concentrates solely on regional varia-
tions in vocabulary, idiom usage and pronunciation, thus ignoring grammar
as well as social differences.
This is contrary to the modern understanding that a “standardised lan-
guage is a language one of whose varieties has undergone standardisation
... consisting of the process of language determination, codification and
stabilisation” (Trudgill 1998: 35). Furthermore, Standard English is inde-
pendent of pronunciation, register and style, which may have been stan-
dardised independently. Standard English is the variety commonly used by
the media, especially in printing, and employed in the educational systems

for both native speakers and learners of English.
As the standard is often associated with the language of educated peo-
ple, social variations are more important than regional ones. Style is nor-
mally linked to formality, ranging from vulgar to very formal, and it de-
pends mostly on lexical choice. It is obvious that speakers may choose
their words independently of the other features of language in such a way
that they are appropriate for the social situation. The same is true for regis-
ter, the vocabulary of which is determined by the subject matter or the ac-
tivity a speaker is engaged in. In spoken language, we can thus encounter
instances of Standard English in a very careful style within a special regis-
ter, but pronounced with a very strong regional or non-native accent; while
it is also possible to encounter instances of non-standard, casual English
but delivered in the prestigious RP accent (cf. Trudgill 1998). Thus, if
Standard English is not connected to a specific form of pronunciation, reg-
ister and style, it must be mainly connected to grammar, i.e. to syntax and
morphology. In addition, “Standard English is not a set of prescriptive
rules”, as Trudgill (1998: 38) points out.
The understanding of norms and standards by grammarians is closely
linked to the concepts of prescriptivism versus descriptivism. Whereas
there was an enormous increase in prescriptive grammars in the second
half of the 18th century (Locher, Chapter 9), there was strong opposition to
prescriptivism already in the Victorian age and especially in the first half
of the 20th century by Bloomfield (1933) and the American structuralists.
Modern reference grammars such as The Cambridge Grammar of the Eng-
lish Language (Huddleston and Pullum 2002), The Cambridge Grammar of
English (Carter and McCarthy 2006) and The Collins COBUILD English
Grammar (Sinclair 1990) and even EFL grammars (for instance the Oxford
Practice Grammar [Eastwood 2006] and Advanced Grammar in Use
[Hewings 2005]) are in fact all descriptive. All these grammars are corpus
based and thus reflect the present state of (mainly written1) English.
Huddleston and Pullum (2002) do not mention the term ‘Standard Eng-
lish’ at all and take a very clear stance against prescriptivism when they
state:
Our aim is to describe and not prescribe: we outline the principles that gov-
ern the construction of words and sentences in the present-day language
without recommending or commending particular choices. ... We report that
sentences of some types are now widely found and used, but we will not ad-
vise to use them. We state that sentences of some types are seldom encoun-
tered ... but we will not tell you that you should avoid them or otherwise
make recommendations about how you should speak or write. Rather, this
book offers a description of the context common to all such decisions: the
linguistic system itself. (Huddleston and Pullum 2002: 2)
Carter and McCarthy (2006: 5), on the other hand, claim that their book
is a grammar of standard British English but state that “issues of accept-
ability are never far from the surface when there is reference to what is
standard in grammar or in language use in general.” They adopt a five-
scale system of acceptability ranging from “acceptable in standard and
spoken English (most forms are in this category)” to “unacceptable in all
varieties of English (for example a structure such as he did must speak),
such forms are excluded from this book.” Most interesting is level 2 con-
sisting of forms “acceptable in standard written and spoken English but not
approved in more prescriptive grammar books.” Their grammar is thus
mainly based on acceptability and not on traditional, often Latin-based
prescriptivism.
Much to the dismay of some traditional teachers of English as a foreign
language, who would like to have a clear distinction between ‘correct’ and
‘incorrect’, modern EFL books have also abandoned prescriptivism to a
large extent. Eastwood (2006: viii), whose grammar is based on the Cam-
bridge International Corpus (www.cambridge.org/corpus) states that “[t]he
emphasis throughout the book is on the meaning and use of the grammati-
cal forms. The explanations of grammar are descriptions of how English
works; they are a guide to help you understand, not rules to be memo-
rized.”
If standards in language are based on acceptability and not on any au-
thoritative decree, how do such standards emerge? Unlike in other coun-
tries (e.g. France, Iceland and to some degree Germany), there has never
been an authoritative institution that was licensed by the government to
impose rules on English language use. As the first step of standardisation,
as mentioned above, is language determination, i.e. the “decisions which
have to be taken concerning the selection of particular languages or varie-
ties of language for particular purposes in the society or nation in question”
(Trudgill 1998: 35), the question is who takes these decisions in the ab-
sence of a legitimised authority? And after these decisions have been
taken, how are they codified and made accessible to a wider public?
According to Milroy and Milroy (1985), “[t]he attitudes of linguists
(professional scholars of language) have little or no effect on the general
public, who continue to look at dictionaries, grammars and handbooks as
authorities on ‘correct’ usage” (Milroy and Milroy 1985: 6). As the public
considers people who can write such books as highly educated and prestig-
ious, they thus indirectly accept them as authorities.
With respect to the question of assigning prestige and authority to a va-
riety, Quirk et al. (1972) state the following:
Educated speech – by definition the language of education – naturally tends
to be given the additional prestige of government agencies, the learned pro-
fession, the political parties, the press, the law court and the pulpit – any in-
stitution which must attempt to address itself to a public beyond the smallest
dialectal community. ... By reason of the fact that educated English is thus
accorded implicit social and political sanctions, it comes to be referred to as
Standard English. (Quirk et al. 1972: 16)
It must be noted, however, that although the decisions taken were often
arbitrary, stigmatizing certain forms and favouring others, they have be-
come standard forms by virtue of being codified and accepted by the ma-
jority of educated people.
There have always been, and probably will always be calls for an Eng-
lish language academy like the Académie française (cf. Swift 1712), but
with no success so far (cf. Milroy and Milroy 1985). Standardisation has
always been instigated by individuals or institutions, whose work was then
interpreted as authoritative at a later stage. The most successful of these
were probably Caxton (1490) and Johnson (1755). Caxton complained
about the varieties and the constant language change and expressed the
need of a standard written form for printing purposes. The English variety
he adopted for printing is seen as contributing to a standardisation process
and has been influential ever since. Johnson’s Dictionary of the English
Language, which was prescriptive in its realisation, on the other hand, has
influenced English orthography so strongly (cf. Milroy and Milroy 1985:
34–35) that even today there is an indisputable consensus in matters of
spelling throughout the English speaking world with basically “two minor
subsystems” reflecting US or GB conventions (cf. Quirk et al. 1972: 16–
17).
‘Standard’ with respect to oral production is closely related to Daniel
Jones, who published the first pronunciation dictionary in 1917. Although
he clearly stated that “[n]o attempt is made to decide how people ought to
pronounce; all that the dictionary aims at doing is to give a faithful record
of the manner in which certain specified classes of people do pronounce”
(Jones 1924: vii, emphasis in original), we read in the editors’ preface to
the 16th edition that “it has become established as a classic work of refer-
ence, both for native speakers of English wanting an authoritative guide to
pronunciation and for users of English as a foreign or second language”

(Roach, Hartman and Setter 2003: iv). With “certain specified classes of
people” Jones meant “Southern English persons who have been educated at
the great public boarding-schools” (Jones 1924: vii-viii), thus opting for a
mainly social model, which he called PSP (Public School Pronunciation).
In his function as a permanent specialist member of The BBC Advisory
Committee on spoken English (cf. Schwyter, Chapter 9), Jones took a
rather prescriptive position. Although this board only lasted from 1926–
1939 and was never accepted as an authority, it “has made us think about
notions such as ‘standard’, RP, ‘correctness’ more than ever before”
(Schwyter, this volume: 188). The influence of the BBC Advisory Board
on broadcasting and pronunciation may be parallelled with the influence of
Caxton and Johnson on printing and spelling.
In the following, the nine chapters in part 1 on “English over time and
space” are discussed in their three subgroups.
2.1. Variation in English varieties and dialects
The chapters contained in this section deal with phonological, morphologi-

cal, syntactical or lexical questions of standard and non-standard features
of the English language. While the first chapter by David Allerton dis-
cusses influences on Standard English as such, the contributions by Wales,
Trudgill, Schreier and Pablé are concerned with the importance of (non-
standard) dialect features. Such microlinguistic topics have been at the
heart of much of Richard Watts’ teaching over the past 30 years, both in a
synchronic and in a diachronic approach. In addition, Watts was involved
in a research project on Swiss English, which tackled the question of an
emerging Swiss variety of English on the micro- and the macrolinguistic
levels (cf. Watts and Murray 2001).
David Allerton starts the first block by studying “Swiss English, Ger-
man English and American English: In grammatical alliance against tradi-
tional British English?” (Chapter 2). Allerton is interested in finding influ-
ences on standard British English that can be explained as being caused by
either American English or by (Swiss-)German English influences. In par-
ticular, he studies grammatical patterns such as the use of the perfect, the
replacement of auxiliary have with be before a past participle, unorthodox
agreement in copular sentences, the lack of tense agreement in cleft sen-
tences, “infinitivitis”, and other features in order to discuss influences from
other varieties of English on British English. Allerton argues that “in some
cases at least, grammatical Immigrant Americanisms, and in some cases

even some German-Americanisms, are a plausible hypothesis” (24) for
innovations in British English.
In the contribution entitled “Regional variation in English in the new
millennium: Looking to the future” (Chapter 3), Katie Wales turns our
attention from influences on standard English more globally to the local
(micro-)level of variation, which is “summarily labelled ‘non-standard’,
and traditionally stigmatised in comparison with standard grammar (and so
often wrongly equated with ‘substandard’)” (61). Wales first gives a his-
torical overview of linguists and language commentators who predomi-
nantly predicted the decline of regional variation in a general levelling and
standardisation process and then continues to debate “the extent of dialect
‘death’ and the prognosis for degree of robustness in the future” (48). By
examining the metaphors used to describe regional variation, the traditions
in dialectology for representing dialect boundaries geographically and
graphically and by investigating the urban variation in Yorkshire, Wales
comes to the conclusion that regional diversity in speech is not in imminent
danger of disappearing. She summarises that “[r]egional ‘norms’ do not
necessarily mean homogenisation or fixity, any more than ‘globalisation’
does” and calls for more research on local patterns and tendencies in order
to capture the continuous “process of ‘hybridisation’” (62).
Peter Trudgill, in turn, develops the ideas of the micro and macro level
with respect to the question of “The role of dialect contact in the formation
of Englishes” (Chapter 4). He claims that, while language contact is often
discussed more generally as having (had) an influence on the development
of the English language, one should not neglect the importance of dialect
contact. The reason for this, he argues, is that “dialect contact has more
often than not led to dialect mixture and to what I have called new-dialect
formation (Trudgill 1986, 2004)” (70). To exemplify this, Trudgill dis-
cusses three examples in more detail: Old English as a colonial mixed dia-
lect, Middle English in Ireland, and the case of North America.
Daniel Schreier focuses on “non-standardisation” (Chapter 5) and dis-
cusses this topic in relation to St Helenian English (StHe). He stresses that
“most off-spring varieties of British English have in fact not standardised
at all” (85) and that speaking a non-standard variety in the English-
speaking world is rather the norm than the exception. He then discusses the
development of St Helenian English by tracing the importance of the Eng-
lish donor varieties, which all happened to be non-standard, and by offer-
ing suggestions for “disentangl[ing] the sociolinguistic complexity of a
variety’s initial stages and identify[ing] the principal donors that shape a
‘new’ non-standard variety” (86). Schreier covers data from the period
between the 1660s and the late eighteenth century and summarises that the
“StHE feature pool was quite diverse, containing donor varieties from Eng-
land, continental Europe, Africa and Portuguese colonies in the Atlantic
Ocean, as well as the Indian subcontinent and Asia” (99). Since the English
input to St Helenian English was non-standard (identified as working-class
southeastern British English), this emerging variety is found to have been
“non-standard to start out with” (100). Like Trudgill, Schreier concludes
that dialect contact “contributes heavily to the continuing spread of non-
standard varieties throughout the English-speaking world” (100).
Adrian Pablé discusses the topic “From ‘standard’ to ‘nonstandard’
grammar. New England in the days of Salem Witchcraft and the Civil
War” (Chapter 6). He is interested in establishing what grammatical fea-
tures that are considered to be non-standard in present-day Standard
American English, were acceptable features in the seventeenth and the
nineteenth centuries. His sources are the Salem Witchcraft Papers and
Civil War correspondence. The features under scrutiny are, among others,
finite indicative be, nonstandard was, nonstandard verbal –s, unmarked
present and past tense verbs, etc. His conclusion is that the “spoken gram-
mar of New England English was fundamentally the same in the seven-
teenth and the nineteenth centuries”, but that “[w]hat did change in the
course of the late eighteenth/early nineteenth century was the social
evaluation – from positive or neutral to negative – underlying these gram-
matical forms and structures” (121).
2.2. Research on English standardisation processes
The second group of chapters in the section on “English over time and
space” contains work with historical data of the English language on ques-
tions of standardisation and the success of standardisation processes (cf.,
e.g. Watts and Bex 1999; Watts 2000). In the chapter “The rise of prescrip-
tive grammars on English in the 18th century” Miriam Locher investigates
the dramatic increase in the number of publications on English grammar in
the second half of the 18th century in England. These texts are discussed in
connection with the process of language standardisation since they propa-
gate a normative, prescriptive view of language. Several interrelated fac-
tors are suggested to account for this increase of publications on English,
one of them being the importance of the notion of politeness for social
climbers, who believed that they could better their situation by learning to
use ‘correct’ language. In addition, one text by the grammarian Fell is in-
troduced in more detail to demonstrate that many of the issues linguists
deal with today, such as prescriptivism versus descriptivism, were already
discussed in the 18th century.
Anita Auer presents research on the use of lest from Early Modern Eng-
lish to the twentieth century in her chapter entitled “Lest the situation dete-
riorates – A study of lest as trigger of the inflectional subjunctive” (Chap-
ter 10). She discusses the inflectional subjunctive and in particular the
conjunction lest as a trigger of the subjunctive by means of diachronic and
synchronic corpus-based research on the actual use of lest, as well as by
investigating comments on lest by prescriptive and descriptive grammari-
ans over time. The results show that
[lest + subjunctive] was still used in Early Modern English, then disap-
peared for 250 years, and it has experienced an enormous revival between
1985–1994 (the end date of the study). The analysis of meta-linguistic
comments by grammarians and language-guardians exhibited that only
eighteenth-century grammarians were particularly concerned with emphasis-
ing that lest necessarily required the inflectional subjunctive. (Auer, this vol-
ume: 165)
Auer argues that the prescriptive grammarians were most concerned
with correctness and forming a standard usage of lest, while the grammari-
ans of the Early Modern English period before still tried to come to terms
with the concept of mood, and the grammarians after the eighteenth cen-
tury “were concerned with distancing themselves from claims made by
eighteenth-century prescriptivists” (166, emphasis in original).
2.3. Issues of standards in oral production
While the chapters in the first and second section of the part on “English
over time and space” focused not only on phonological features but also on
morphological, syntactic and lexical issues, the two chapters introduced in
this sub-section are centred exclusively on the study of standards in oral
production.
Jürg Schwyter presents work on “The BBC advisory committee on spo-
ken English or How (not) to construct a ‘standard’ pronunciation” (Chapter
9). Schwyter focuses on the period from 1926 to 1939 and discusses the
development and success that the BBC advisory committee had in stan-
dardizing pronunciation. He concludes that the advisory committee moved
through three phases, from a “strict and dogmatic prescriptivism” (186) to

a “listening BBC” (187), i.e. the BBC was open to feedback from the gen-
eral public and an enlarged body of experts, and finally towards “linguistic
professionalization” (187). Schwyter maintains that the result of the com-
mittee’s endeavours was not the standardisation of a fixed pronunciation,
but the “emergence of a kind of ‘broadcast English’ or ‘broadcast style’”
and the general “raised awareness of language issues among the popula-
tion” (187).
Franz Andres Morrissey explores questions of phonological style in re-
cordings of popular music in his contribution entitled “Liverpool to Louisi-
ana in one lyrical line: Style choice in British rock, pop and folk singing”
(Chapter 10). In particular, he pursues the notion of an American reference
style in music, a ‘standard’ that was adopted by many British interpreters.
He traces the development of and deviations from such a style over the last
decades and argues that there are several factors which influence style
choice in rock, pop and folk singing next to socio-linguistic considerations,
such as “musical genre, song topics and cultural considerations” (196). In
addition, the aspect of performance and its impact on pronunciation, in
particular the crucial element of sonority, is highlighted as a potential ex-
planation for inconsistencies in style choice.
3. English usage in non-native contexts
The second part of this collection contains six chapters and is concerned
with the study of English used in non-native contexts. The contributions by
Bex, Dürmüller, Stotz and Strässler discuss aspects of the question of Eng-
lish usage in the classroom. The authors raise issues such as which stan-
dards of English should be taught in school and how standard and non-
standard forms of English can be used, exploited and learnt in the class-
room. These issues are discussed against the backdrop of the ongoing de-
bate on teaching English as a Foreign Language (EFL) and the use of Eng-
lish as a lingua franca (ELF), i.e. the continuing internationalisation of
English in a global context (cf., e.g. Gnutzmann and Intemann 2005; Grad-
dol 2006; Jenkins 2003; Watts 1990; Watts and Murray 2001).
In the first of these contributions, Tony Bex discusses “‘Standard’ Eng-
lish, discourse grammars and English language teaching” (Chapter 11). He
presents a lucid discussion of the different points of view on what ‘stan-
dards’ should be used when teaching English to students in countries
where English is not spoken and in countries where English is a native
language. In addition, he argues that corpora work, such as the CANCODE

project, which, among other issues, allows linguists to study discourse
patterns, are a great resource to study native speaker varieties of English,
but that the results of these analyses should not directly flow into teaching
of English as a foreign language. Bex thus takes up Phillipson’s (1992)
warning against transporting cultural beliefs and values together with lan-
guage teaching and concludes that “the native-speaker English teacher has
become an expensive irrelevance for those countries which teach English
as a second language, and that such teaching should be restricted to coun-
tries which have English as a first language” (221).
Urs Dürmüller continues this theme with his contribution “Towards a
new English as a Foreign Language curriculum for Continental Europe”
(Chapter 13). Like Bex, Dürmüller discusses the problem of what ‘stan-
dard’ of English to choose in foreign language teaching. He claims that,
due to the increase of the importance of English on an international level,
the school curricula in countries where English is not a first language
should be redesigned. By looking at one particular example of a teacher
training college in Switzerland, Dürmüller identifies the issues and prob-
lems that arise in such a reformulation of the targets for teaching and he
discusses the suggestions proposed, which can result in “a new orientation
for EFL teacher training” (239). Dürmüller concludes that “English must
be perceived as a truly international language, increasingly expressive of
many and differing cultures, literatures and lifestyles, …” (251).
Daniel Stotz also focuses on a language situation in which English is a
foreign language. In his contribution entitled “Immersion or integration?
Discourses and practices in Swiss language classrooms” (Chapter 12), he
traces the development of reforms in foreign language schooling in Swit-
zerland. Stotz questions “the notion of standard, in particular with respect
to the diglossic situation in Swiss schools, and points to the additional so-
ciolinguistic complexities that the standard/dialect dichotomy introduces”
(256). His argument is that
while immersion-style provision of language classes may look like a persua-
sive solution to the school system’s task of promoting students’ multilingual
competence, it comes with the inherent contradiction that it attempts to in-
troduce, as a medium of instruction, standard forms of language in an envi-
ronment characterised by various forms of non-standard use such as dialect,
learner language and hybrid mixing of different varieties. (Stotz, this vol-
ume: 256)
Stotz exemplifies these issues with an analysis of classroom interaction and

a discussion of a recorded information forum for adolescents and parents in
a suburb in Zurich. In addition, special attention is given to the particular
role assigned to English in Switzerland, which is not a national language,
but which is perceived by many as an indispensable language for students
to learn.
In his contribution “Can academic writing style be taught?” (Chapter
14), Jürg Strässler analyses and compares papers written by students of two
Swiss universities at the beginning and the end of specialised writing skills
courses. Using the QSUM technique (Farringdon 1996), which is one of
the most widely recognised methods for attributing texts to specific authors
in the framework of forensic linguistics, he shows that the standards for
academic writing taught in the courses are so rigid that they leave hardly
any room for individual features. Whereas the texts written at the begin-
ning display a high degree of integrity and homogeneity, those written at
the end of the courses lack these qualities. As this development runs
counter to Farringdon’s claim that each person’s QSUM ‘fingerprint’ re-
tains consistency across different genres, Strässler juxtaposes personal and
academic texts written by two of the lecturers. As the same picture emerges
from the respective analyses, we have to assume that academic writing is
strongly influenced by a genre-specific standard, and that native and non-
native academics experience similar difficulties (cf. Swales 2004; Devitt
1997). Strässler suggests that English has not only become the Lingua
Franca in the globalised world of academia, but that a genre-specific writ-
ing style has developed to which native as well as non-native authors seem
to adhere.
The remaining two chapters in the part on English usage in a non-native
context focus no longer on questions of English in the classroom, but on
two very different language situations. The first chapter by Adam Jaworski
and Ingrid Piller deals with language employed in the tourism sector, while
Elke Hentschel discusses the incorporation of English loanwords into the
German grammar system. In their chapter “English in the Swiss tourism
sector” (Chapter 15), Jaworski and Piller investigate language ideologies,
and study “the linguistic knowledge about Switzerland that is produced and
reproduced” in British newspaper travelogues (315). They found that Ro-
mansch, as the smallest national language, receives disproportionate atten-
tion, and that German, French and Italian are exoticised to a lesser degree
and mainly used to refer to “local terms for local cuisine, landmarks or
cultural events” (316). The travelogues do not report on any of the many
immigrant languages that actual travellers are likely to be exposed to in the
tourism sector, such as Croatian, Portuguese, Serbian, or Spanish. The

authors interpret this fact as a process of “standardisation” in the sense that
“the travelogues do not in any way deviate from the official version of the
Swiss linguistic landscape” (316). In addition, the authors argue that “Eng-
lish is naturalized as the language of tourism to Switzerland”, as it is else-
where (316). Piller and Jaworski conclude their contribution by stating that
“tourism as a practice carries with it its own form of linguistic standardiza-
tion” (316).
The final chapter in this part is by Elke Hentschel, who writes on “The
rules of ‘Denglish’” (Chapter 16). She studies the use of Anglicism in
German Internet texts that are collected from the area of every day com-
puter usage. She is interested in finding out how the users deal with the
orthographical and morphological challenges of incorporating English
loanwords into the German language system. Hentschel argues that “mor-
phology and orthography, taken together, offer the best possible view on
the development of new standards” (325). The term standard here refers to
“naturally evolving rules” rather than to prescriptive norms (325). Overall,
Hentschel demonstrates that, surprisingly, the verbs investigated showed a
decrease in integration into the German orthographic and morphological
system despite the fact that that the overall usage of these verbs increased
in the time period studied (2003–2006). Hentschel concludes that
“[s]tandardization … can neither be foreseen nor does it follow the rule
that increasing frequency leads to higher integration” and that “additional,
so far unknown factors play an important role in the development of new
standards for foreign and loan words” (344).
4. Issues on politeness and impoliteness
Politeness research is another important field of linguistic enquiry for

which standards and norms are a central concern. Richard Watts has been
working on this topic since the late Eighties (cf. 1989, 1992, 1999, 2003;
Locher and Watts 2005) and has contributed considerably to a re-
evaluation of the seminal work by Brown and Levinson ([1978] 1987). In
recent years, Watts has also turned his attention to the opposite of harmo-
nious relational work and has started to look at impoliteness (cf. Watts in
press; Locher and Watts 2008). Watts stresses that the study of politeness
(and impoliteness) must entail the study of norms and standards against
which judgements of politeness are being made by social actors of a par-
ticular discursive practice. He introduced the concept of ‘politic behaviour’

to the terminology of politeness research, which he defined as
[s]ocio-culturally determined behaviour directed towards the goal of estab-
lishing and/or maintaining in a state of equilibrium the personal rela-
tionships between the individuals of a social group, whether open or closed,
during the ongoing process of interaction. (Watts 1992: 50)
The concept of ‘politic behaviour’ highlights the idea of appropriate-
ness, which is linked to judgements based on the norms of a particular in-
teraction. Politeness, Watts argues, is a positively marked version of ap-
propriate behaviour.2 In this field of research standards and norms are thus
seen as sets of expectations that serve as a framework for judgements on
relational work.
Politeness and impoliteness are, in this view, evaluative notions that
cannot be regarded without considering their context. This emphasis on in
situ judgements on interaction by participants has added a new angle to
politeness research. The focus of study is on first order, layperson’s inter-
pretations. The terms politeness and impoliteness are thus not used as tech-
nical terms for linguistic analysis, but are discussed in the light of the par-
ticipants’ understanding of what they might have constructed as polite or
impolite behaviour in interaction. The stress is thus on the discursive na-
ture of the evaluative terms, in the sense that the standards and norms
against which such judgements are made are constantly being negotiated,
confirmed or subtly changed over time. This negotiation of norms and
standards adds a historical aspect to the study of politeness phenomena
(cf., e.g. Watts 1999).
As a consequence of freeing the term politeness from its technical
Brown and Levinsonian meaning, i.e. taking it again as first order notion, a
slight shift in interest has occurred with respect to the study of interper-
sonal communication. While much of the classic politeness research con-
centrated on mitigation strategies, in recent years researchers have increas-
ingly turned to the study of other aspects of interpersonal behaviour, such
as impoliteness (see, e.g. Bousfield and Locher 2008; Bousfield 2007;
Gorji 2007). It remains to be stressed that any investigations of the rela-
tional aspect of language focusing on first order notions will have to dis-
cuss the norms and standards evoked by the participants in an interaction.
In the third part of the collection the issues of politeness and impolite-
ness already touched upon in Chapter 9 on the prescriptive grammars
(Locher) are taken up in two papers. The issues of norms and standards are
at the heart of the contributions by Juliane House and the team of scholars
Mercedes Viejobueno, Carol G. Preston and Dennis R. Preston. In her con-

tribution entitled “The role of politeness in discourse in English as a lingua
franca” (Chapter 17), Juliane House studies whether the use of English as a
lingua franca in interactions between members of different linguistic and
cultural backgrounds can be called ‘impolite’, judged with native speaker
norms in mind. She draws on data from three case studies that discuss in-
teractions by non-native speakers who use English as their lingua franca.
The results of the analyses show that the ELF speakers create their own
habitus, or politic behaviour, which is particular to the interaction in ques-
tion, and which does not seem to be perceived as inappropriate or impolite
by the interactants in question.
In their chapter “How to be impolite: Rating offensive strategies”
(Chapter 18), Mercedes Viejobueno, Carol Preston and Dennis Preston
investigate impoliteness and rudeness rather than politeness. They thus turn
to a field of research in which surprisingly little empirical studies have
been done, given the enormous interest in politeness. The authors point out
that impoliteness is so common that “a descriptive framework is necessary,
but those that have been proposed are incomplete” (368). They then set out
to offer a cross-cultural study in which they first established lay-people’s
evaluations of a set of interactions (such as annoying, rude, impolite,
mocking, mean, etc.) and then, in a second study, they investigate how
these terms are ranked by participants with respect to degree of offence.
The results show that “the perception of direct and sarcastic offenses
strongly depends on the face that is being attacked, the social distance be-
tween the interlocutors, or the interaction between these two factors”
(382).
5. Concluding remarks
The theme of this collection is a discussion of the notions of ‘norms’ and

‘standard’, which are studied from a variety of angles, but always in rela-
tion to the English language. As mentioned above, these terms are to be
understood in a very wide sense, allowing discussions of topics such as the
norms we orient to in social interaction, the benchmark employed in teach-
ing, or the development of English dialects and varieties over time and
space and their relation to the standard language. The notions of standards
and norms are equally important concepts for historical linguists, sociolin-
guists with a variationist background, applied linguists, pragmaticians and
discourse analysts. We trust that this collection offers a unique view on
how the concepts of norm and standard are of importance in the different
fields of linguistics touched on here.
Notes
1. However, it should be stressed that the majority of the corpora which these
Grammars are based on predominantly represent written English. Since the
written mode adheres more generally to the notion of standard English, much
of the variation observed in the vernacular is lost (both with respect to lexical
as well as grammatical features). Modern pedagogical grammars such as Hew-
ings (2005) are innovative in that they also include sections on oral usage.
2. For thorough introductions to politeness research, cf. Watts et al. (1992/2005),
Eelen (2001), Watts (2003) and Locher (2004).
References
Bloomfield, Leonard
1933 Language. New York: Holt, Rinehart & Winston.
Bousfield, Derek
2007 Impoliteness in Interaction. Amsterdam: John Benjamins.
Bousfield, Derek and Miriam A. Locher (eds.)
2008 Impoliteness in Language. Studies on its Interplay with Power in
Theory and Practice. Berlin: Mouton de Gruyter.
Brown, Penelope and Stephen C. Levinson
1987 Politeness. Some Universals in Language Usage. Cambridge: Cam-
bridge University Press. [1978]
Carter, Ronald and Michael McCarthey
2006 Cambridge Grammar of English: A Comprehensive Guide: Spoken
and Written English Grammar and Usage. Cambridge: Cambridge
University Press.
Caxton, William
1490 Prologue to Eneydos. Reprinted in (1966) The English Language.
Essays by English and American Men of Letters 1490–1839, Whit-
ney F. Bolton (ed.), 1–4. Cambridge: Cambridge University Press.
Devitt, Amy J.
1997 Genre as a language standard. In Genre and Writing, Wendy Bishop
and Hans Ostrum (eds.) , 45–55. Portsmouth: Boynton / Cook.
Eastwood, John
2006 Oxford Practice Grammar. Oxford: Oxford University Press.
Eelen, Gino
2001 A Critique of Politeness Theories. Manchester: St. Jerome Publish-
ing.
Farringdon, Jill M.
1996 Analysing for Authorship. A Guide to the Cusum Technique. Cardiff:
University of Wales Press.
Gnutzmann, Claus and Frauke Intemann (eds.)
2005 The Globalisation of English and the English Language Classroom.
Tübingen: Narr.
Gorji, Mina (ed.)
2007 Rude Britannia. London: Routledge.
Graddol, David
2006 English Next. London: The British Council.
Hewings, Martin
2005 Advanced Grammar in Use. Second Edition. Cambridge: Cambridge
University Press.
Huddleston, Rodney and Geoffrey K. Pullum
2002 The Cambridge Grammar of the English Language. Cambridge:
Cambridge University Press.
Jenkins, Jennifer
2003 World Englishes. A Resource Book for Students. London: Routledge.
Johnson, Samuel
1755 Dictionary of the English Language. London: W. Strahan.
Jones, Daniel
1924 English Pronouncing Dictionary. Revised Edition, with Supplement.
London: Dent and Sons.
Locher, Miriam A.
2004 Power and Politeness in Action: Disagreements in Oral Communi-
cation. Berlin: Mouton de Gruyter.
search 1 (1): 9–33.
2008 Relational work and impoliteness: Negotiating norms of linguistic
behaviour. In Impoliteness in Language, Derek Bousfield and
Miriam A. Locher (eds.), 77–99. Berlin: Mouton de Gruyter.
Milroy, James and Lesley Milroy
1985 Authority in Language. London: Routledge and Kegan Paul.
Oxford English Dictionary
1933 Oxford: Oxford University Press.
1989 (2nd Ed.). Oxford: Oxford University Press [online].
Phillipson, Robert
1992 Linguistic Imperialism. Oxford: Oxford University Press.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik

1972 A Grammar of Contemporary English. Harlow: Longman.
Roach Peter, James Hartman and Jane Setter
2003 Editors’ preface to the 16th edition. Cambridge English Pronouncing
Dictionary. 16th Edition, Daniel Jones, iv. Cambridge: Cambridge
University Press.
Sinclair, John (ed.)
1990 The Collins COBUILD English Grammar. London: Harper Collins.
Swift, Jonathan
1712 A Proposal, for Correcting, Improving and Ascertaining the English
Tongue. Reprinted in (1966) The English Language. Essays by Eng-
lish and American Men of Letters 1490–1839, Whitney F. Bolton
(ed.), 107–123. Cambridge: Cambridge University Press.
Swales, John M.
2004 Research Genres. Exploration and Application. Cambridge: Cam-
bridge University Press.
Trudgill, Peter
1986 Dialects in Contact. Oxford: Blackwell.
1998 Standard English: What it isn’t. European English Messenger 7 (1):
35-39. Reprinted in (1999) Standard English: The Widening Debate,
Tony Bex and Richard J. Watts (eds), 117-128. London: Routledge.
2004 Dialect Contact and New-Dialect Formation: The Inevitability of
Colonial Englishes. Edinburgh: Edinburgh University Press.
Watts, Richard J.
1989 Relevance and relational work: Linguistic politeness as politic behav-
ior. Multilingua 8 (2-3): 131-166.
1992 Linguistic politeness and politic verbal behaviour: Reconsidering
Ehlich (eds.), 43-69. Berlin: Mouton de Gruyter.
1999 The social construction of Standard English: Grammar writers as a
Press.
in press Rudeness, conceptual blending theory and relational work. Journal
of Politeness Research.
Watts, Richard J. and Franz Andres (eds.)
1990 Zweisprachig durch die Schule / Le bilinguisme à travers l’école.
Berne: Haupt.
Watts, Richard J., Sachiko Ide and Konrad Ehlich (eds.)

1992 Politeness in Language: Studies in its History, Theory and Practice.
Berlin: Mouton de Gruyter.
2005 Politeness in Language: Studies in its History, Theory and Practice
(2nd revised and expanded Ed.). Berlin: Mouton de Gruyter.
2001 Die fünfte Landessprache? Englisch in der Schweiz. Zürich: vdf.
Hochschulverlag.
Part I
English over time and space
Chapter 2
Swiss English, German English and
American English: In grammatical alliance
against traditional British English?
D. J. Allerton
1. Introduction1
Like Dick Watts, I was a professor of English linguistics at a Swiss univer-

sity for more than twenty years. Being at heart a grammarian and having
come straight from England and maintained my links through international
commuting, I was, in my early years in the Helvetic Republic, struck by the
grammatical differences (not to mention the phonological and lexical ones)
between British English2 and the English spoken in Germanophone Swit-
zerland. Some time later Peter Trudgill, Dick and I had to confront these
issues more urgently when we jointly directed the Swiss National Research
Council project on English in Switzerland. One of the key concerns of the
project was whether there is such an entity as Swiss English, or whether
there are just specially Swiss varieties of German English, French English
and Italian English. The answer seems to be that Swiss English is a dubi-
ous entity, and that the English used in (Swiss-)German-speaking Switzer-
land is more closely related to the English used in Germany than to the
English used in other parts of Switzerland (cf. Dröschel et al. 2005).3
One of my experiences as a grammatical observer throughout my career
in Switzerland was that, although starting out from the assumption that any
non-British grammatical feature I heard was Swiss-German or German, I
often discovered that the feature in question was actually found in Ameri-
can English. Looking at things more analytically, we should say that there
are some grammatical features (as well as some lexical features) that are
shared by (Swiss-)German English and American English but not found in
traditional British English.4 Such a situation could in principle arise in a
number of different ways:
(A) British English has undergone a new development that has not taken
place in American English whereas German English has retained the
24 D. J. Allerton
traditional (in this case American) form. We can call this development
an original British English innovation, or in brief a BRITICISM. This
term should be distinguished from the more general term “Britishism”,
which covers all specifically British features, whether innovatory or
conservative. When the older form is preserved in American English,
this can be termed an “American conservatism”.
(B) American English has undergone a spontaneous new development that
has not taken place in British English, and that this new development
has then been taken over from American English by German English.
Such a development can be referred to as an original American Eng-
lish innovation, or in short an AMERICISM. Again, this term needs to
be kept apart from the broader term “Americanism”, which covers
both innovatory and conservative American features. Equally, when
the older form is preserved in British English, this can be termed a
“British conservatism”.5
(C) American English has undergone a new development under the influ-
ence of one or more of the immigrant languages (other than English),
and this development has not taken place in British English. Since
non-native speakers of English have always been present in substan-
tial numbers, this is not so surprising a development. It could be de-
scribed as an IMMIGRANT AMERICANISM. If it is clear that the immi-
grant language concerned is specifically German (especially through
the large number of German-speaking immigrants into the United
States in the 19th and early 20th centuries6), this development (C1)
can be termed a German-influenced American English innovation, or
in brief a GERMAN-AMERICANISM. But the precise origin of any lin-
guistic innovation can rarely be pinpointed with accuracy.
All three of these major scenarios have probably occurred, but the last one,
(C) Immigrant Americanisms (with [C1] German-Americanisms as a sub-
variety) may seem somehow to be the least probable. The aim of this arti-
cle is to suggest that, in some cases at least, grammatical Immigrant
Americanisms, and in some cases even some German-Americanisms, are a
plausible hypothesis.
From the viewpoint of speakers of British English, all Immigrant
Americanisms, are simply a variety of Americanisms. They are moreover
Americanisms not only in the sense that they are points that distinguish
American English from British English, but also in the sense that they are
recent introductions into British English from American English. It is
sometimes not pointed out that the general international trend towards
Swiss English, German English and American English 25
Americanization affects English-speaking Britain at least as much as it

affects non-English-speaking countries.
Although the principal focus of this article is on German-like features in
American English that are grammatical in nature, it is perhaps best to begin
with some examples of the corresponding lexical phenomenon, because
these seem to demonstrate the case more clearly. A well-known example is
the word hopefully used as a sentence adverb with the meaning ‘it is to be
hoped that ...’. This is first recorded in British English in 1932 according to
the O.E.D., which declares it to be of U.S. origin, suggesting the German
word hoffentlich as a source.
A further lexical example is provided by the word dumb, which in Brit-
ish English until the 1960s only had the meaning ‘unable to speak’. Since
that time it has however taken on a second meaning, ‘stupid’, which was
already present in American English. This development is nicely demon-
strated by comparing the LOB corpus (1960s British English) and the
Brown Corpus (1960s American English) with the comparable Freiburg
corpora for the 1990s, F-LOB (for British English) and Frown (for Ameri-
can English). The following figures (kindly supplied by Christian Mair of
Freiburg) make the point:
LOB 3 occurrences (all meaning ‘unable to speak’)

Brown 9 occurrences (both meanings)
F-LOB 9 occurrences (both meanings)
Frown 18 occurrences (both meanings)
Although the origin of the second meaning (‘stupid’) in American English

is not certain, it seems more than just an accident that the German word
dumm means ‘stupid’.
Further examples of potential lexical German-Americanisms are:
(a) the verb reach in the sense of ‘pass something by hand’ on the model
of German reichen. The American Heritage Dictionary describes this
use as informal;
(b) the verbs bring and take differentiated not, as in British English, ac-
cording to whether the movement is towards (bring) or away from
(take) the speaker or addressee, but rather, as in German, according to
whether the destination (bring) or the origin (take) of the movement is
emphasised;
(c) bathtub (cf. German Badewanne), rather than simply bath, for the
household “receptacle” in which the bath is taken.
26 D. J. Allerton
Other possible examples might include: the phrasal verb fill out (a form)
for traditional British English fill in or fill up (cf. German ausfüllen); and
the verb hire in the sense of ‘take on as an employee’ (cf. German an-
heuern) rather than the traditional British sense of ‘borrow (a thing) for a
fee’. Some cases can be said to involve underdifferentiation (Weinreich
1968), e.g. the British English distinction between at home with a purely
static meaning (be at home, stay at home, live at home) as against home
with a movemental meaning (go home, get home, arrive home) or a post-
movemental one (be home ‘have arrived home’). It is clear that some of
these changes are purely semantic shifts, while others are additionally af-
fected by the phonological similarity of the English and German lexemes
and can therefore be categorised as “false friends” (Allerton and Wieser
2005). But these lexical examples are not the central issue here; rather, the
concern is with possible immigrant Americanisms (including German-
Americanisms) in the field of grammar.
2. Loss of the perfect
Grammatical features are slightly more intangible than lexical ones. Lexi-
cal items can be recognised from their phonological form, even when this
has been slightly distorted during the borrowing process (as in the fill
out/ausfüllen example cited above). Grammatical patterns, on the other
hand, are more abstract in nature, consisting, as they do, of function words,
affixes, concord, word order, etc. This means that it is more difficult to
unequivocally identify the source of a new grammatical pattern, since po-
tential models in the form of similar (but not identical) patterns may be
found in various languages. It is particularly problematic when a rather
idiosyncratic grammatical distinction in a particular language is lost, and
there is doubt about which of its neighbouring languages (if any) is respon-
sible for the change, since none of these neighbours has that particular
distinction. A classic case of this is the decline of the perfect in modern
American English7 as noted, for instance, in Trudgill and Hannah (2002:
69), cf. also Elsness (1997), and Algeo (2006: 267–268).
In traditional British usage the perfect (with the form HAVE + past parti-
ciple) contrasts with both the simple present and the simple past. Although
the precise interpretation of these forms depends on the actual situation (or
“Aktionsart”) referred to by the lexical verb and its elaboration in terms of
objects, etc., the semantic basis of the contrast is the period of time the
tense-aspect form refers to, roughly as follows;
– the verb form called PAST (with the regular form: V + -ed1) refers to a
period beginning and ending in the past (no matter how recent)
– the verb form called PERFECT (with the regular form: HAVE + V + -ed2)
refers to a period beginning in the past and ending at the present
– the verb form called PRESENT (with the regular form: V + 0/-(e)s) refers
to a period beginning in the past and continuing through the present into
the future.
This account, however, gives the impression that the three forms are on a
linear scale, which is not quite the full picture. There is one essential dif-
ference between the perfect and the other two: whereas the period referred
to by the past (begins and) ends at an arbitrary point in past time, and that
referred to by the present (having begun in the past) ends at an arbitrary
point in the future, the perfect refers to a period with a very clear end-
point, the present moment. For this reason it is better to refer to it as the
present perfect, contrasting it with the past perfect (with the regular form:
HAD + V + -ed2) which refers to an event with an equally clear end-point at
a specified or implied time in the past.
The basic meaning of the present perfect, then, is ‘past event connected
to the present’. However, the way the meaning of the perfect (or for that
matter of the progressive) is interpreted is intimately connected to the kind
of event referred to by the verb structure it is applied to. The most relevant
factor in the case of the perfect is the distinction between momentary (or
punctual) events and those that have duration, i.e. states and processes.
Also relevant is the question of whether the event is a single event or one
that involves repetition.
Consider the following examples:
(1) Gordon has waited patiently for promotion.
(2) Gordon has often waited for the children.
(3) Gordon has occasionally blundered.
(4) Gordon has blundered.
Non-punctual processes, as illustrated by (1), have duration, as do repeti-

tions of events, whether non-punctual or punctual, as exemplified by (2)
and (3) respectively. When such verb structures representing such eventu-
alities are used with the perfect, the natural interpretation is that the even-
28 D. J. Allerton
tuality in question (including the series of repetitions, where there is one)

has continued right up to the present moment. Non-repeated punctual
events like those of (4), however, are limited in time to a particular mo-
ment, so that when they are used with the perfect they require a different
interpretation of the connection to the present time.
In fact quite generally the connection between the past event (referred
to by the perfect) and the present time can be either of two kinds: the past
event (or the period when it took place, e.g. this morning) may have con-
tinuation right up to the present; or it may have some logical connection to
the present, which makes it relevant at the present time. In the case of non-
repeated punctual eventualities only the latter interpretation (i.e. relevance
for the present, rather than continuation up to the present) is possible.
The general meaning of the present perfect, ‘past event connected to the
present’, has important implications for the use of adverbials of time: it
means that adverbials that have a purely past time meaning (e.g. yesterday,
last week, at the meeting last night) are excluded from use with the present
perfect. Another way of formulating this restriction is to say that the pre-
sent perfect requires not only a link with present time, but also that the past
time implicitly referred to is left unspecified.
Time adverbials that do have the possibility of occurring with the pre-
sent perfect must therefore have no explicit reference to a particular past
time. They must also be linked to the present by covering a period of time
that is linked to the present in one of the following ways:
(a) The event took place at a point in time that is so close to the present that
it is still felt to be current (e.g. lately, just ‘a few moments ago’). This is
thus one type of ‘relevance for the present’.
(b) The period is a very general expanse of past time that begins with the
coming into existence of the entities under discussion and extends right
up to the present, but the actual events are individual ones that were
completed in the past, although they have relevance for the present (e.g.
already, (not) yet). This is therefore another case of ‘relevance for the
present’ (rather than ‘continuation up to the present’).
(c) The period is again a very general expanse of past time that begins with
the coming into existence of the entities under discussion and extends
right up to the present, but this time the eventuality itself (either be-
cause it is a process or because it involves repetition) extends through
time right up to the present, and possibly even into the future (e.g. so
far, up till now). It is therefore a case of ‘continuation up to the pre-
sent’.
(d) The period ends at the present, having begun at some particular time in
the past that is specified either by naming its starting point (e.g. since
last Tuesday) or by giving its duration (e.g. for the last six days). Again
this is clearly a case of ‘continuation right up to the present’.
The change that is taking place in the speech of some younger speakers of
British English, probably on the model of American English or perhaps on
the model of non-native English in general, is that the perfect uses of (a)
and (b) are being replaced by the simple past while the uses of (c) and (d)
are being replaced by the present. It is worth looking at some of the adver-
bials individually, starting with type (a).
Consider, firstly, the time adverb just, which referred traditionally to the
very recent past (cf. category (a) above), in sentences such as:
(5) I’ve just arrived.
meaning roughly that I arrived recently and that I am still here as a new
arrival; or:
(6) She’s just had a hip operation.
meaning roughly that she had a hip operation recently and is still possibly
feeling the effects of it. Rather surprisingly, perhaps, this temporal just was
probably the first of the typical perfect-aspect adverbials to become used in
British English in a simple past context.8 By the first decade of the 2000s it
has now become common for younger speakers, especially in informal use.
It is often found, for instance, in advertisements, cf.
(7) The best just got better.
[advertisement for Orion Optics in Sky at Night magazine, June 2006]
Why should the perfect have been dropped so readily in this particular
context? Perhaps because the usage was rather subtle and complex, as the
following examples of the older usage illustrate:
(8) (i) I’ve just seen Mary.
(ii) I saw Mary a few minutes ago.
(iii) I saw Mary just a few minutes ago.
(iv) I just saw Mary a few minutes ago.
The point that the examples of (8) demonstrate is that, although just with
its ‘recency/current relevance’ meaning is naturally used with the perfect,
this use becomes impossible as soon as an explicit past time adverbial ap-
30 D. J. Allerton
pears in the same clause. Even the time adverbial just now normally re-
quires a past verb form. It is perhaps only natural that such subtleties of
usage should prove difficult for native speakers of other languages, includ-
ing immigrants into English-speaking countries, such as the United States.
The adverb lately was placed above in category (a) alongside just, be-
cause it typically refers to an event that took place at a point in time that is
so close to the present that it is still felt to be current, but the two adverbs
differ in their reference to punctuality or duration. Whereas just typically
refers to a single punctual event that is linked to the present through cur-
rent relevance, lately typically refers to a repeated or habitual event, which
thus has duration, with the period of repetitions in principle extending right
up to the present. For this reason the most common replacement for the
perfect is the present rather than the past tense. Looking at examples of
lately found in corpora including sources provided by internet search en-
gines like Google, we find that 90% of examples involve some form of the
perfect. It very commonly occurs in questions, a classic case being of the
type “(Have you) read any good books lately?”, although statements of the
form “I haven’t been feeling too well lately” are also common. A further
typical example is found in the Country and Western song:
(9) Have I told you lately that I love you? ... Well, darling, I’m telling you now.
This example also nicely points up the contrast with typical adverbs of
present time (whether present simple or present progressive) like now.
However, as already noted, lately is being used more and more with the
present, as illustrated by the following examples (all taken from U.K.
sites):
(10) Lately I just cannot be in a monogamous relationship. But there are people ...
(11) Why do I feel nauseous lately?
(12) I’m not using the Ferrari’s much lately.
(13) What’s up with the Pope lately?
It is also occasionally found used with the simple past.

Turning to type (b) adverbials, which cover a wide period extending
right up to the present but highlight completion in the past, we find already
to be the principal member of the group. In traditional British English this
is a classic adverb for use with the perfect, and indeed Google “results
from the U.K.” did give (on 4th October, 2006) the following result:
“already done”: about 1,020,000 hits

“already did”: about 73,900 hits.
But, although this may seem a very good result in favour of perfect use,
corresponding to roughly 93.3%, it would at one time have been over 99%;
and the figure for spoken language may well now be less than 90%. Typi-
cal examples of already10 with the simple past are:
(14) I don't need bribing by the police to "shop" a drunk driver, I already did so. I
saw a man who was clearly drunk preparing to drive away in a car that …
(15) The problem is not that I cannot update the filter myself: I already did that.
The problem is rather that when our students download references from ...
The temporal adverb yet is the negative and interrogative equivalent of

already and is coming to be used in a similar way with the simple past.
Our third category (c) involves adverbials that refer to an activity or
state that extends right up to the present, typical examples being so far, up
till now. In present-day British English such adverbials, which traditionally
strongly favoured the use of the perfect, are occurring more and more with
the present, as the following examples show:
(16) So far the mood is carnival-like ...

[part of news report on demonstration at power station, BBC Radio 4 11 a.m.
news summary, 31st August, 2006]
(17) So far things are going well.

[words spoken by programme presenter on Grand Designs, shown on More 4
television (U.K.) 22nd August, 2006]
In traditional British English the verb forms has been and have been going
respectively (rather than is and are going) would have been used in these
sentences.11
For our last category (d) the period referred to ends at the present, hav-
ing begun at some specified time in the past. This particular past time,
whether given directly (with an adverbial introduced by since) or through a
specification of the length of the period up to the present (with an adverbial
32 D. J. Allerton
introduced by for), seems to cause the perfect to be replaced with the sim-
ple past, as the following British examples12 testify:
(18) He was here since his school days.
[said by Jean Swainbank, character in Dalziel and Pascoe: Fallen Angel,
BBC 1 television, 10th September, 2006]
(19) We have to understand one thing: that since September the 11th the world did
change.
[Gordon Brown, M.P., the Today programme, BBC Radio 4, 25th September,
2006]
Occasionally the perfect is replaced with the present, cf.:
(20) Since 10 October 1994, all driving licences contain the holder’s consent to be
an organ donor, and ...
[official U.K. information leaflet “How to fill in your renewal driving licence
form”, edition 07/05, p. 5]
All four types of time adverbial that traditionally show solidarity with the
perfect thus show signs of allowing either the past or the present in its
place. There is also evidence that when the perfect is situationally rather
than linguistically conditioned, it suffers the same fate: for instance, the
traditionally used exclamation of euphoric success on completing a diffi-
cult task I’ve done it! is apparently being replaced by children with I did
it!, possibly under the influence of American electronic media. This gen-
eral decline of the British English perfect and its partition between simple
past and simple present is something I first pointed to in a lecture at the
University of Stuttgart in 1980. Now, more than 25 years later, it seems to
have gathered momentum.
As noted at the beginning of this section, it is difficult to identify the
source of the loss of the perfect versus past tense distinction in American
English and later British English. Of the numerically significant American
immigrant languages only Spanish has a comparable distinction. But Ger-
man is at least a prime candidate for influencing the loss of the distinction,
since it has distinct perfect and past verb forms which differ stylistically
more than semantically.
3. Replacement of auxiliary have with be before a past participle
This phenomenon is related to the one just discussed, since it involves

what is, superficially at least, a perfect verb form. As we have already
noted, this is regularly formed in English with the auxiliary have, whilst in
older English, as well as in other languages, such as French, German and
Italian (though not, say, Spanish or Swedish) a minority of (mostly intran-
sitive) verbs formed their perfect with the auxiliary be. This feature has
also appeared in certain cases in American English, and latterly also in
British English. It may be an Americism (in the sense defined above), but it
is more likely to be an American conservatism or an immigrant American-
ism. As regards British English the pattern seems to be slipping in almost
unnoticed. How can this be?
Consider the following example with the meaning ‘The liquid assets
have disappeared’:
(21) The money’s gone.
The reduced auxiliary ’s is of course ambiguous between the auxiliary

verbs have (i.e. has) and be (i.e. is). But since go is an intransitive verb, it
would seem logical to assume that the reduced auxiliary is in fact has, and
that has gone is the perfect form of go. This is the traditional British usage.
The negative form of (21) (with a reversed polarity tag attached) would
therefore be:
(22) The money hasn’t gone (, has it?).
However there is a growing tendency in British English (which probably

originates from American English) to use (23) in place of (22)
(23) The money isn’t gone (, is it?).
The preference for be over have also becomes evident when the subject is
plural, because dudm the reduced forms of be and have are distinguished
(though in their unstressed forms only as /?/ are and /?u/ or /u/ have re-
spectively, with the distinction resting on the presence of a lenis labioden-
tal fricative).
It is important to note that for speakers who use the be + past participle
construction in this way, it is not a generalised grammatical pattern for all
intransitive verbs with this kind of meaning, since they would not say (24)
but rather (25):
(24) *The money isn’t disappeared (, is it?).

34 D. J. Allerton
(25) The money hasn’t disappeared (, has it?).
The verbs that seem to be adopting the pattern exemplified by be gone (in
place of have gone) are essentially intransitive change-of-state verbs, al-
though most of them also have a transitive use with a different meaning.
They then come to be treated like “ergative pattern” verbs, which allow
two different valency patterns. Let us consider for a moment the three basic
verb types, transitive-only (e.g. modify), ergative i.e. dual use as transitive
with ‘affected’ object and intransitive with ‘affected’ subject (e.g. im-
prove), and intransitive-only (e.g. deteriorate).
In the case of purely transitive change-of-state verbs, we can observe
the following pattern, exhibited by the verb modify:
(26) (a) They have modified the service.

(b) The service has been modified.
(c) The service is modified.
(d) *The service has modified.
The (a) sentence is an active use of the verb with the perfect form of the
verb; it refers to an action that “they” have done. The (b) sentence is the
equivalent passive sentence, which also refers to an action, but it is about
something that has happened to the service; it is a so-called “short pas-
sive”, which lacks a by-phrase referring to the agent. The (c) sentence is
descriptive, using the past participle as an adjective that describes a state
that has resulted from an action. The (d) sentence is ungrammatical, be-
cause it would involve an intransitive use of modify.
Now consider the possibilities for a verb like improve, which can be
used both transitively (with the changer as the subject, and the changed
entity as the object) and intransitively (with the changed entity as the sub-
ject):
(27) (a) They have improved the service.

(b) The service has been improved.
(c) The service is improved.
(d) The service has improved.
This time the fourth sentence is grammatical; but there is a slight differ-
ence in meaning between (27c) and (27d), in that the former, in which im-
proved is an adjectival past participle, refers to a new, different state, while
the latter refers to the change of state that has taken place.
Things look different for our third class of verbs, those that are purely
intransitive:
(28) (a) *They have deteriorated the service.
(b) *The service has been deteriorated.
(c) *The service is deteriorated.
(d) The service has deteriorated.
This time only the change of state construction with have is possible.
The change in British English we are now considering involves using an
intransitive verb such as go (which was previously used just like deterio-
rate) in a new way, so that it behaves in the perfect like an ergative verb
such as improve. Another frequently used example involves the past parti-
ciple of the verb finish with the verb be (in place of the traditional auxiliary
have) when the meaning relates to the completion of a task, so that He’s
finished is interpreted as involving auxiliary be rather than have. As a re-
sult, the first of the following alternatives is regularly replaced by the sec-
ond:
(29) (a) Has he finished?

(b) Is he finished?
In traditional British English the two questions of (29) would have had
different meanings, (a) ‘Has he completed the task?’; (b) ‘Is he ruined?’.
An interesting rather similar example is (30):
(30) QUESTION...How do I get the right picture format?

ANSWER: ....Highlight ‘Save New Setting’, press ‘Select’, and you’re done!
[Sky Magazine, February 2006, 101.]
where the traditional form would be ...and you’ve done in dialect, but
...and you’ve finished in standard English. Indeed the form be done in Brit-
ish English suggests a different meaning entirely (‘be swindled’).13
It is interesting to note that if we translate into German our earlier ex-
ample (21) (The money’s gone) alongside (29a), we get (31) and (32) re-
spectively:
(31) Das Geld ist weg.
(32) Ist er fertig?

36 D. J. Allerton
In both cases the corresponding sentence involves an adjectival construc-

tion with be. It is tempting to think that the use of the verb be in corre-
sponding constructions in English may have been influenced by the pat-
terns used in the English of nineteenth century German-speaking
immigrants into the American Mid-West and/or twentieth century German-
speaking Jewish immigrants.
The phenomenon has however gone much further than this, perhaps be-
cause it has linked up with the conservative use of be gone etc. found in
earlier English (for instance, that of the King James Bible of 1611). Indeed
the use of be gone has even been preserved in traditional British English
with the meaning ‘be away from a place’ in contexts such as How long has
he been gone? In any case, by now a whole group of further verbs seem to
have been reassigned in this way, including agree, change, finish, do (in
the sense of ‘finish’), grow up, move (in the sense ‘move house’), qualify,
retire, stop. The effect of this is that be agreed is used equivalently to have
agreed, etc.
Consider the following examples:
(33) ‘Nicole’ is all grown up

[Headline in motoring section of The Independent (London) 21st February,
2006]
(34) Light the torch and move it slowly and evenly over the surface until all the
sugar is melted, golden and bubbling.
[Reply to query about crème brûlée, Good Housekeeping (London), March
2006, p. 76]
(35) The atmosphere at the conference is changed.

[example adapted from statement by expert on Russian politics on the Today
programme, BBC Radio 4, 24th February, 2006]
(36) Now, are you guys moved out, or ...?

[Kirstie Allsop, in “Relocation, relocation” on Channel 4 television, 8th
March, 2006]
(37) ...particularly when you’ve got batsmen who are settled.

[Ian Botham, Sky Sports television cricket commentary, 19th March, 2006]
(38) The blood on the back of his head is all dried up.
[words spoken by a pathologist in the Scottish detective series episode
Taggart: a death foretold]
(39) The drinks break is now ended.

[Henry Blofeld, cricket commentator, BBC Test Match Special, 28th July,
2006]
In all of these cases, traditional British English (as found in the present
writer’s usage) would have preferred auxiliary have to be, and for tradi-
tional speakers the use of be suggests a link to a passive form with have
been. For instance, in the last example in the list, the interpretation is that
the drinks break has been ended by someone. This is presumably not the
case for speakers (whether American or British) who have the new ergative
use for such verbs. They can use auxiliary have and auxiliary be with all
these verbs with the subtle difference in meaning described above for the
ergative verb improve. An example of this awareness is found in the fol-
lowing extract:
(40) First it is the case that absolute emissions are and have increased in recent
time [sic!].
[Paula Dobriansky, Head of the U.S. delegation to the Climate Change Con-
ference at Monterey (in Mexico), speaking on the Today programme, BBC
Radio 4, 5th October, 2006]
With are increased the speaker is presumably suggesting a new state,

whereas with have increased she must be referring to a change of state. But
only the form have increased would be available to traditional British
speakers.
Finally on this developing replacement of have with be, it is perhaps
worth asking if there is a link to the previous phenomenon discussed, i.e.
the loss of the perfect. Once the perfect has disappeared from use, the sta-
tive perfect of result found in uses like The money has gone lacks a suitable
form of expression, the simple past (The money went) not having the right
association with a result. The perfective meaning thus has nowhere to go,
and has to seek a different form of expression.
4. Unorthodox agreement in copular sentences
A third phenomenon that seems to be entering current British English from

American English and is possibly partly German in origin is a matter of
grammatical agreement in copular sentences. At first sight it seems clear
that in a “nominative-accusative” language like English (as opposed to an
“ergative-absolutive” language like Basque) the finite (or “tensed”) verb
38 D. J. Allerton
regularly agrees with the subject, in both transitive and intransitive clauses
(even when the verb is of the ergative-absolutive type). In traditional Brit-
ish English this generalization also holds for nominal copular clauses, i.e.
those with a verb like be, become, remain, seem. It is one subset of these,
i.e. equative clauses with different grammatical number in subject and
predicative, that gives rise to the problem.
If we consider an example14 like:
(41) Their principal crop during the later part of the year is/are potatoes.
we find that the traditional analysis would identify the initial (preverbal)
noun phrase their principal crop during the later part of the year as sub-
ject, and that this singular noun phrase would demand the singular verb
form is. There is, however, a growing tendency in British English, follow-
ing what seems to be a generalised pattern in modern American English, to
prefer a plural verb form, in this case are.
What is the source of this preference for plural concord in cases where
the following predicative noun phrase is plural, but the preverb subject
noun phrase is singular? There seem to be at least two different factors
involved: the first is the distance between the subject noun phrase and the
agreeing verb, with the nearest noun phrase triggering so-called “contact
agreement”; the second seems to be a possible reinterpretation of which
noun phrase constitutes the subject.
Consider these examples:
(42) The advantage are the extra tools.

[HydrogenAudio website, observed 3rd February, 2006]
(43) The other party that’s going to be a key player in the talks are the Ulster Un-
ionists.
[Shaun Ley, The World at One, B.B.C. Radio 4, 1st February, 2006]
(44) RANDALL BOE: It does do a lot. It sets uniform rules for commercial e-
mailers. It provides a lot of additional penalties to use against the outlaw
spammers. I mean, the problem is not commercial e-mailing – generally. The
problem is not the L.L.Beans of the world to pick on them. The problem are
the outlaws.
[www.pbs.org website 3rd February, 2006: online news hour]
In these equative sentences with the verb be one of the noun phrases refers
to some sort of entity that can be readily identified, whereas the other noun
phrase names an epithet in the form of a definite description that is claimed

as appropriate for the identified entity. In some languages the identified
entity demands subjecthood, so that, for instance, in the following German
translation of the first of the above sentences, the noun phrase die
zusätzlichen Werkzeuge must be the subject:
(45) Der Vorteil sind die zusätzlichen Werkzeuge.
Of course the German subject is not bound to clause-initial position in the

same way that it is in English. In traditional British English the pre-verb
noun phrase is taken as the subject and therefore controls agreement. On
the other hand, speakers of Modern English have long felt unsure about
agreement, and have tended towards the kind of “contact agreement” to
which Fowler (1931: 75) draws attention (e.g. The Ordination services of
the English Church states …). When the preverb noun phrase has a long
postmodifier (such as the relative clause in the second example) that sepa-
rates the head noun from its potentially agreeing verb, the attraction of the
postverb noun phrase can thus be stronger. As regards the influence of non-
native speakers, a word-order predicative-be-subject is more common in
German than in French or Spanish, for example. The tendency of German
immigrant speakers of English to prefer a more strongly referential entity
as subject may then have been a crucial factor in establishing the current
trend.
5. Other grammatical patterns
One grammatical pattern that may possibly reflect the influence of German
is the lack of tense agreement in cleft sentences. In British English it-cleft
sentences are typically of a form such as:
(46) It is in Blackpool that the conference is taking place.
(47) It was Margaret who/that led the party at that time.
The only point about such sentences that concerns us here is that the verb
be in the it-clause agrees in tense with that of the that/wh- clause. If, for
instance, the first sentence had had was taking place, it would have re-
quired was in the it-clause. This at least is the normal pattern in traditional
spoken British English, as both Quirk et al. (1986: 1386) and Huddleston
40 D. J. Allerton
and Pullum (2002: 1419–1420) acknowledge. Both authorities also point to

the possibility of a present tense in the it-clause preceding a past tense in
the that/wh- clause, though perhaps failing to point out that such usage can
sound stilted or even unnatural,15 at least in British English. The tendency
to use a uniform present tense in the it-clause of all cleft sentences is, how-
ever, not limited to German. It is equally normal in French, for example.
But German speakers may nevertheless have significantly contributed to its
use in the United States.
A further pattern that may reflect German usage but equally well a num-
ber of other languages is what has been called “infinitivitis” (Allerton
1988). English is unusual in having two partly contrasting non-finite verb
forms, the infinitive and the gerund; and non-native users of the language
find this distinction difficult, usually preferring to generalise use of the
infinitive, even when tradition requires the gerund. In standard British Eng-
lish, verbs such as like and love have a systematic difference in meaning
between a less common infinitive construction which suggests doing some-
thing habitually because one likes it, and a more common gerund construc-
tion which means that one likes it however rarely one does it. A random
Google search (carried out on 7th February, 2007) came up with the fol-
lowing statistics:
Table 1. liked to work versus liked working

UK the web
liked to work 693 (= 4.68%) 78,000 (= 30.12%)
liked working 14,100 (= 95.32%) 181,000 (= 69.88%)
TOTAL 14,793 (= 100%) 259,000 (= 100%)
This seems to demonstrate that international internet writers overuse the

infinitive compared with British English norms. In a similar way non-
native speakers come to use such sequences as the possibility to achieve
success rather than the possibility of achieving success. A random Google
search (carried out on 10th October, 2006) produced the following results:
Table 2. possibility to achieve versus possibility of achieving

UK the web
the possibility to achieve 382 (= 1.76%) 44,100 (= 17.78%)
the possibility of achieving 21,300 (= 98.24%) 204,000 (= 82.22%)
TOTAL 21,682 (= 100%) 248,100 (= 100%)
The infinitive construction is clearly more popular amongst American

and/or non-native speakers, but there is no way of knowing the precise
contribution to this that has been made by German-speakers.16
A grammatical usage that is more likely to have been influenced by
German speakers is the adverbial use of some and any in such American
English sentences as:
(48) (a) It really hurt some. [British English: It really hurt (quite a bit).]
(b) Did it hurt any? [British English: Did it hurt at all?]
It seems reasonable to suggest a substratum effect of German etwas, which

is used as both a determiner corresponding to some/any and as an adverb in
the sense of ‘a bit, rather’.
A further minor grammatical pattern that could well represent a calque
on a German construction involves the omission of the verb go after a mo-
dal verb when a direction adverbial follows, as in cases like (49a) and
(50a), which can be linked to the German pattern of (49b) and (50b) re-
spectively:
(49) (a) I must out.

(b) Ich muss heraus.
(50) (a) He wants in.

(b) Er will herein.
It is true that similar patterns are to be observed in Scottish and Irish Eng-
lish, but these, too, may involve a substratum effect. For this phenomenon
the case is not proven, and it may well be that it involves a conservatism
reinforced by a German-Americanism.
No doubt there are still further German-Americanisms. Some of these
have not yet reached British English. In some cases this is because their
use is not universal in American English. For instance, the use of non-
volitional would in if-clauses (If I would be late tomorrow, ... etc.) or the
use of take me with, etc. without a following noun phrase indicating the
companion, are German-Americanisms that are not yet prominent enough
in American English to cross the Atlantic.
42 D. J. Allerton
6. Conclusion
After this brief survey we can unfortunately only come to a very tentative
conclusion. Swiss English and German English seem to be subvarieties of a
developing Global English used by millions of non-native speakers. Be-
cause of the mounting domination of American political and cultural influ-
ence in the world today this new Global English is closer to American Eng-
lish than to British English. Indeed it is perhaps becoming ever closer, as
American English (and secondarily also British English) is affected by the
number of non-native speakers. Speakers of German, particularly immi-
grants into the United States in the 19th and early 20th centuries have cer-
tainly had a part to play in the development, as evidenced by Brogan’s
(1999: 488) figure of 7 million German-Americans for 1919. Without a
detailed examination of the historical records, however, the precise extent
of this influence is difficult to assess.
Notes
1. I am indebted to Christian Mair (University of Freiburg-im-Breisgau) for his

insightful comments and helpful suggestions on an earlier version of this article.
I am also grateful to Miriam Locher and Jürg Strässler for some further im-
provements.
2. Although I am a speaker of English English, I believe that most of the linguistic
features referred to in this article as “British English” apply equally in Scottish
English and Welsh English. An exception is the phenomenon exemplified by
examples (49) and (50) below.
3. If Swiss English did exist, it would presumably be a variety of Euro-English
(Görlach 2001, Jenkins et al. 2001) and ultimately a subvariety of World Eng-
lish (Cheshire 1991; Allerton et al. 2002; Trudgill and Hannah 2002; Jenkins
2003; Mollin 2006).
4. Actually some have since started appearing in recent British English.
5. The notions “British conservatism”, “Britishism” and “Briticism”, of course,
represent oversimplifications, in that they do not distinguish English English
from Scottish English and Welsh English. But for most of the phenomena dis-
cussed here the issue is the contrast between American English and all forms of
British English.
6. German-speaking immigrants include of course speakers of German dialects
and can be taken to include Yiddish speakers.
7. This phenomenon is now being found in advanced British English.
8. This usage can be found in British English literary texts as early as 1951, as
witnessed by this example from Anthony Powell’s A Buyer’s Market (as it ap-
pears in A Dance to the Music of Time: Spring, London: Arrow Books, page
505): ‘I think I may be seeing something of Prince Theodoric’, he said. ‘I be-
lieve you just met him.’
9. The fact that such uses involve ambiguity for the word just (1. ‘a few mo-
ments ago’; 2. ‘simply’) seems not to inhibit this new use of the past tense
form.
10. A further German-Americanism not discussed here is the general over-use of
already, particularly in an emphatic rather than a temporal sense, as in the
words of the American father impatient for his child to develop linguistically:
“So DO something already.” in Rick Kirkman and Jerry Scott’s (2000: 21)
Baby Blues. I am indebted to Nadja Nesselhauf (University of Heidelberg) for
this example.
11. It is possible that so far is being used in such cases by news reporters and
others in a new sense, roughly ‘all I have been able to witness up to the pre-
sent is that ...’. But the effect in any case is to loosen the bond between perfect
aspect and adverbials that refer to a period ending at the moment of speaking.
12. It is virtually impossible to provide unequivocal examples of the change from
perfect to past for adverbials of the type for the last week/two weeks/month/
etc., because the resulting sequence is always a possible sentence; but for the
traditional British speaker it signals the wrong meaning, suggesting a process
that has ended before the present, when the process referred to actually con-
tinued up to the present.
13. This use of be done is attested in American English as early as 1939, as dem-
onstrated by this quotation from I. Anderson’s review of The Big Sleep: “Be-
fore the story is done, Marlowe just misses being an eyewitness to two mur-
ders ...”
14. Taken from Allerton (1992), who gives many more examples of such prob-
lematic sentences.
15. Cf. the following cited by Quirk et al. (1986: 1386 [note a]) as an example: “It
is Miss Williams that enjoyed reading novels as a pastime.” It is difficult to
imagine any British speaker actually saying this.
16. German-speakers and perhaps other native speakers may also be influenced
by a non-native tendency to use the noun possibility as a synonym for oppor-
tunity.
44 D. J. Allerton
References
Algeo, John
2006 British or American English? A Handbook of Word and Grammar
Patterns. Cambridge: Cambridge University Press.
Allerton, D. J.
1988 ‘Infinitivitis’ in English. In: Studies in Descriptive Linguistics. (Es-
says on the English language and Applied Linguistics on the Occa-
sion of Gerhard Nickel's 60th Birthday), Josef Klegraf and Dietrich
Nehls (eds.), 11–23. Heidelberg: Julius Groos.
1992 Problems of Modern English grammar II – Disagreement about
agreement (findings); and new topic. English Studies 73: 458–470.
Allerton, D. J., Paul Skandera and Cornelia Tschichold (eds.)
2002 Perspectives on English as a World Language. International Cooper
Series in English Language and Linguistics 6. Basle: Schwabe.
Allerton, D. J., Cornelia Tschichold and Judith Wieser (eds.)
2005 Linguistics, Language Learning and Language Teaching. Interna-
tional Cooper Series in English Language and Linguistics 10. Basle:
Schwabe.
Allerton, D. J. and Judith Wieser
2005 The true nature of “false friends”. In: Allerton et al.(eds.), 57–83.
American Heritage® Dictionary of the English Language.
2000 4th Edition.
Brogan, Hugh
1999 The Penguin History of the U.S.A. 2nd edition. Penguin: London. 1st
edition, 1985.
Cheshire, Jenny (ed.)
1991 English around the World: Sociolinguistic Practices. Cambridge:
Dröschel, Yvonne, Mercedes Durham and Lukas Rosenberger
2005 Swiss English or simply non-native English? A discussion of two
possible features. In: Allerton et al. (eds.), 161–176.
Elsness, Johan
1971 The Perfect and the Preterite in Comtemporary and Earlier English.
Berlin: De Gruyter.
Fowler, H. W.
1926 A Dictionary of Modern English Usage. Oxford: Clarendon Press.
Görlach, Manfred
2001 A Dictionary of European Anglicisms: A Usage Dictionary of Angli-
cisms in sixteen European Languages. London: Oxford University
Press.
Huddleston, Rodney and Geoffrey Pullum

Jenkins, Jennifer
2003 World Englishes: a Resource Book for Students. London: Routledge.
Jenkins, Jennifer, Marko Modiano and Barbara Seidlhofer
2001 Euro-English. English Today 68: 13–19.
Mollin, Sandra
2006 Euro-English: Assessing Variety Status. Tübingen: Narr.
Quirk, Randolph, Sydney Greenbaum, Geoffrey Leech, and Jan Svartvik
1986 A Comprehensive Grammar of the English Language. London:
Longman.
Trudgill, Peter and Jean Hannah
2002 International English. 4th edition. London: Arnold. 1st edition 1982.
Weinreich, Uriel
1968 Languages in Contact. 2nd edition. The Hague: Mouton. Original
edition: New York: Publications of the Linguistic Society of New
York.
Source of literary quotations

Anderson, Isaac
1939 Review of The Big Sleep, Raymond Chandler. New York Times.
Kirkman, Rick and Jerry Scott
2000 Baby Blues. Chicago: Contemporary Books.
Powell, Anthony
2000 A Dance to the Music of Time: Spring. Reprint. London: Arrow
Books. Original edition, London: Heinemann, 1951.
Chapter 3
Regional variation in English in the new
millennium: Looking to the future
Katie Wales
1. Introduction: Prophesying the future of English
It is no surprise that, as the twentieth-century neared its close, and a new

millennium beckoned, futurology in many disciplines became very promi-
nent. English language studies were no exception, particularly in relation to
conjectures about the continuing hegemony of English as the major global
language: see particularly Crystal (1997) and Graddol (1997). However,
prophecies about the future of English are by no means confined to the last
ten years or so. Although documentation from earlier centuries is tantalis-
ingly scarce, there is Samuel Daniel’s speculative question in his poem
Musophilus right at the end of the sixteenth century, when the idea of Eng-
lish as a world language would have been undreamed of:
And who in time knows whither we may vent
The treasure of our tongue, to what strange shores
The gain of our best glory shall be sent,
T’enrich unknowing nations with our stores? (1965: 86)
Speculations about whether international varieties of English, notably

American English, would markedly ‘diverge’ from English English or
‘converge’ with it, have been aired most famously by Noah Webster (1789)
over 200 years ago, in his Dissertations. He (unlike Crystal) favours a ‘di-
vergence’ theory: there will result “a language in North America as differ-
ent from the future language of England, as the modern Dutch, Danish and
Swedish are from the German, or from one another”. His optimism was
certainly partly willed by an agenda of North America’s political separa-
tion from England. As Jonathan Boucher (d.1804) explicitly put it: “in no
very distant period, [Americans’] language will become as independent of
England as they themselves are” (cited Quirk 1972: 2). Most recently
Burchfield believes the process of “multiplying” into a thousand varieties
will take “hundreds of years”, but nevertheless “it will happen”, despite the
48 Katie Wales
pull of mass media, global communications and travel (cited McCrum,

Cran and Macneil 1992: 393–394; see also Burchfield 1994: 15). Henry
Sweet’s time-scale for this process was much shorter: a mere hundred
years: in 1977 “England, America and Australia will be speaking mutually
unintelligible languages, owing to their independent changes of pronuncia-
tion” (cited Crystal 1997: 134). Clearly Burchfield was right to be more
cautious in his own prediction.
I shall return to the issue of convergence and divergence in my conclu-
sion; my concern here is with predictions about the English language on
the micro- rather than macro-level, internal to England, rather than external
to it. It has to be said in general that any sense of optimism is sadly lack-
ing: the overall tone of any prediction about English dialects is decidedly
apocalyptical, whether at the end of a century or the beginning, as their
decline and even disappearance is anticipated. It is in regional variation
that language change is dramatically revealed to layperson and scholar
alike, and the kind of language change that is viewed the most apprehen-
sively, so significant is location, locale for one’s identity. The concerns I
shall be debating here include the extent of dialect ‘death’ and the progno-
sis for degree of robustness in the future.
It should be noted at the outset that I shall be using ‘dialect’ as a con-
venient synonym for ‘variety’. I recognise that there are no clear-cut dialect
boundaries, and that clines or continua of variation exist, diachronically
and synchronically. I would not therefore go so far as Upton (2006) in
avoiding the notion of ‘dialect area’ altogether. If it is a ‘fiction’ (Upton
2006: 311) it is a highly salient concept nonetheless in perceptual terms for
speakers of English. It is also a highly convenient term to describe pithily
urban as well as rural variation, whether in Liverpool, Newcastle or Som-
erset and Cumbria. As Trudgill (1998: 38) writes, “we use names for dia-
lects that we happen to want to talk about for a particular purpose as if they
were discrete varieties”.
Over two hundred years ago we find a London-based reviewer of Fran-
cis Grose’s Provincial Glossary (1787) (cited Bailey 1996: 267) suggesting
that “frequent communications with the provinces will soon make little
variety, except in the wilds of Exmoor, the barren heaths of Northumber-
land, and the sequestered mountains of Lancashire”: anticipating modern
ideas of dialect ‘levelling’ to which I shall return. Just before this, ‘Tim
Bobbin’, or John Collier of Urmston in the border area of Lancashire and
Cheshire, casts an interesting light from the provinces on what appears to
be the encroachment of ‘standardisation’. In the ‘Observations’ which
Regional variation in English in the new millennium 49
mark the preface to the 1775 edition of his extremely popular ‘dialogue’
Tummus and Meary (first published 1746) he comments:
But as Trade in a general way has now flourish’d for near a Century, the in-
habitants not only Travel, but encourage all Sorts of useful Learning so that
among Hills and Places formerly unfrequented by Strangers, the People be-
gin within the few years of the Author’s Observations to speak much better
English. If it can properly be called so … (Collier 1775: 10)
In the nineteenth century, antiquarians continuing in the long tradition
of recording dialect words from John Ray (1674) and Francis Grose on-
wards, were clearly alarmed at the implications for the long-term future of
rural speech of not only the encroachment of Standard English, but also
better education, increased mobility, improved communications and, by
then, industrialisation. We can note the title of Holloway’s A General Dic-
tionary of Provincialisms, Written with a View to Rescue from Oblivion the
Fast Fading Relics of By-Gone Days (1839). The founding of the English
Dialect Society in 1873, and the comprehensive works of Alexander Ellis
(1888) and Joseph Wright (1905) for example, bear witness again to the
zeal to record ways and words of rural speech before the ‘adulteration’ by
urban speech and inevitable recession and oblivion. (See, for example,
Wright’s preface to his Grammar 1905). The Reverend Heslop laments in
his own preface to his Northumberland Words (1892: xx–xxi) that “the
tendency to assimilate the form of the dialect with the current English of
the schools is increasing”; and Nicholson’s Folk Speech of East Yorkshire
(1889) opines that “steam, electricity and education are surely killing dia-
lects” (cited Mugglestone 2006: 293). The rapid growth of industrial cities
North of the Trent was seen as a threat to ‘pure’ or ‘genuine’ rural speech
by ‘adulteration’ from the new urban varieties. So Addy (1888: viii) la-
ments the incursion of immigrant cutlery trade workers into Sheffield, to
the detriment of local speech.
Move forward half a century and Harold Orton at the University of
Leeds is initiating after the Second World War the only full-scale and sys-
tematic Survey of rural dialects across the whole of England, precisely out
of his concern that unless he did so, valuable information for the history of
English would be lost for ever. By the 1960s, certainly, Received Pronun-
ciation was deeply embedded in England’s social consciousness; and tele-
vision could be added to the list of improved communications. Yet Richard
Hoggart writing in 1957 laments the potential loss of local and close-group
entertainments in the face of the onslaught of mass media.
50 Katie Wales
Move forward another half century – and to the end of the millennium,
and as one might expect, predictions suggest either a general ‘levelling’
towards greater uniformity (in the direction of London English); or ‘reces-
sion’ and even complete disappearance. The British press is particularly
prone to the gloomy and sensational: “Scouse accent sinking into the estu-
ary”, proclaimed the Daily Telegraph 1st June 1999, “in danger of extinc-
tion from the relentless northward march” of Estuary English. The Inde-
pendent (April 2nd 2004) reports that “some have forecast that Geordie
could disappear within 30 years”. Ironically, these particular accents, of
Liverpool and Newcastle respectively, played no part in the surveys and
dialect studies of the past, being essentially the ‘new’ products of the ur-
banisation and industrialisation which philologists had tended to ignore.
But even these significant varieties are perceived as threatened, and espe-
cially by the southern force of ‘Estuary English’, to which I return below.
In general terms, even allowing for his particular focus on Northern
English, Griffiths (1999), an editor of the journal Northern Review, repre-
sents the very commonly held view that the distinctive traditional dialect
vocabulary has ‘shrunk’ or ‘eroded’ quite considerably in the last fifty
years under the influence of urbanisation: consider the end of century/
millennium surveys by J. Widdowson (1999) in the Sheffield area,
Simmelbauer (2000) in Northumberland and Rhodes (2000) in West York-
shire. Griffiths also notes how dialect grammar (especially written) is very
close to that of the Standard; Widdowson (1993: 8) again predicts “less
and less variation … in grammar and syntax”. Griffiths and others follow
those other commentators in the last hundred years who have pointed to the
influence of schooling, pressure to ‘get on’ socially, the dissolution of tra-
ditional class structures, as well as mobility and migration.
2. Metaphors, models and caveats
It is well known in futurological studies generally that predictions depend

not only on the data, but also on the nature of the model which frames the
evidence. What is very obvious from this summary of predictions about
English dialects since the late eighteenth century is the sense of the inevi-
tability of loss. And yet, regional variation in English has not totally disap-
peared, but on the contrary is still very much in evidence today (see also
section 4 below). Moreover, in the early years of this millennium, as in the
1980s, and the 1950s, it is always the older generations who sound the
‘broadest’ to the younger (see Elmes 2005, passim). Of course, however,
inevitably, there have been linguistic changes over the centuries. A com-
mon metaphor is of death and degree of moribundity or ‘decline’, as if
dialects are living creatures: not unsurprising in one sense, since they are
spoken by human beings, and so the metaphor functions metonymically.
And, just as, globally, languages may ‘die out’ if the speakers die out, like
threatened species of plant and animal life, so dialects also are seen ecol-
ogically in a similar way.1 This ecological image finds a symbiosis most
aptly in dialect lexis. Just as many species of flora and fauna have disap-
peared or are under threat, so local words are no longer used or needed.
Even the common spuggy (‘sparrow’ in the North-east) is dwindling in
numbers.
Other common images relate dialects more directly to the landscape:
dialectologists are prone to talk of the ‘recession’ of features, particularly
in pronunciation and grammar, which, in this millennium certainly evokes
images of receding icebergs in the face of global warming. Most particu-
larly, they are also prone to talk of the ‘erosion’ of vocabulary, like the
erosion of soil or rock; or they speak of ‘attrition’ (literally, ‘rub away’).
Again, by a kind of metonymy, human beings live their lives in a regional
landscape, and their speech is identified with this: so Northerners’ accents
are commonly perceived as ‘gritty’ and ‘rough’ like the granite scenery and
harsh weather. Particular occupations also have close relations with the
local geography, like coal-mining, herring-fishing or cockle-picking. When
such industries disappear for cultural or environmental reasons, their par-
ticular dialect forms (especially lexis) are prone also to disappear, and this
is both patently observed and also negatively commented on. The danger is
that vernaculars are too closely identified with such unique activities and
their lexicons and that, in another metaphor, those bemoaning loss wish to
“preserve [them] in an invariant state of suspended animation” (Widdow-
son and Seidlhofer 2003: 308).2
Even the now common image of dialect ‘levelling’, used by social dia-
lectologists, and to which I shall return, which involves the eradication of
marked variations (L. Milroy 2003: 155) can evoke in the subconscious
ideas of the levelling of the landscape in our post-industrial era: the eradi-
cation of slag-heaps, or the bulldozing of artisan dwellings in industrial
cities to make way for a new set of uniform housing estates and shopping
malls. So the perceived ‘homogenisation’ of society is matched by the per-
ceived increased homogenisation in speech, particularly in grammar and
vocabulary, and particularly under the ‘standardising’ influence of Stan-
dard English.
52 Katie Wales
In the rest of this chapter I shall be continuing to interrogate some of

these common metaphors and other terms used in dialectology which pre-
suppose certain well-established models of language behaviour and
change. The aim is not necessarily to change the models, but to suggest
that metaphors can blind us to neglecting other vital areas of attention or to
overlooking other significant tendencies. The focal point for initiating dis-
cussion will be the only detailed set of predictions, and a corresponding
map, produced by a social dialectologist at the end of the twentieth cen-
tury, namely Peter Trudgill’s in his Dialects of England (1999). Trudgill’s
own bold attempt will thus act as a catalyst, so to speak, for renewed
speculations about the future of English dialects in the twenty-first century.
3. Trudgill’s future dialect areas
It has to be said at the outset, however, that Trudgill’s maps of past (‘tradi-
tional’), present (‘modern’) and ‘future dialect areas’ that punctuate his
text are in effect labelling accent areas rather than dialect. His main con-
cern throughout is segmental phonology. In one sense this certainly under-
lies the perception that regionality is primarily marked by accent. However,
we must not forget other significant markers of regional variation in accent
and speech such as prosody: remarkably understudied in dialectology, but
of great import to speakers and observers as a means of distinction of one
variety from another, and commonly labelled by such folk-labels as
‘twang’, ‘whine’, ‘sing-song’ or ‘tuneful’. The significant issue of spoken
grammar I shall return to below (Section 4), which has tended to be under-
estimated in contemporary language studies. The other point to make about
Trudgill’s predictions is that there is no explicit time-scale projected: per-
haps wisely.
There are actually thirteen dialect areas specifically labelled in Trud-
gill’s map, but what is very striking visually is the predominance of the
label ‘London’: comprising most of the East, the South-east and the South
of England, and part of the Central Midlands. It is not too far graphically
from his map of modern dialect areas, except that the label ‘Home Coun-
ties’ has been replaced by ‘London’, and the South Midlands and Suffolk
have disappeared. Visually, then, the map reflects what sociolinguists ar-
gue and what lay-people believe, namely that London English (now popu-
larly known as ‘Estuary English’) is the main source of linguistic ‘diffu-
sion’ or ‘spreading’, because of the capital’s social and cultural
dominance; and because of the massive growth of suburbs and dormitory-
towns along commuting routes with corresponding emigration of London

English speakers from the inner city districts. The result is the further lev-
elling of distinctive accents under the influence of London pronunciations
in the south and east. Viewed ecologically, and especially in the media,
London English is rather like a rampant weed or deadly parasite which is
destined to take over or choke to death many of England’s remaining ac-
cents. Again, the visual isolation on any map of Cornwall and Devon to the
south-west, reflecting their geographical distance from the metropolis,
likewise that of North Lancashire and the Lake District in the North-west,
and Norfolk in the East, suggests that these areas, remote from London,
will somehow escape potential ‘contamination’. And these are areas, of
course, as the metropolitan reviewer of Francis Grose revealed (section 1),
whose provincial inhabitants have for centuries been remarked upon for
their decidedly marked habits of speech. What is interesting, however, is
that in this centre versus periphery model or ‘gravity model’ as it is com-
monly termed, Trudgill’s Northern limit of ‘London’ appears to correspond
still to the Northern limit (from the south) of the traditional STRUT-FOOT
distinction, implying that for the foreseeable future at any rate this impor-
tant perceptual boundary will remain, to keep ‘Northerners’ and ‘Southern-
ers’ apart (see further Wales 2006a, chapters 1, 3 and 5).
Trudgill’s projection about the ‘spread’ of London English is presuma-
bly premised on the basic assumption that London will continue to be the
country’s centre of gravity for political, economic, cultural and hence lin-
guistic activity. Alexander Ellis in 1889 had certainly predicted that “Lon-
don and its environs, the south-east, will continue to be a source of linguis-
tic innovations” (cited Ihalainen 1994: 262). Historians, however, are well
aware that the national status quo is not eternal (see Musgrove 1990; Tay-
lor 1993); and we have only to think of the political and cultural signifi-
cance of Winchester before the Norman Conquest, and of York in the
North. The history of the country over the past 1500 years has revealed
periods of ‘centralisation’ and ‘decentralisation’. The recent moves to po-
litical ‘devolution’ in Wales and Scotland certainly stirred the North of
England into some consideration of it, a region which over the same long
period of time has always been strongly anti-metropolitan in many ways.
Moreover, a study of linguistic change reveals that processes of diffusion
have not all been out from London but also into it, from the periphery.
Since the medieval period London English has absorbed or assimilated
features from East Anglia, the central Midlands and even Northern Eng-
lish.
54 Katie Wales
In the immediate post-Second World War period, however, linguists

might not have predicted that popular London English, or ‘Estuary Eng-
lish’ (Rosewarne 1994), rather than educated London English or Received
Pronunciation, the traditional ‘enemy’ of dialectal variation, would be con-
sidered a source of innovation and change. By 1982 Wells was able to state
most sweepingly that London’s “working-class accent is today the most
influential source of phonological innovation in England and perhaps in the
whole English-speaking world” (1982: 301). However, Ellis again, in 1889,
predicted that certain Cockney features would work their way up in social
status and become ‘polite’ pronunciations in another one hundred years;
although tike for ‘take’ and bowt for ‘bought’ have not yet done so (cited
Ihalainen 1994: 262). At the turn of the twentieth century, again, a member
of the English Association, S. K. Ratcliffe, talked of the English language
‘going to pieces’ especially under the influence of the ‘debased’ dialect of
the Cockney, “which is spreading from our schools and training-colleges
all over the country” (cited Doyle 1989: 36).
However, we cannot always assume that Estuary English is necessarily
the source, or only source, of contemporary sound changes that appear to
mark diverse urban pronunciations from Bristol to Derby, Middlesbrough
to Glasgow. The ‘diffusion’ model may be too weak in its explanatory
power on its own, especially in relation to the actual mechanisms of the
alleged spreading of certain features of pronunciation especially amongst
the younger generations. Commonly cited, for example, by sociolinguists
as being due to the influence of Southern or Estuary English are T-
glottalling, H-dropping, and TH-fronting (as in bovver ‘bother’ and fing
‘thing’): see, e.g. Kerswill (2003), and Kerswill and Williams (2002), who
state quite categorically that in general “consonantal features spread from
the South to the North” (2002: 96). But there is strong evidence for such
features as TH-fronting (and also T-glottalling) in nineteenth-century dia-
lects well away from the London area (see Milroy, Milroy and Hartley
1994; Upton 2006: 306–307; Wales 2006a: 176–177). I myself would pre-
fer therefore an alternative model, and if a naturalistic analogy must be
made, that of ‘latency’ or ‘dormancy’, akin to Chambers’ (2004) idea of
‘vernacular universals’: such variants prone to appearing independently at
different times in different localities, but probably often suppressed be-
cause of their social stigma. At the present time it certainly looks as if they
are being stimulated or ‘revived’ in urban communities under the influence
of their popularity in street-wise London English and / or the yoof [sic]
media, so the whole process can be seen as a reflex of an internal develop-
ment and ‘borrowing’. So as J. Milroy (2003: 217) concludes, who quite
coincidentally seems to have arrived at a similar kind of model, there is

absolutely no guarantee that one specific community such as London is the
original source for such changes. There is also a striking parallel to be
drawn with the case-history of L-vocalisation, as in miuk for ‘milk’, also
seen as a relatively recent change, and centred upon popular London Eng-
lish. But as Ihalainen (1994: 265) points out, “it appears to have occurred
in several dialects and at different times in the history of English”: he cites
early examples from the ‘North-Country’ section of John Ray’s dictionary
(1674).
To return to Trudgill’s further conjectures, five of his areas are labelled
according to major cities: Newcastle, Liverpool, Manchester, Birmingham
and Bristol, a reflection of the inheritance of dramatic nineteenth-century
and twentieth-century industrial developments and mass immigration par-
ticularly in the North and the Midlands. “We can predict”, he says
(Trudgill 1999: 83), “the survival of the Northeast area focussed on New-
castle, Merseyside focussed on Liverpool … and the Northwest Midlands
focussed on Manchester as distinctive areas”; and that, generally, future
dialect areas are likely to be “city-based” (Trudgill 1999: 84). Certainly
Trudgill seems to be underlining a very significant development in the
second half of the twentieth century, namely that conurbations such as
Liverpool, Manchester and Newcastle have become the locus for what are
termed ‘mainstream dialects’, ‘supra-regional norms’ or ‘regiolects’: ‘lev-
elling’ occurring within a region, between the city and hinterland, but not
necessarily, in the latter part of the twentieth century, under the direct in-
fluence of the exo-normative Received Pronunciation (or even of ‘Estuary
English’), certainly amongst the working classes. Påhlsson (1972) pre-
dicted this trend over thirty years ago:
In my view it may be argued that the position and influence of RP today is
likely to be less prominent and operative than it used to be … It seems not
unlikely that local regional standards will exert a relatively greater influence
in the future, as melting-pots … . (Påhlsson 1972: 27)
So the region labelled ‘Newcastle’ extends northwards to the Scottish bor-
der, and southwards to the River Tees, and so would include the conurba-
tion of Middlesbrough. Indeed, ‘Geordie’ today, as the dialect of Newcas-
tle, Tyneside and (more loosely) the Northeast is known, exerts
considerable covert prestige. The research of Llamas (1998, 2000) has
shown how the younger inhabitants of Middlesbrough identify themselves,
not with the historical Yorkshire, but with the Northeast, and this is re-
56 Katie Wales
flected in their accent, with the emergence of a High-Rise Tone, as well as

glottalisation (but see also section 4 below).
Interestingly, while it is tempting to assume that a big city will inevita-
bly act as the source of diffusion into the smaller rural neighbourhoods,
this is not always the case. Innovations amongst the urban working class
youngsters with close ties and little occupational mobility can be drawn
from the relatively ‘conservative’ vernacular of the hinterland, whether in
phonology or lexis, and so there is a ‘pulling inwards’, so to speak, as well
as a ‘pushing outwards’, towards the local community norms. A rounded
vowel in poke, stone in Newcastle speech may have come from the coastal
area, and the intensifier geet (Beal 1993: 208). Charver (‘townie’) and
gadgie (‘old man’) have come into Newcastle from the border region and
local Romany dialects. A reduction in regionally marked forms therefore
does not necessarily happen. To echo Påhlsson’s phrase (1972), the regio-
lect is a rich ‘melting-pot’ for the continued diversification of features; and
for innovations even on the lexical level. (See also Upton 2006: 324;
J. Widdowson 2005: 13, former proponents of the ‘erosion’ model.)
Given the existence of significant urban varieties in England today, we
might be surprised, however, that Trudgill has not identified further areas
in this way. There is no Hull or Humberside (the latter marked on his
‘modern’ dialect map); no ‘East Midland’ urban varieties such as Notting-
ham or Derby; and no Yorkshire urban varieties such as Leeds, Bradford or
Sheffield: they are subsumed under the label ‘S.Yorks’ or South Yorkshire
(as distinct from ‘N.Yorks’ or North Yorkshire). Presumably, their hinter-
lands are not perceived even in the future to be as large as that of Liverpool
or Manchester, for example, but this is surprising, especially for Humber-
side. However, let us now project a different future for urban variation in
Yorkshire, in section 4.
4. Dialect, identity and difference
A possible foreshadowing of Trudgill’s own conjecture is to be found a

hundred years ago. The editor of the Barnsley-based Pogmoor Olmenack
‘Tom Treddlehoyle’ in 1906 arbitrates over a dispute between ‘Dame Shef-
field’ and ‘Mother Leeds’ as to which is the leading Yorkshire city. Tom
predicts that their rapid urban spread will result in the settling of differ-
ences in a merged great metropolis to rival Manchester, Liverpool or even
London (cited Russell 2004: 127). However, on the face of it, Liverpool
and Manchester are themselves poised to be conjoined, at either end (once)
of the great Ship Canal built in the 1890s to facilitate Manchester’s cotton
exports. Yet their industrial history is one of rivalry, and rivalry exists to
this day, if manifested more specifically in football teams and teen-age
fashions. We should not be surprised, therefore, that there is continued
pressure for the Liverpudlians and Mancunians to maintain their distinct-
iveness in their speech habits, despite the physical blurring of their subur-
ban boundaries, and despite, historically, the Lancashire base to both of
their dialects. The accents of both cities, for instance, share the [g]-
articulation in nasal groups of words like long. Indeed, recent research
reported in the national press early in 2006 suggested that the accents of
Liverpudlians or ‘Scousers’ were getting stronger precisely because they
are determined to sound distinctive. This may also account for the mainte-
nance of the characteristic aspirated plosives. So much for the Scouse ac-
cent ‘sinking into the Estuary’ (section 1); and the general view expressed
by Kerswill (2003: 239) that “few researchers have been able to demon-
strate the opposite [to levelling, namely] divergence or diversification in
local varieties”. It seems extremely likely that urban varieties all over the
country will therefore maintain rather than lose their distinctiveness, and so
Sheffield speech remain distinctive from that of other Yorkshire cities.
A similar cultural rivalry to that of Liverpool and Manchester certainly
exists between Newcastle on the Tyne and Sunderland on the Wear, even
though both cities, once in Northumberland and Durham respectively, are
now part of the Tyne and Wear civil administration. (See Townsend and
Taylor 1975 for a survey of local antipathies and loyalties in the 1970s.) So
while the inhabitants of both cities may well be contributing to a regional
norm of ‘Newcastle’ speech in the future, as predicted by Trudgill, none-
theless they are acutely aware of linguistic differences between them: to
the extent that the local nickname for Sunderlanders is Mackems, reflecting
the distinctive Wearside pronunciation of verbs like make and take (and
not Newcastle ma-ek, ta-ek) (see also below and further Beal 2000). Inter-
estingly, one of the reasons why the Government’s Referendum on a
Northeast Regional Assembly failed a couple of years ago, with only 22%
in favour of this kind of ‘devolution’, was precisely that people in Durham
(also on the river Wear) resented the idea of Newcastle’s political domi-
nance. As a headline in the Independent so tellingly summarised it, “Sup-
port from Geordie icons only served to whip up tribal conflicts” ( 6th No-
vember 2004). Even Llamas’s young Middlesbrough informants, noted in
section 3 above, showed a hostility to the city of Newcastle, preferring a
more general identity with ‘Northeast’. It is not too fanciful, therefore, to
imagine a future sub-variety of ‘Northeast’ English centred upon Middles-
58 Katie Wales
brough itself, and clearly distinctive also from any North Yorkshire speech.
Already, for instance, Beal (personal communication) notes the NURSE-
SQUARE merger which distinguishes Teesside speech from the rest of the
Northeast as well as North Yorkshire; as indeed Hull from the rest of York-
shire, and Liverpool from the rest of Lancashire. Similarly, for the un-
stressed vowel at the end of HAPPY words, the tense or lax realisations act
as markers of difference between regions.
It is worth reminding ourselves therefore of the commonplace that dia-
lect and accent not only signal a bonding with others who speak ‘the same’,
share the same group identity, but also serve to mark one group from an-
other, along with other social symbols:
The individual creates for himself the patterns of his linguistic behaviour so
as to resemble those of the group or groups with which from time to time he
wishes to be identified, or so as to be unlike those from whom he wishes to
be distinguished. (Le Page and Tabouret-Keller 1985: 181)
As J. Milroy (1993: 219) succinctly states: “To put it crudely, some groups
in society may not particularly want to be understood by other groups” (see
also Wales 2006b). Cameron (1995: 9) argues that the potential for nor-
malcy “is latent in every communicative act, and the impulse behind it
pervades our habits of thought and behaviour”, but the impulse for differ-
entiation is probably stronger. Regional features are both salient markers of
difference and markers of a ‘secret code’. The ‘tribal’ instinct both to con-
gregate and to delimit territory, so deeply rooted in England since Anglo-
Saxon times, still survives in some manifestation or another, including
speech. And while we might expect the publications of the English Dialect
Society in the nineteenth century to reveal and comment upon the distinc-
tion of speech habits in communities from neighbouring dale to dale, and
also expect social commentators like Hoggart (2003) to indicate variation
from district to district in cities like Leeds in the 1920s, and between Shef-
field and Leeds, we might not expect such nuances, or even perception of
such nuances, at the turn of the millennium, in this age of increased mobil-
ity and looser social network ties. But they are indeed there. As Stanley
Ellis concluded in 1992, one of the key field-workers on Harold Orton’s
Survey of English Dialects published 30 years before, “Something still very
local keeps going, something to differentiate one place from another”
(1961–72: 9; his italics).
This has been most strikingly revealed in the nationwide BBC dialect
survey (2004–5) published by Simon Elmes (2005). Time and again the
survey found that villagers and townspeople all over the country were still
acutely aware of differences in speech ‘just down the road’, in accent,

grammar and vocabulary: whether between rural Norfolk and Suffolk,
Birkenhead and Liverpool, and even North and South Liverpool. More-
over, despite the diffusion of ‘Estuary English’ and the blanketing, so to
speak, of much of the South of England under ‘London English’ generally,
Elmes’s interviewees were able to distinguish accents ‘north’ and ‘south’
of the Thames, and between inner and outer suburbs of London.3 Often the
differences were prosodic: subtle distinctions in intonation, voice quality,
stress or rate of delivery – features still strikingly under-researched by dia-
lectologists. The BBC survey actually corroborates research such as Wil-
liams and Kerswill (1999) on accent differences between adolescents in
Milton Keynes and Reading; and Britain (2003) on three distinct speech
communities in the eastern, central and western Fens in the south Mid-
lands.
The perception of linguistic difference between communities is interest-
ingly paralleled by the folklinguistic habit of blazon populaire, which
seems to flourish today as much as it did in earlier centuries: namely, the
labelling of rival towns and villages, particularly at sporting events such as
football matches. Just as Mackem picks out a salient feature of pronuncia-
tion, so does Dee-dah, the pejorative label for a Sheffield inhabitant (the
traditional user of the second person pronoun thou). Wigan football fans
are labelled ‘pie-eaters’, those from Hartlepool ‘Monkey-hangers’ (recall-
ing an historical event from the Napoleonic Wars!); and all Teessiders are
smoggies (because of local air pollution). Ifracombe folk in North Devon
still look down on the villagers of Combe Martin five miles to the west as
Shammicks, or slum-dwellers, who once smelted silver in the late eight-
eenth century. The deep-rooted hostility between town and countryside is
reflected in the label children-munchers given by Carlislers to the Cumbri-
ans generally; and carrot-crunchers to the fennies of East Anglia by ‘town-
ies’. The traditional cultural link between local region and local food is
reflected in labels such as Cornish pasties, Lancashire hotpots, tiddy-
oggies (Plymouth; naval slang for ‘pasty’, Elmes 2005: 46), and most pro-
bably the well-known Scouse(r), once a kind of stew (lobscouse; note also
lobbygobblers for Leigh in Lancashire). A recent formation in Liverpool is
plastic Scouser, for those unfortunate to live in the posher suburbs of the
Cheshire hinterland (Chris Montgomery, personal communication). While
many of such nicknames are clearly pejorative in origin when used by ‘out-
siders’, some are appropriated by ‘insiders’ as proud emblems of their
identity. They are as much an illustration of what Russell (2004: 278)
60 Katie Wales
terms our ancient ‘territorial mentality’ as the more widespread regional

nicknames such as Lanky, (Yorkshire) tyke, Geordie and Cockney.
It is arguable that such overt features as nicknames, jokes and stereo-
types are as salient markers of local or community identity as dialect fea-
tures themselves; moreover, that both blazon populaire and linguistic
variation have a very clear function in contemporary society in the process
of what Townsend and Taylor thirty years ago termed ‘image formation’.
This has as its basis a set of common attributes perceived as distinctive to
the region, and, most significantly, “is a historical process of a dynamic
and continuing character” (1975: 387). At one extreme it might be said that
regional Tourist Boards actively promote image formation, and even the
commodification of regional culture, in the promotion of symbolic com-
munities comprising seasonal festivals and rituals, local cuisine (stotty
cake, singing hinnies, Cornish pasties), as well as tea-towels and mugs
emblazoned with dialect sayings.
At another extreme, the advent of the internet has brought a new kind of
textual community (Duncan and Duncan 1988) to replace or supplement
the older or obsolescent community centred upon the ‘shared readings’, in
their terms, of dialect literature and folk-song which have traditionally
played a significant role in the cultural formation of the English regions.
So Teessiders, at home or abroad, can log on to www.smoggy.org.uk; and
Common Ground tries to involve people in taking responsibility for their
own locality and its distinctiveness (www.commonground.org.uk). The
‘virtual communities’ of the internet chatrooms and focus groups both sup-
plement the initiatives of the traditional metadiscursive communities like
dialect societies and also function more proactively and effectively than
local radio or television. As Crystal (2006: 404) states, the internet offers
“an unprecedented degree of … public presence to small-scale regional and
social groups, and thus a vast potential for representing local identities”.
One general conclusion, therefore, might be that regional variation not only
helps in the construction of regional identity and image, but, conversely,
the latter, in whatever form, will be important for the maintenance of re-
gional variation in the future.
Significant also in this new millennium for image formation and the
construction of regional identity is likely to be the performance of dialect
at the individual rather than the group level; and I use the word ‘perform-
ance’ deliberately. While code- or style-switching has been much studied
in the context of bilingualism (and see Crystal 2000), more research needs
to be done on bi-dialectalism, and in particular the conscious stylised use
of dialect and accent across the classes for a variety of functions as part of
a public ‘repertoire’, or the projection of an image or persona. (See also

Coupland 1988: 139; Rampton 1999). Elmes (2005: 17) was certainly
made aware of a “whole portfolio of styles and variants” used by young
people nationwide. We need to see dialect therefore, and indeed the speech
community itself, in a more flexible and multi-dimensional way, relating
degrees of ‘broadness’ not simply to status or education as in the past, but
to a strategy of identity construction which arises in shifting day-to-day
interactions or experiences. On a microcosmic level, this could well in-
volve a degree of variability in the use of dialect features from utterance to
utterance. As Chambers (2002: 351) succinctly states, membership of a
speech community is determined “by consensus about community norms
but not by conformity in their use”.
In such a context it is also likely that regional grammar will flourish,
summarily labelled ‘non-standard’, and traditionally stigmatised in com-
parison with standard grammar (and so often wrongly equated with ‘sub-
standard’). It has always been the case that in informal spoken registers
specific features local to an area will be heard; and, in any case, there is
much more distinctive syntactic and morphological variation in English
speech generally than is commonly recognised by grammarians or has been
tabulated by dialectologists: e.g. in modality, adverbials and pronouns.
There are also very distinctive discourse markers, speech tags, terms of
address, greetings and leave-taking signals, etc.: features which hardly ever
find their way into print, because of their phatic and community functions.
I would also go further and predict (along with Crystal 2006) that in the
general prevailing climate of informalisation in speech and in the new
oracy of electronic communication, the hegemony of ‘standardness’ (and
prescriptivism with it) is likely to be dismantled to an extent not seen since
its inception in the Renaissance. As Crystal states (2006: 404), the internet
is not controlled by the traditional custodians of ‘standardness’, such as
grammarians, lexicographers or publishers. Standard English could be but
a phase in the ongoing history of English.4
5. Conclusion: Convergence or divergence?
Despite reiterated fears during the twentieth century, that the new secon-
dary forms of oracy such as radio, film and television would ‘kill off’ re-
gional diversity in speech, this has not yet happened, neither on a local nor
on a global level. Even on this scale, notions of both ecology and imperial-
ism or hegemony are complex (see further Brutt-Griffler 2002), and the
62 Katie Wales
microcosm of internal regional variation reflects this. Regional ‘norms’ do

not necessarily mean homogenisation or fixity, any more than ‘globalisa-
tion’ does; but more research needs to be done on how local patterns and
tendencies map on to global. Local cultures are constantly changing, and
the result is a continuous process of ‘hybridisation’, as cultural geogra-
phers argue (see Johnston 1991: 50). It is also time then for sociolinguists
and dialectologists to engage seriously with the work of such disciplines as
cultural geography, and also linguistic and social anthropology. So Cohen
(1985: 36) has argued that the creativity of speakers should not be underes-
timated, and that the monolithic conformity of English is not inevitable,
and is simply a ‘myth’. In his own image, the homogeneity of speech is
merely a ‘veneer’ (1985: 44). I conclude with a vivid variant on this image
from the decorative arts: to assume that regional speech is becoming more
uniform, says Hoggart (2003: 10–11), would be “like plonking obliterating
paint over a multi-coloured if faded fresco”.
Notes
1. As Crystal points out in his review of Dalby (2002), ‘loss’ is probably a better
word than ‘death’ because usually people don’t die but switch to other lan-
guages (THES 14.06.02). But note the title of his own book (2000).
2. Widdowson and Seidlhofer’s (2003) discussion arises in the context of
Trudgill’s suggestion (2002) that dialectologists should take some responsibil-
ity for the ‘preservation’ of dialects. In widening their argument to language
intervention generally, they unfairly see Trudgill’s remarks as expressing the
same interest as writers like Sprat and Orwell in a “virtuous vernacular sim-
plicity” (2003: 299–300).
3. This tabulated regional variety struck a new chord with journalists. In August
2005 the Sunday Times proclaimed: “Odd a thar’t it? British dialects get
stronger” (cited Dent 2006: 135). So much for the devouring monster of Estu-
ary English.
4. In the post-nuclear holocaust of the twenty-fourth century that is the setting
for Russell Hoban’s dystopia Riddley Walker (1980), Standard English is no
more; in its place a demotic speech and semi-phonetic spelling.
References
Addy, Sidney
1888 A Glossary of Words Used in the Neighbourhood of Sheffield. Lon-
don: English Dialect Society.
Bailey, Richard
1996 Nineteenth-century English. Ann Arbor: University of Michigan
Press.
Beal, Joan
1993 The grammar of Tyneside and Northumbrian English. In Real Eng-
lish, James Milroy and Lesley Milroy (eds.), 187–213. London:
Longman.
Beal, Joan
2000 From Geordie Ridley to Viz: Popular literature in Tyneside English.
Language and Literature 9: 359–375.
Britain, David
2003 Exploring the importance of the outlier in sociolinguistic dialectol-
ogy. In Social Dialectology: In Honour of Peter Trudgill, David
Britain and Jenny Cheshire (eds.), 191–208. Amsterdam: John Ben-
jamins.
Brutt-Griffler, Janine
2002 World English: A Study of its Development. Clevedon: Multilingual
Matters.
Burchfield, Robert
1994 Introduction. In Cambridge History of the English Language, vol. 5,
Robert Burchfield (ed.), 1–19. Cambridge: Cambridge University
Press.
Cameron, Deborah
1995 Verbal Hygiene. London: Routledge.
Chambers, Jack
2002 Patterns of variation including change. In Handbook of Language
Variation and Change, Jack Chambers, Peter Trudgill and Natalie
Schilling-Estes (eds.), 349–372. Oxford: Blackwell.
2004 Dynamic typology and vernacular universals. In Dialectology Meets
Typology, Bernd Kortmann (ed.), 127–145. Berlin: Mouton de
Gruyter.
Cohen, Anthony Paul
1985 The Symbolic Construction of Community. London: Tavistock Publi-
cations.
Collier, John of Urmston
[1746] 1775 The Miscellaneous Works of Tim Bobbin Esq., Containing his
View of the Lancashire Dialect. Manchester: for the author and Mr
Haslingden.
64 Katie Wales
Coupland, Nikolas
1988 Dialect in Use: Sociolinguistic Variation in Cardiff English. Cardiff:
Crystal, David
1997 English as a Global Language. Cambridge: Cambridge University
Press.
2000 Language Death. Cambridge: Cambridge University Press.
2006 Into the twenty-first century. In Oxford History of the English Lan-
guage, Lynda Mugglestone (ed.), 394–413. Oxford: Oxford Univer-
sity Press.
Dalby, Andrew
2002 Language in Danger: How Language Loss Threatens our Future.
London: Penguin.
Daniel, Samuel
[1599] 1965 Musophilus. West laFayette, Indiana: Purdue University Studies.
Dent, Susie
2006 The Like, Language Report for Real. Oxford: Oxford University
Press.
Doyle, Brian
1989 English and Englishness. London: Routledge.
Duncan, J. and Duncan, N.
1988 (Re)reading the landscape. Society and Space 6 (2): 119–26.
Ellis, Alexander
1888 On Early English Pronunciation, Part V. London; Philological Soci-
ety / Asher and Co.
Ellis, Stanley
1992 Forty years on: is dialect dead? Transactions of the Yorkshire Dialect
Society. Part 92 vol.17: 6–16.
Elmes, Simon
2005 Talking for Britain: A Journey through the Nation’s Dialects. Lon-
don: Penguin.
Graddol, David
1997 The Future of English. London: Open University/ British Council.
Griffiths, Bill
1999 Retaking the language: The need for dialect initiatives. Northern
Review 8 (2): 126–132.
Grose, Francis
[1787] 1968 A Provincial Glossary. Menston: Scolar Press reprint no. 63.
Heslop, Rev. Richard
1892 Northumberland Words. London: English Dialect Society.
Hoban, Russell
1980 Riddley Walker. London: Cape.
Hoggart, Richard
[1957] 1976 The Uses of Literacy. London: Pelican.
2003 Everyday Language and Everyday Life. New Brunswick, NJ: Trans-
action Publications.
Holloway, William
1839 A General Dictionary of Provincialisms. Sussex, Lewes: Baxter and
Son.
Ihalainen, Ossi
1994 The dialects of England since 1776. In Cambridge History of Eng-
lish, vol. 5, Robert Burchfield (ed.), 197–274. Cambridge: Cam-
Johnston, R. J.
1991 A Question of Place: Exploring the Practice of Human Geography.
Oxford; Blackwell.
Kerswill, Paul
2003 Models of linguistic change and diffusion: New evidence from dia-
lect levelling in British English. In Social Dialectology, David Brit-
ain and Jenny Cheshire (eds.), 223–243. Amsterdam: John Benja-
mins.
Kerswill, Paul and Ann Williams
2002 ‘Salience’ as an explanatory factor in language change: evidence
from dialect levelling in urban England’. In Language Change: The
Interplay of Internal, External and Extra-linguistic Factors, Mari
Jones and Edith Esch (eds.), 81–110. Berlin: Mouton de Gruyter.
Le Page, Robert and A. Tabouret-Keller
1985 Acts of Identity: Creole-based Approaches to Language and Ethnic-
ity. Cambridge: Cambridge University Press.
Llamas, Carmen
1998 Language variation and innovation in Middlesbrough: A pilot study.
Leeds Working Papers in Linguistics and Phonetics 6: 97–114.
2000 Middlesbrough English: Convergent and divergent trends in a part of
England with no identity. Leeds Working Papers in Linguistics and
Phonetics 8: 123–48.
McCrum, Robert, William Cran and Robert Macneil
1992 The Story of English (New and Revised Edition). London: Faber and
Faber / BBC.
Milroy, James
1993 On the social origins of language changes. In Historical Linguistics:
Problems and Perspectives, Charles Jones (ed.), 215–236. London:
Longman.
2003 When is a sound change? On the role of external factors in language
change. In Social Dialectology, David Britain and Jenny Cheshire
(eds.), 209–221. Amsterdam: John Benjamins.
66 Katie Wales
Milroy, James Lesley Milroy and Sue Hartley

1994 Local and supra-local change in British English: The case of glottali-
sation. English World-wide 15 (1): 1–33.
Milroy, Lesley
2003 Social and linguistic dimensions of phonological change. In Social
Dialectology, David Britain and Jenny Cheshire (eds.), 155–171.
Amsterdam: John Benjamins.
Mugglestone, Lynda
2006 English in the nineteenth century. In Oxford History of English Lan-
guage, Lynda Mugglestone (ed.), 274–304. Oxford: Oxford Univer-
sity Press.
Musgrove, Frank
1990 The North of England: A History from Roman Times to the Present.
Oxford: Blackwell.
Nicholson, J.
1889 Folk Speech of East Yorkshire. London: Simpkin, Marshall and Co.
Orton, Harold et al.
1961–1972 Survey of English Dialects. Leeds: E.J.Arnold
Påhlsson, Christer
1972 The Northumbrian Burr. Lund: CWK Leerup.
Quirk, Randolph
1972 The English Language and Images of Matter. Oxford: Oxford Uni-
versity Press.
Rampton, Ben
1999 Crossing. Journal of Linguistic Anthropology 9 (1–2): 54–56.
Ray, John
[1674] 1969 A Collection of English Words. Menston: Scolar Press reprints
no. 145.
Rhodes, Barrie
2000 Changes in knowledge and use of non-standard words over the last
80 years-research in West Yorkshire. Transactions of the Yorkshire
Dialect Society, Part 100, vol.19: 11–26.
Rosewarne, David
1994 Estuary English- tomorrow’s RP? English Today 37 (10): 3–8.
Russell, Dave
2004 Looking North: Northern England and the National Imagination.
Manchester: Manchester University Press.
Simmelbauer, Andrea
2000 The Dialect of Northumberland: A Lexical Investigation. Heidelberg:
Carl Winter.
Taylor, Peter
1993 The meaning of the North: England’s foreign country within. Politi-
cal Geography 12 (2): 136–155.
Townsend, A. R. and C. C. Taylor

1975 Regional culture and identity in industrial societies: The case of
North-east England. Regional Studies 9 (4): 379–393.
Trudgill, Peter
1998 Standard English: What it isn’t. European English Messenger 7 (1):
35–39. Reprinted in (1999) Standard English: The Widening Debate,
Tony Bex and Richard J. Watts (eds.), 117–128. London: Routledge.
1999 The Dialects of England. [2nd edition] Oxford: Blackwell.
2002 Sociolinguistic Variation and Change. Edinburgh: Edinburgh Uni-
versity Press.
Upton, Clive
2006 Modern regional English in the British Isles. In Oxford History of
English, Lynda Mugglestone (ed.), 305–333. Oxford: Oxford Uni-
versity Press.
Wales, Katie
2006a Northern English: A Social and Cultural History. Cambridge: Cam-
2006b Dialects in mental contact: A critique of perceptual dialectology. In
The Wider Scope of English, Herbert Grabes and Wolfgang Viereck
(eds.), 57–66. Frankfurt: Peter Lang.
Webster, Noah
1789 Dissertations on the English Language. Boston.
Wells, John
1982 Accents of English. Cambridge: Cambridge University Press.
Widdowson, Henry and Barbara Seidlhofer
2003 The virtue of the vernacular: On intervention in linguistic affairs. In
Social Dialectology, David Britain and Jenny Cheshire (eds.), 299–
309. Amsterdam: John Benjamins.
Widdowson, John
1993 English dialects: Our living heritage. Transactions of the Yorkshire
Dialect Society, Part 93 vol. 18: 6–16.
1999 Sheffield dialect on the eve of the millennium. Transactions of the
Yorkshire Dialect Society, Part 99 vol. 19: 9–19.
2005 Time to move on: Changing perceptions of English regional dialects
in the twenty-first century. Transactions of the Yorkshire Dialect So-
ciety Part 105 vol. 21: 9–16.
Williams, Ann and Paul Kerswill
1999 Dialect levelling: Change and continuity in Milton Keynes, Reading
and Hull. In Urban Voices, Paul Foulkes and Gerard Docherty (eds.),
141–161. London: Arnold.
Wright, Joseph
1905 The English Dialect Grammar. Oxford: Clarendon Press.
Chapter 4
The role of dialect contact in the formation of
Englishes
Peter Trudgill
1. Introduction
Most histories of English note the role of language contact in the history of
the language. Some of the following languages generally get at least a men-
tion as having been in contact with English in the British Isles: Welsh, or
its ancestor, on the first arrival of the Anglo-Saxons in Britain; Old Norse,
as a result of Viking colonisation; Norman French, after 1066; and Scottish
Gaelic and Irish Gaelic.
As for colonial varieties of English, they have experienced forms of lan-
guage contact with indigenous languages, which seems to manifest itself
mostly in terms of the acquisition of loanwords from these languages.
South African English, for example, has forms such as karroo ‘desert pla-
teau’ and dagga ‘wild hemp’ from Khoi, and bundu ‘rough country’ and
impala ‘type of antelope’ from Bantu languages (Branford 1994). Canadian
English has borrowed saskatoon and muskeg from Cree (Brinton and Fee
2001); North American English generally has borrowed lexical items such
as skunk and caucus from Native American languages (Romaine 2001),
and New Zealand English has borrowed mana ‘prestige, charisma, author-
ity’ from Maori.
Colonial Englishes have also experienced forms of language contact
with other European languages in the colonial situation (Romaine 2001).
Much of this also involves lexical borrowing. For example, Canadian Eng-
lish has borrowed coulee from French (Brinton and Fee 2001); North
American English has borrowed lexical items such as cookie ‘biscuit’ from
Dutch and key ‘islet’ from Spanish; while South African English has bor-
rowed veld ‘flat, open country’ from Dutch/Afrikaans. We can also sup-
pose that North American grammatical constructions such as Are you com-
ing with? and I like to skate (as opposed to I like skating) are the result of
German and/or Yiddish influence (see Trudgill 1986; Mufwene 2001:
162).
70 Peter Trudgill
In this paper I want to focus my attention on another type of contact,

dialect contact, in order to suggest that in a number of respects this is of
equal importance in the history of the language. This is because dialect
contact has more often than not led to dialect mixture and to what I have
called new-dialect formation (Trudgill 1986, 2004).
Ever since its inception, the English language has experienced a more
or less continuous process of geographical expansion, sometimes slower,
sometime faster, as a result of colonisation and settlement in new areas by
its speakers. In nearly all cases, this expansion has been accompanied in
the newly settled area by a mixing of English speakers from different re-
gions of the homeland, wherever that was. In very many recent cases we
have documentary historical evidence showing that this is so.
For example, Branford (1994: 487) supports a dialect contact origin for
South African English, citing the fact that the initial English-speaking im-
migrants to South Africa came from London, Ireland, Lancashire, York-
shire and Scotland. Turner (1994: 278), too, writes that early Australia
“was a situation of ‘dialects in contact’”.
However, as has been pointed out by a number of writers, it does not
necessarily follow that dialect contact would have led to dialect mixture. It
is perfectly possible that in the new location a single dialect prevailed and
no mixture resulted form the contact. Some writers have rejected the im-
portance of dialect mixture in the development of colonial dialects alto-
gether and plumped for a monogenetic origin for particular colonial varie-
ties. Hammarström (1980), for example, claims that Australian English is
simply transplanted Cockney. The same Cockney-origin point is argued for
by Wall (1938) for New Zealand English. Bauer (1994: 391) also argues
that “it is clear that New Zealand English derives from a variety of English
spoken in the south-east of England”. He goes on to argue further that New
Zealand English arrived as a ready-formed entity transplanted in its entirety
from Australia: “the hypothesis that New Zealand English is derived from
Australian English is the one which explains most about the linguistic
situation in New Zealand” (Bauer 1994: 428). These two points are of
course entirely consistent if one accepts that Australian English, too, as
Hammarström (1980) believes, arrived from the south-east of England.
And Lass (1990) argues that South African English is essentially southeast
of England in origin.
The same point has been made concerning colonial varieties of other
languages. For example, Rivard (1914) suggested that Quebec French is
entirely due to the dialects of Normandy. Lipski (1994: 36) similarly out-
lines a theory which developed amongst hispanicists that Latin American
The role of dialect contact in the formation of Englishes 71
Spanish was in origin basically a form of transported Andalusian Spanish

from the Iberian Peninsula. This andalucista theory was proposed early on
by Wagner (1920), amongst others.
If we cannot assume the dialect contact inevitably leads to dialect mix-
ture, we then have to argue for the importance of the dialect contact/dialect
mixture/new-dialect formation scenario, with particular respect to the de-
velopment of different varieties of English, on the basis of the linguistic
evidence. I have argued, in connection with the development of New Zea-
land English, that one cannot use socio-historical demographic information
to support linguistic hypotheses in the face of linguistic evidence to the
contrary. I agree with McWhorter (2000: 165) when he says, of historical
work on creole language origins, that: “Many creolists tend to see the so-
ciohistorical evidence as decisive, and as a welcome improvement upon
linguistic evidence considered merely tangential, capable only of ‘sugges-
tion’. In fact, one can argue that things are precisely the other way around.”
There are two types of linguistic evidence, I would suggest, that we can
employ. The first, which is easier to apply in the case of more recent stages
of the language, is to compare the linguistic characteristics of a new colo-
nial variety with those of the homeland and see if it is necessary to cite
sources for different individual features in different dialects in the home-
land. If features cannot all be given an origin in a single homeland dialect,
then mixture is the obvious explanation, although one does have to be care-
ful to allow for subsequent innovations (Trudgill 2004).
This is the approach adopted by Lanham (1967: 104), who writes of
South African English that it is derived from “at least 20 regional (geo-
graphical) dialects”, and says that “out of a welter of English dialects there
grew up in a remarkably short space of time a form of English which was
not identical with any one of them but presented a unique set of dialectal
features deriving probably from several British dialects”. For New Zea-
land, Bauer (2000) supplies very strong lexical evidence in favour of the
mixed origins of its English. His paper shows that typical New Zealand
English vocabulary items can be traced back to many different dialect loca-
tions in England and Scotland. This is the view also adopted in Trudgill
(2004) where I argue that early New Zealand English is for the most part a
mixture of features from different parts of the British Isles. And Schreier
(2003) discusses the origins in many different British and American dia-
lects of the features of Tristan da Cunha English.
Much of the literature on colonial varieties of languages other than Eng-
lish also takes this approach. An early proponent was Frings (1957), who
described the eastward colonial expansion of German into formerly Slavic-
72 Peter Trudgill
speaking areas as resulting in a new variety which was the result of a mix-
ture of Dutch, Low German, Central German and Upper German dialect
forms. This was later to provide the basis for Standard German, which
shows a mixture of forms such as the southern pronominal distinction be-
tween accusative and dative forms e.g. mich ‘me (acc.)’ and mir ‘(to) me
(dat.)’, which is absent from northern German, and the central diminutive –
chen rather than northern –kin or southern –lein.
Moag (1977) shows that Fiji Hindi is the result, amongst other things,
of a mixing of north Indian varieties. Combrink (1978: 72) suggests that
Afrikaans is in part the result of “Germanic dialects in close contact”, with
these dialects – 73% of the immigrants – coming from the dialect areas of
North Holland, South Holland, and Utrecht. And dialect contact was also
clearly important in the development of Canadian French (Morin 1994).
Mougeon and Beniak (1994: 26) say that:
the particular conditions associated with emigration to New France [Canada]
and with life in the new colony on the St Lawrence led to a unique mixture
[brassage] of central French, varieties of regional French and perhaps even
of certain patois … From this mixture there developed a new form of the
French language … a unified, coherent, distinctive Quebec variety. (my
translation)
They also speak of a “fusion dialectale” (Mougeon and Beniak 1994: 26).
Poirier (1994: 256) reports that “philological studies have proved the exis-
tence, in the middle of the 17th century, of a koiné along the St Lawrence
which was strongly influenced by the dialects of different French prov-
inces” (my translation).
Darot and Pauleau (1993: 295) point to the presence in the colonial
French of New Caledonia in the South Pacific of forms from southern as
well as Parisian French; and Hollyman (1979: 623) also cites the occur-
rence of features from the north of France as well as from the region of
Toulouse.
Lipski (1994: 45) also says of colonial Spanish that “it would seem that
a linguistic alchemy acted on the kaleidoscopic jumble of Peninsula lan-
guages and dialects to yield Latin American Spanish”. Mattoso Camara
(1972: 20) says of Brazilian Portuguese that “immigrants from both the
north and south of Portugal, seemingly in approximately equal numbers,
carried their respective dialects to Brazil”. Quoting da Silva Neto (1950:
10), he also suggests that “the coming together of such diverse dialects in a
single overseas center must have set up conditions leading to a sort of lin-
guistic compromise, a new kind of dialect”.
There is, however, a second approach to the study of the role dialect
mixture in the new-dialect formation of colonial varieties. This approach
enables us to achieve a greater time depth in our analyses. It consists in
acknowledging that mixed dialects have certain linguistic characteristics
which distinguished them to a certain extent from other varieties, and to
look for such characteristics. Usually, however, as we shall see, an element
of comparison with homeland varieties is still involved. In the rest of this
paper I concentrate on this second type of evidence.
2. Old English as a colonial mixed dialect
In Trudgill (1986), I make the rather obvious point that the development of
a new focussed dialect, in the sense of Le Page and Tabouret-Keller
(1985), out of a diffuse dialect mixture involves a reduction of variability –
a reduction in the number of extant variants for any given feature. I also
point out, however, that “an important fact about dialects that have recently
coalesced out of dialect mixtures is that even after focusing has taken
place, many of them continue to retain, at least for some generations, a
relatively high level of variability” (Trudgill 1986: 108).
This observation can be utilised in the following case. The earliest ex-
ample of English colonial dialect mixture involves the actual development
of English itself. We have evidence of various non-linguistic sorts that
southern and eastern England and southeast Scotland were initially settled
by Germanic speakers coming from all along the north sea littoral from
Jutland to the mouth of the Rhine – Jutes, Angles, Saxons, Frisians – and it
is reasonable to assume that they spoke different dialects. We also know
that Germanic speakers arrived from further inland, for example from place
name evidence. Swaffham in Norfolk, for instance, was originally Swae-
fasham, which meant ‘home of the Swabians’. But did contact between
these different Germanic dialects lead to dialect mixture? Can we say that
Old English was in origin a mixed dialect, just as we are claming for, say,
modern South African English?
Nielsen (1998: 78–79) answers this question in the affirmative, and sup-
plies the linguistic evidence. He argues that Old English was the result of a
mixture of West Germanic dialects from continental Europe. He points out,
for example, that it is because of dialect mixture that Old English had ini-
tially a greater degree of variability than the other Germanic languages
where no colonial dialect mixture had been involved. He cites the follow-
ing examples.
74 Peter Trudgill
(1) Old English had a remarkable number of different, alternating forms corre-
sponding to Modern English ‘first’. This variability, moreover, would ap-
pear to be linked to origins in different dialects from the mainland European
mainland: ærest (cf. Old High German eristo); forma (cf. Old Frisian
forma); formesta (cf. Gothic frumists); and fyrst (cf. Old Norse fyrstr).
(2) Similarly, OE had two different paradigms for the present tense of the verb
to be, one apparently resembling Old Norse and Gothic, and the other Old
Saxon and Old High German:
Goth O.Norse O.English I O.English II O.Saxon OHG
1sg. im em eom beom bium bim
2sg. is est eart bist bist bist
(3) Old English also exhibited variation, in all regions, in the form of the inter-
rogative pronoun meaning ‘which of two’. The form hwæCer relates to
Gothic hvaÞar and W. Norse hvaCarr, while the alternative form hweder
corresponds to O. Saxon hweCar and OHG hwedar.
We cannot know, of course, whether this type of variation was constrained

in any way. Most often (Trudgill 2004), where more than one variant from
a dialect mixture situation survives in the new dialect, reallocation takes
place (see below).
The suggestion is that even if we did not know from other non-linguistic
evidence that southern and eastern Britain were initially settled from many
different locations on the continent, there would have been at least some
linguistic evidence would have pointed us in the direction of that supposi-
tion.
3. Middle English in Ireland
In Trudgill (1986) I introduced the term interdialect. There, and in Trudgill

(2004), I define interdialect forms as forms which are not actually present
in any of the dialects contributing to a dialect mixture but which arise out
of interaction between them. Such forms are of three types:
a) They may be forms which are simpler or more regular than any of
the forms present in the original dialect mixture.
b) Interdialect forms may also be forms which are the result of hy-
peradaptation. The best known form of this is ‘hypercorrection’, in
which speakers attempt to use forms from higher status accents, but
employ an incorrect analysis and extend changes to items where

they are inappropriate. That this can have permanent consequences
for colonial varieties is shown by the fact that in many forms of Ja-
maican English (Cassidy and Le Page 1980), as well as in Tristanian
(see above), all stressed-vowel-initial words can actually begin with
/h/.
c) Interdialect forms may also be intermediate forms. In Trudgill
(1986) I suggested that these are most usually forms which are pho-
netically intermediate between two contributing forms in the mix-
ture. Mougeon and Beniak (1994: 26) write, for Canadian French, of
“innovations such as intermediate compromise forms” (my transla-
tion). They cite (1994: 40) an example originally provided by Rivard
(1914: 59): three different French forms of gens ‘people’, Standard
/Y`}/, Picard /Yd}/ and Saintongeais /g`}/, have combined to produce
Québécois /gd}/.
Samuels (1972: 108) uses both of the types of evidence for showing prior
dialect mixture, as outlined above, in connection with late mediaeval colo-
nial English in Ireland. First, he says that the available 14th and 15th cen-
tury texts show that the English settlers “must have been predominantly
from the West Midlands and South-West England”, but the language of the
texts
tallies with the dialect of no single restricted area of England; it consists
mainly of an amalgam of selected features from the different dialects of a
number of areas: Herefordshire, Gloucestershire, Somerset, Devon, Shrop-
shire, and to a lesser extent Cheshire, Lancashire, and possibly Wales.
(Samuels 1972: 108)
Most of the forms are found in large parts of the West Midlands and South-
west, but they do not all have the same geographical provenance. For ex-
ample, streynth ‘strength’ and throY ‘through’ are from the West Mid-
lands, while hyre ‘hear’ and ham ‘them’ arte Southwestern in origin.
Secondly, he focusses on a particular feature which is of special interest
because it clearly constitutes a feature of the type I have referred to as an
intermediate form. Samuels shows that in England, the Middle English
form for ‘each’ was uch in the West Midlands, and ech in the Southwest.
However, in the Anglo-Irish texts the norm is euch. It is true that there is a
small area in England, intermediate between the West Midlands and the
Southwest, just to the south of the towns of Worcester and Hereford, where
this form also occurs. But the fact that it is the norm in Anglo-Irish texts is
76 Peter Trudgill
of course not because Ireland was settled from southern Herefordshire.

Rather, euch was as a form which developed in Ireland, or survived from
England, or both, because it was intermediate between the two major com-
peting forms in the mixture that developed in Ireland, uch and ech. In
Trudgill (2004) I indicate that this kind of compromise development will
occur only when the surviving intermediate form is linguistically interme-
diate between two (or more) majority variants of roughly equivalent demo-
graphic strength.
Once again, then, it is clear that Ireland was settled by English-speakers
from a number of different dialect areas, and that the outcome was a new,
mixed dialect, consisting of a collection of forms from different dialect
areas, plus interdialect forms arising out of interaction between speakers of
different dialects.
4. North America
In Trudgill (1986) I further argued that in the process of new-dialect forma-

tion out of dialect mixture situations, it is generally the case that eventually
all variants in the mixture except one will be levelled out. Occasionally,
however, more than one competing variant left over from the original mix-
ture may survive. Where this happens, reallocation will occur. Indeed, we
can say that survival is normally actually dependent on reallocation. Vari-
ants originally from different regional dialects may, for example, be reallo-
cated functions as social class variants or stylistic variants in the new dia-
lect (see Trudgill 1986). Penny (2000: 54) shows how the original Iberian
Spanish regional differentiation between dialects, such as Old Castilian,
which had lost /h/ from Latin word-initial /f/ as in /umo/ ‘smoke’ and other
dialects, such as Extremaduran, which had retained it, as in /humo/, led
through colonial dialect contact to a situation where today “throughout
Spanish-speaking America” /h/ is found in lower-class and Ø in upper-
class sociolects.
I suggest that one such case is convincing evidence for the mixed nature
of Canadian English as a new colonial dialect. In what is now the USA,
there were a number of early anglophone settlements at different points
along the east coast. None of these was settled from a single location in
England, and in each case there would have been contact between different
English regional dialects. A proponent of dialect mixture scenarios would
therefore assume that, very early on, contact between different British dia-
lects would have occurred in the American settlements which would have
led to the appearance of new, mixed dialects not precisely like any dialect
spoken in the homeland. Each would have been, as Kurath has said, “a
unique blend of British types of speech” (1949: 1). The fact of modern
regional variation along the east coast of the USA would thus be explained,
not only in terms of different linguistic changes having taken place in dif-
ferent areas during the last 400 years, but also, more crucially, by the fact
that the initial mixtures – and therefore the outcomes of these mixtures –
were different in the different places from the very beginning (see Algeo
2001). In the words of McDavid (1985: 19) we can assume “dialect mix-
ture from the beginning in each colony … with different results”.
Canadian English would seem then to be even more the result of mix-
ture. The external, sociohistorical development of Canadian English would
appear to make it a mixed dialect par excellence. Proponents of the dialect
mixture hypothesis can argue for the development of Canadian English
along the following lines.
It descends for the most part from the mixing of these different, already
mixed USA dialects, as a consequence of a number of different waves of
immigration northwards, at different times, from different parts of the
American colonies. If we then add the fact of heavy 19th century immigra-
tion to Canada from England, Scotland and Ireland, the situation seems to
be crying out for a dialect mixture explanation.
This, at least, is what the immigration history and the demographics –
and the proponents of dialect mixture explanations – would suggest. But is
it really so? Can we support it with linguistic evidence?
Canadian Raising. There is considerable evidence that Canadian Rais-
ing, according to which distinctive and markedly different allophones of
the PRICE and MOUTH vowels occur before voiceless consonants (Cham-
bers 1991), is the result of, and therefore evidence for, dialect mixture. I
argued in Trudgill (1986) that so-called Canadian Raising may have arisen
out of a process of structural reallocation. General Canadian English, as is
well known, is characterised by very distinct allophones of /ai/ as in PRICE
and /au/ as in MOUTH. Before voiced consonants and word-finally these
diphthongs have open onsets, and before voiceless consonants central on-
sets: ‘out loud’ Z?tsk`tc]. The suggestion in Trudgill (1986) was that this
was due to the process of new-dialect formation which led to the develop-
ment of Canadian English as a new mixed variety. Scottish and Irish-type
diphthongs, with central onsets in all environments, and south of England-
type diphthongs, with open onsets in all environments, were both present in
the original dialect mixture which preceded the eventual focussed variety.
Both variants survived, but were reallocated a new function as positional
78 Peter Trudgill
allophonic variants. This reallocation was according to the very logical

principle that the narrower diphthongs were confined to the pre-voiceless
consonant environment where all English vowels have shorter allophones
(Laver 1994: 446).
This dialect mixture and structural reallocation thesis is greatly
strengthened by the following rather remarkable fact. ‘Canadian Raising’ is
by no means purely Canadian. It is also found in a number of other colonial
varieties. Indeed, it occurs “in nearly every form of non-creolised, mixed,
colonial English outside Australasia and South Africa” (Trudgill 1986:
160), although the actual phonetics is not necessarily the same as in Can-
ada.
For example, Tristan da Cunha English has clear Canadian raising for
the PRICE vowel. Schreier and Trudgill (2006) write that Tristan da Cunha
English “has a PRICE vowel which is a ‘fast’ diphthong [5H] before voice-
less consonants, but a ‘slow’ diphthong [P∂D] elsewhere, the latter demon-
strating both Diphthong Shift (Wells 1982: 256) and Glide Weakening.”
The mixed English of East Falkland has Canadian raising for both of
the diphthongs. Trudgill (2004), following Sudbury (2000, 2001), shows
that Falkland Islands English has a MOUTH vowel which is generally [8| ~
2|] before voiceless consonants, as in south, but [z∂T ~ `∂T] elsewhere as in
now. Similarly, PRICE is [5H ~ 2H] before voiceless consonants, as in right,
but [@∂H] elsewhere, as in time.
Canadian Raising is also found in the English of White Bahamians
(Trudgill 1986), of White Bermudans (Trudgill 1986), and of St Helena
(Trudgill 1986). It also occurs in certain areas of the American South.
Kurath and McDavid (1961) write that:
[i]n two rather sharply delimited areas of the South, (1) Virginia and adjoin-
ing parts of Maryland and North Carolina, and (2) the coastal belt of South
Carolina, Georgia and Florida, these “fast” diphthongs, with centralised be-
ginning occur only before voiceless consonants, as in twice, night, and differ
sharply from the slow diphthongs [`∂D, `∂?, @∂H] current in these areas before
voiced consonants and finally, as in five, nine, good-bye. In Virginia the dif-
ference between these positional variants is especially great, as in twice [?î]
vs. nine [`∂D, `∂?]. (Kurath and McDavid 1961: 109)
The fact that Canadian Raising is not found anywhere in Britain – with one
exception – further strengthens the hypothesis that it arose out of colonial
dialect mixture.
The one British exception, moreover, actually strengthens the hypothe-
sis, because it involves a relatively recent geographical spread of English
speakers and can also be shown – and shown in convincing chronological

detail – to be a dialect-contact feature in origin. We are able at this point to
“use the present to explain the past”. The dialect of the originally uninhab-
ited central English Fens is characterised by what Britain (1997, 2002)
calls Fenland Raising of PRICE. The PRICE vowel has central onsets [?H]
before voiceless consonants, but has [@H] or [@9 ] elsewhere.
Britain explains this as follows. Ellis (1889) and the Survey of English
Dialects (see Britain 1997) show that to the east of the Fens speakers have
conservative, raised onsets of /ai/ of PRICE in all phonological positions,
while speakers to the west have much more advanced open ones, including
monophthongs.
The developmental process which resulted in the emergence of reallo-
cated allophony in the Fens can be reconstructed as follows. During the
19th century this marshy area was drained and reclaimed, allowing large
numbers of migrants from both east and west to move in and take advan-
tage of the improved agricultural prospects.
The initial post-reclamation speech community would have included
both speakers of dialects containing western [ai]-type forms and eastern
[?h]-type forms. Importantly, Britain points to the fact that neither the west-
ern nor the eastern area appears to have dominated in providing migrants.
Following the settlement, children, perhaps over two or three genera-
tions, sought to impose regularity on this linguistic mêlée, and began to
focus on a new koinéised variety. In the case of /ai/, speakers simplified the
mixture of variants by reallocating them according to principles of phono-
logical naturalness – [?i] before voiced consonants and [@i] before other
environments.
This is not just guess-work on Britain’s part, however. In his variation-
ist study, Britain gives quantitative evidence across generations and across
geographical space which illustrates statistically the gradual sorting out of
variants into positional allophones. That is, he has been able to follow, in
apparent time, the development of Raising as it actually happened.
Further evidence for a contact origin for allophony in the Fens is the
absence of allophonic variation of /au/ – Canadian Raising of /au/ does not
occur in the Fens at all. For /ai/, the emergence of allophony was depend-
ent on the existence in the unfocused dialect of allophones with onsets of
differing degrees of openness which could be reallocated to different pho-
nological contexts. In the case of /au/, however, the onsets used by speak-
ers on both sides of the Fens were very similar, a half-open front ZD] (Brit-
ain and Trudgill 1999). Since there were no different onset realisations in
the dialect mix, reallocation simply did not occur.
80 Peter Trudgill
My suggestion then is that the three above examples indicate that it may
be possible to reveal, on linguistic grounds, the role that dialect mixture
has played in the formation of a new dialect, even at a time depth of some
centuries.
References
Algeo, John
2001 External history. In The Cambridge History of the English Lan-
guage, Vol. 6: English in North America, John Algeo (ed.), 1–58.
Cambridge: Cambridge University Press.
Bauer, Laurie
1994 English in New Zealand. In The Cambridge History of the English
Language, Vol. 5: English in Britain and Overseas – Origins and
Development, Robert Burchfield (ed.), 382–429. Cambridge: Cam-
2000 The dialectal origins of New Zealand English. In New Zealand Eng-
lish, Allan Bell and Koenraad Kuiper (eds.), 40–52. Amsterdam:
Benjamins.
Branford, William
1994 English in South Africa. In The Cambridge History of the English
Language, Vol. 5: English in Britain and Overseas – Origins and
Development, Robert Burchfield (ed.), 182–229. Cambridge: Cam-
Brinton, L. and M. Fee
2001 Canadian English. In The Cambridge History of the English Lan-
guage, Vol. 6: English in North America, John Algeo (ed.), 422–440.
Britain, David
1997 Dialect contact and phonological reallocation: ‘Canadian raising’ in
the English Fens. Language in Society 26: 15–46.
2002 Diffusion, levelling, simplification and reallocation in past tense BE
in the English Fens. Journal of Sociolinguistics 6 (1): 16–43.
Britain, David and Peter Trudgill
1999 Migration, new-dialect formation and sociolinguistic refunctionalisa-
tion: Reallocation as an outcome of dialect contact. Transactions of
the Philological Society 97: 245–256.
Cassidy, Frederick and Robert B. Le Page
1980 Dictionary of Jamaican English. Second edition. Cambridge: Cam-
Chambers, Jack
1991 Canada. English around the World: Sociolinguistic Perspectives.
Jenny Cheshire (ed.), 89–107. Cambridge: Cambridge University
Press.
Combrink, Johan
1978 Afrikaans: Its origins and development. In Language and Communi-
cation Studies in South Africa, Len Langham and K. P. Prinsloo
(eds.), 69–85. Cape Town: Oxford University Press.
Darot, Mireille and Christine Pauleau
1993 Situation du Français en Nouvelle-Calédonie. In Le Français Dans
l’Espace Francophone, Didier de Robillard and Michel Benjamino
(eds.), 283–301. Paris: Champion.
da Silva Neto, Serafim
1950 Introdução ao Estudo da Língua Portuguêsa no Brasil. Rio de Ja-
neiro: Presença.
Ellis, Alexander
1889 On Early English Pronunciation, Vol. 5. London: Trübner.
Frings, Theodor
1957 Grundlegung einer Geschichte der deutschen Sprache. Third edition.
Halle: Niemeyer.
Hammarström, Göran
1980 Australian English: Its Origins and Status. Hamburg: Buske.
Hollyman, K. Jim
1979 Le Français en Nouvelle-Calédonie. In Le Français Hors de France,
Albert Valdmann (ed.), 621–629. Paris: Champion.
Kurath, Hans
1949 A Word Geography of the Eastern United States. Ann Arbor: Uni-
versity of Michigan Press.
Kurath, Hans and Raven McDavid
1961 The Pronunciation of English in the Atlantic States. Ann Arbor:
University of Michigan Press.
Lanham, Len
1967 The Pronunciation of South African English. Cape Town: Balkema.
Lass, Roger
1990 Where do extraterritorial Englishes come from? In Papers from the
5th International Conference on English Historical Linguistics, Syl-
via Adamson, Vivien Law, Nigel Vincent and Susan Wright (eds.),
245–280. Amsterdam: Benjamins.
Laver, John
1994 Principles of Phonetics. Cambridge: Cambridge University Press.
Le Page, Robert B., and Andrée Tabouret-Keller
1985 Acts of Identity: Creole-based Approaches to Language and Ethnic-
ity. Cambridge: Cambridge University Press.
82 Peter Trudgill
Lipski, John
1994 Latin American Spanish. London: Longman.
McDavid, Raven
1985 Dialect areas of the Atlantic seaboard. In American Speech: 1600 to
the Present, Peter Benes (ed.), 15–26. Boston: Boston University
Press.
McWhorter, John
2000 The Missing Spanish Creoles: Recovering the Birth of Plantation
Contact Languages. Berkeley: University of California Press.
Mattoso Camara, Joaquim
1972 The Portuguese Language. Chicago: University of Chicago Press.
Moag, Rodney
1977 Fiji Hindi. Canberra: Australian National University Press.
Morin, Yves-Charles
1994 Les sources historiques de la prononciation du Français du Québec.
In Les Origines du Français Québécois, Raymond Mougeon and
Édouard Beniak (eds.), 199–236. Sainte-Foy: Laval University Press.
Mougeon, Raymond and Édouard Beniak (eds.)
1994 Les Origines du Français Québécois. Sainte-Foy: Laval University
Press.
Mufwene, Salikoko
2001 The Ecology of Language Evolution. Cambridge: Cambridge Univer-
sity Press.
Nielsen, Hans Frede
1998 The Continental Backgrounds of English and its Insular Develop-
ment until 1154. Odense: Odense University Press.
Penny, Ralph
2000 Variation and Change in Spanish. Cambridge: Cambridge University
Press.
Poirier, Claude
1994 La langue parlée en Nouvelle-France: Vers un convergence des ex-
plications. In Les Origines du Français Québécois, Raymond
Mougeon and Édouard Beniak (eds.), 237–274. Sainte-Foy: Laval
University Press.
Rivard, Adjutor
1914 Études sur les Parlers de France au Canada. Quebec: Garneau.
Romaine, Suzanne
2001 Contact with other languages. In The Cambridge History of the Eng-
lish Language, Vol. 6: English in North America, John Algeo (ed.),
154–183. Cambridge: Cambridge University Press.
Samuels, Michael
1972 Linguistic Evolution: With Special Reference to English. Cambridge:
Schreier, Daniel
2003 Isolation and Language Change: Sociohistorical and Contemporary
Evidence from Tristan da Cunha English. London: Palgrave Macmil-
lan.
Schreier, Daniel and Peter Trudgill
2006 The segmental phonology of nineteenth-century Tristan da Cunha
English: Convergence and local innovation. English Language and
Linguistics 10 (1): 119-141.
Sudbury, Andrea
2000 Dialect Contact and Koinéisation in the Falkland Islands: Develop-
ment of a New Southern Hemisphere English? Colchester: University
of Essex PhD thesis.
2001 Falklands Island English: A Southern hemisphere variety? English
World-Wide 22: 55–80.
Trudgill, Peter
2004 Dialect Contact and New-Dialect Formation: The Inevitability of
Colonial Englishes. Edinburgh: Edinburgh University Press.
Turner, George W.
1994 English in Australia. In The Cambridge History of the English Lan-
guage, Vol. 5: English in Britain and Overseas – Origins and Devel-
opment, Robert Burchfield (ed.), 277–327. Cambridge: Cambridge
University Press.
Wagner, Max
1920 Amerikansich-Spanisch und Vulgärlatein. Zeitschrift für Romanische
Philologie 40: 286–312, 385–404.
Wall, Arnold
1938 New Zealand English: How It Should Be Spoken. Auckland:
Whitcombe and Tombs.
Wells, John C.
1982 Accents of English. (Three Volumes). Cambridge: Cambridge Uni-
versity Press.
Chapter 5
Non-standardisation
Daniel Schreier
1. Introduction
Given the amount of attention Standard English has received in the litera-
ture over the last few years (with general focus on the historical dimension
of standardisation, the origins of Standard English, mono- or pluricentric
standards of English around the world, reference varieties for English lan-
guage teaching (ELT), etc.; see Graddol, Leith and Swann 1996; Leith
1983; Cheshire and Stein 1997; Bex and Watts 1999; Lippi-Green 1997;
Watts 1999, 2000; Wright 2000; Bonfiglio 2003; Crowley 2003), it is
worth stressing every once in a while that most off-spring varieties of Brit-
ish English have in fact not standardised at all. Standardisation is the ex-
ception rather than the norm, and the amount of variability within the Eng-
lish-speaking world is so immense that it is by no means exaggerated to
claim that the majority of native speakers of English speak a non-standard
variety. This becomes particularly clear when considering the so-called
“Inner circle varieties” from Kachru’s highly influential (1985) model, i.e.
those English varieties that developed in a context of restricted bi- and/or
multilingualism and where a native-speaker tradition was maintained
throughout the evolutionary phase. Though there is no doubt about the
existence of British and American standard varieties, it is highly debatable
whether (and to what extent) Australian or New Zealand English are in the
process of standardisation or not (cf. Schneider 2003). Few would go as far
as to suggest that standardisation has operated in the Outer and Expanding
circles, and as far as I know, nobody has argued in favour of the existence
of Standard Bahamian, Standard Hong Kong or Standard Swiss English
(though it has been discussed how these varieties function and develop,
whether and to what extent they display substratum processes, and whether
there is any linguistic justification for these terms; cf. Rosenberger 2006).
Consequently, anybody interested in the spread of English as a world
language must be interested in the evolution, function and diversity of
‘non-standard English’ as well, for descriptive and theoretical reasons.
86 Daniel Schreier
Given the fact that all the post-colonial varieties of English have developed
with more or less influential impact of British donor varieties (many of
which were non-standard to start out with; see below), a fruitful (yet by
some sadly neglected and, most famously by Honey (1997), even contested
and ridiculed) question is how non-standard varieties have developed all
over the English-speaking world, and how the transplantation of British
English to settings outside the British Isles has given rise to the emergence
of ‘new’ varieties, in the form of pidgins, creoles, koinés, etc. How do non-
standard varieties come into existence, and how do they vary and change
and evolve? And why, in analogy to decreolisation processes (Mühlhäusler
1997), should the speakers of non-standard varieties not strive towards
standardisation, since their variety is not socially valued?
These questions have been addressed before, of course, and sociolin-
guists have advanced various explanations for the continuing use of non-
standard varieties (see, among others, Trudgill 1972 on the concept of cov-
ert prestige, or Milroy and Milroy 1987 on the impact of social network
structures). This paper has two aims: (1) to discuss non-standardisation as
a global phenomenon, approaching it as a contact-induced process, namely
by asking how contact between several language and/or dialect varieties
gives rise to a ‘new’ non-standard variety; and (2) to address the question
how historical linguists can reconstruct and retrace the origins and devel-
opment of such a variety. Given the fact that a contact scenario’s feature
pool (Mufwene 2001) is characterised by extreme diversity and social,
regional and individual differentiation, the question pursued here is how
we can disentangle the sociolinguistic complexity of a variety’s initial
stages and identify the principal donors that shape a ‘new’ non-standard
variety, thus contributing to ongoing non-standardisation in the English-
speaking world. The analysis is strictly historically-oriented; by means of
illustration and due to limitations of space, I will focus on the English vari-
ety that has developed on the island of St Helena in the South Atlantic
Ocean (see Schreier 2008).
2. English as a non-standardising language
English is one of the best-studied languages in the world, and this for a
number of reasons. The spread of English from a language that was locally
confined to one spoken all over the globe, i.e. its development from an
isolated variety spoken by perhaps 5–7 million speakers before 1600,
mostly monolinguals, to ‘the language on which the sun never sets’, spoken
Non-standardisation 87
on all continents with an estimated number of more than one billion speak-
ers, occurred over the last four centuries only and is thus a comparatively
recent development (Crystal 2003). English has undergone a period of evo-
lution and diversification in a not-too-distant time frame, for which we
have written and, particularly over the last century, ample spoken evidence,
and this allows us to describe successive stages of the variety and to trace
its lineage and development. The contact scenarios English became em-
bedded into saw the emergence of a whole range of so-called ‘new’ varie-
ties, from koinés (which originate in contact between mutually intelligible
and structurally similar varieties, i.e. through dialect contact) to pidgins
and creoles, i.e. the eventual output of language contact, involving English,
other European and indigenous languages (see below). This allows histori-
cal and genetic linguists to gain insights into how a language diversifies as
a result of regional expansion, and these insights can in turn be projected
historically (in Sankoff’s [2006: 121] words: “slicing through the present
to the past”). Consequently, the recent development of English may serve a
larger purpose, namely to trace general processes of linguistic diversifica-
tion and language birth, perhaps even the development of ‘new’ language
families. The question as to whether English is in the process of breaking
up into a subset of ‘new’, linguistically distinct and mutually non-
intelligible varieties is a hotly debated issue (Quirk 1990; Kachru 1991; see
the discussion in Graddol, Leith and Swann 1996).
Of particular importance for the present paper is the fact that
(post-)colonial varieties practically always originate in contexts that in-
volve contact between several linguistic systems and subsystems. It cannot
be emphasised enough that relatively ‘pure’ contact scenarios are rare (the
case of the Falkland Islands perhaps coming closest). Practically all colo-
nial varieties, independent of population size and location, are formed in
environments that witness the simultaneous or sequential input of a number
of varieties. Typical contact scenarios can be roughly classified as follows:
1. Contact between English as a colonising language and local indige-

nous languages (for example along the Slave and Gold Coasts in
West Africa), thus typically in linguistically heterogeneous, multi-
lingual environments
2. Contact between various coloniser’s languages (for instance between
English and French, which coexisted when England and France si-
multaneously claimed Vanuatu)
3. Contact between several varieties of one colonising language (such
as on the Falkland Islands, which did not have an indigenous popula-
88 Daniel Schreier
tion and were mostly settled by colonists from Southwest England

and the Scottish Lowlands)
If sociolinguistic contact sees the formation of (mostly non-standard) varie-

ties, then the question is how the socio-historical evolution of ‘new’ varie-
ties can be reconstructed, and what methodology is most reliable for this
purpose. Generally speaking, there are two approaches: a more synchronic-
ally-oriented one, i.e. by collecting data of a variety as it is now and then to
retrace its steps and reconstruct earlier stages and putative sources, or to
opt for a more diachronic approach, namely by locating earlier evidence
and interpreting attestations from various sources. Whereas the former
technique has produced much of what we know about English as a world
language (see for instance the articles collected in the monumental A
Handbook of Varieties of English, Schneider et al. 2004), a more dia-
chronic approach is indispensable for a reconstruction of earlier develop-
mental stages. Crucially, though, it is much more unreliable and sketchy.
The pitfalls for the analysis of written documents have been described in
detail by Montgomery et al. (1993) or Schneider (2001), so it is sufficient
to list just some of the most important ones. Written evidence poses the
following problems for analysis: (1) it is often unclear whether the sources
are reliable and trustworthy (who wrote the documents we base our hy-
potheses on, for what purpose were they produced, and how accurate are
they?); (2) did the features reported really make an appearance or are they
just attributed?; (3) there is often no information as to where the feature in
question occurred – in the speech of small groups, across the community or
in single speakers’ idiolects only?; and (4) are the sources diachronically
representative?
In the following, I would like to discuss issues related to the location
and evaluation of historical data and show how (given necessary caution
and a step-by-step evaluative process) written sources can help historical
linguists in their endeavours to outline the origins and evolution of a so-
called ‘new’ non-standard variety, i.e. one that is English-derived and that
developed in a context of British colonial involvement, in an environment
that was continuously inhabited by native speakers of (mostly non-
standard) British English, yet witnessed admixture of speakers from other
places of origin and with other different linguistic backgrounds. The basic
question is not only how the variety under scrutiny came to be the way it is
and why it underwent this particular trajectory (and not another one), but,
from a methodological perspective, how researchers, making use of social,
socio-historical, sociodemographic and sociolinguistic evidence, can iden-
tify and assess the various contact types and thus research non-
standardisation processes. These questions are discussed with reference to
St Helenian English (StHE), a variety of South Atlantic English and in fact
the oldest variety of Southern Hemisphere English. I begin with a brief
discussion of the island’s social history before going on to present, discuss
and assess some earlier evidence of the variety.
3. St Helena: A brief social history
The volcanic island of St Helena lies in the mid-central South Atlantic

Ocean, 1,930 km west of Angola and just south of the equator. Its nearest
neighbour (geographically speaking) is Ascension Island, more than 1,000
kilometres to the northwest. St Helena covers 122 square kilometres, and
its topography mostly consists of steep, relatively barren and rocky terri-
tory, unsuitable for cultivation. The island’s capital and only town is
Jamestown, although there are other smaller settlements such as Halftree
Hollow, Blue Hill, Sandy Bay and Longwood (the latter being the resi-
dence of Napoleon Bonaparte, who was exiled on the island from 1815 to
his death in 1821). St Helena’s population of approximately 4,000 (2006)
is almost without exception of mixed European, African and Asian origin,
and English is the only language spoken on the island.
Originally uninhabited, St Helena was discovered in 1502 by the Portu-
guese, who, just like the other European seafaring nations who followed in
their wake, used the island as a refreshment station and sick bay on their
journeys to and from the East. Until claimed by the British East India Com-
pany (EIC) in 1658, the island was never permanently or formally settled
(Gosse 1938). From this date, a concerted settlement policy was imple-
mented, and soldiers, servants and planters (employed and contracted by
the EIC, who held direct control over the island until the 1830s) were re-
cruited to St Helena, along with slaves supplied on request by the Com-
pany’s ships. Even though the exact origins of the British settlers are not
known, there is socio-historical evidence that most of them came from
southern England. Moreover, the majority of the planters had working
class origins and the EIC recruited many of its soldiers (and settlers as
well, for that matter) from among the unemployed in England (Gosse 1938:
72). We know that many of them were illiterate (as evidenced by an entry
in the St Helena Consultations (quoted in Brooke 1808) on February 2nd,
1774: “On 31st Jan. six soldiers deserted in the night taking two Boats …
90 Daniel Schreier
The deserters were illiterate men of bad character and only a few days pro-
visions and must inevitably perish at sea.”).
The origins of the non-white population are better documented, and
various records show that slaves were imported from the Guinea Coast, the
Indian sub-continent and Madagascar, and to a lesser extent from the Cape
and Larger Table Bay area, the West Indies, Indonesia and the Maldives. In
1789, the importation of slaves officially ended, but the lack of cheap la-
bour was compensated by Chinese indentured labourers who arrived on the
island from 1810 onwards. However, very few, if any, stayed on perma-
nently and slavery was finally abolished in 1832 (Melliss 1875). In 1815,
the total population was 3,342, comprising 694 whites, 1,517 slaves, 933
non-permanent army personnel, as well some 300 indentured labourers
from China (see Table 1, adapted from Schreier 2008).
Table 1. The sociodemographic situation on St Helena in 1815
Men Women Children Total
White 109 202 383 694
Non-white 342 245 610 1,197
Army person- 933 (combined) 933
nel (white)
Company 98 (combined) 98
slaves (Black)
Free blacks 420 (combined) 420
Total 3,342
The situation was to change dramatically, when in 1834, St Helena’s ad-

ministration was transferred from the EIC to the British government and St
Helena officially became a crown colony. Poverty led to out-migration and
the remainder of the 19th century was characterised by extreme hardship.
Crucially, this period saw an increase in mobility and also ethnic mixing.
Governor Charles Elliot remarked in 1868 that “there can be no position on
the face of the earth where it would be more difficult to discriminate be-
tween the various strains of blood of which the body of the population is
composed than here in St Helena” (quoted in Gosse 1938). The population
loss due to out-migration was compensated for by the arrival of immigrant
groups: the indentured labourers from China, liberated African slaves,
brought to the island after 1840, when St Helena was used as base for re-
habilitating slaves from captured slave ships (some of them chose to stay
while the majority were sent on to the West Indies or repatriated back to
the African mainland) and hundreds of Afrikaans-speaking Boer War pris-

oners in 1902, only very few of whom stayed behind upon their release.
The increasing use of steam-driven ships and the opening of the Suez
Canal voided the island’s strategic purpose as a refreshment station. With
the exception of a short-lived flax industry (which ended in 1965 when the
British postal service switched to cheaper synthetic fibre), no industry has
provided a viable means of sustaining the island. There is no airport and
the single government-subsidised ship that connected the island with the
United Kingdom has changed its route to become more cost-efficient, now
only serving Ascension, St Helena and Cape Town (the annual run to Tris-
tan da Cunha was cut as well). Today, many Saint Helenians (or ‘Saints’,
as they call themselves) take on contract work on the military bases on
Ascension and the Falkland Islands; since 1999, when the British Govern-
ment conceded full citizenship rights to the islanders, they have had full
access to United Kingdom workplaces. This has affected the community
heavily, as perhaps up to 30 per cent of the population, mostly younger
Saints, left the island in search of better job opportunities.
4. The sociolinguistic evolution of St Helenian English
We need to address a number of questions in order to reconstruct the his-

tory of StHE and by doing so, to understand how StHE came to be the way
it is. The issues, by nature historical, social and sociodemographic, linguis-
tic and sociolinguistic, are addressed in turn, before I evaluate them and
reach some first conclusions and guidelines for further research.
To begin with historical, social and sociodemographic issues, there is
no doubt that settlers, planters, administrative staff, soldiers and slaves
came to the island from three continents: Europe, Africa and Asia. The
origins of the settlers (and founders of the colony) can be identified as
follows:
– England (mainly south(-eastern) England)
– France
– St Iago, Guinea, Nigeria, southern Africa, India, Indonesia, Mada-
gascar, Maldives
– China (indentured labourers in the 1820s and ‘30s)
– South Africa (POWs from the Boer War, 1898–1902)
– West Africa (liberated slaves from 1850s)
92 Daniel Schreier
The strong early presence of English settlers and planters is evidenced by

the first account of life on St Helena: Henry Gargen’s (1665) diary, in
which he describes the little community in quite some detail, as the follow-
ing excerpts show:
(1) Now I shall enbold my selfe to give you account of ye inhabitants and there
[sic] habitations. Firstly Mr Robert Stringer and his family which is his wife
and his sonnen and sonne in law. And his two daughters 2 maid servants and
one man servant and 4 blacks. Mr Butler second: with his wife and his
childe and one servant: now for ye rest: John Coulson His wife and childe,
John Wood His wife and childe; William Young Steward His wife and
childe; John Evans Militiaman His wife and childe; Thomas Harper His wife
and childe; John Waily His wife and childe; William His wife and sonne in
law; Thomas Holton and his wife; Now for ye rest as are single persons:
William Rogers: William Fox: William Gates: Richard Price: Frances Mold:
James Hastings: Thomas Evans: Edward Carre: James Codner: John Young:
Thomas How: John Greentree: Henry (Pride) … William Goy … Thomas
Collins: Richard Swally Thomas Hallicross staied in my birth: Mr Charles
Higginson; … now as for ye rest of ye Englishmen there is 3 that run away
… Now I will sattisfie you of ye freemen and there [sic] plantation and hoe
they lie there 4 of them in all. John Coulson; John Wood; Thomas Harper;
John Waily.
All the historical sources agree that there was a strong British presence on
the island. However, they also evidence that slaves were on the island from
the early 1660s onwards. Their precise origins are unknown, but order was
given that “[i]n case at St. Iago you can procure five or six blacks or Ne-
groes, able men and women, we desire you to buy them, provided they may
be had at or under 40 dollars per poll” (Company Instructions, quoted in
Gosse 1938: 46–47). A later entry in the St Helena Consultations (col-
lected by Janisch 1885) points to Guinea as a place of origin: “There are
but twelve out of 42 Guinea Blacks now living yet more dyed from want of
care or victuals.” From the 1700s on, the preferred place of origin of slaves
was Madagascar, and requests were made to provide slaves from that is-
land exclusively. Other sources (Brooke 1808; Janisch 1885; Gosse 1938)
inform us that some of the inhabitants came from France (Huguenots, who
fled in the 1590s). Consequently, then, the historical account of the local
community allows us to pinpoint the founders and to speculate on their
sociolinguistic input, which is particularly important for the question as to
whether there was a potential for “founder effects” (Mufwene 1996). His-
torically speaking, these were the English planters and their offspring
(other groups were too small in size and keen to integrate and accommo-
date, e.g. the French Huguenots, or they left the island before they could
make an impact on the evolution of the local variety, e.g. the South African
POWs), as well as the slaves, the majority of whom came from Madagas-
car.
Sociodemographically, the question is how the populace developed gen-
erally and how stable or transient it was. In the years following the Dutch
Interregnum (when St Helena was taken over by the Dutch for a few
months in 1672), there were countless conflicts in the island community,
affecting soldiers and planters alike. Promises of land grants were broken
and the late 17th century was characterised by quasi-feudal relationships
between the Company directors in London, the local administration and the
settlers and their families, who were in fact nothing else than indentured
labourers. Most of the planters were dissatisfied with these conditions and
left the island at the first occasion. There were fluctuations in the army and
administrative staff as well. The fort saw revolts and rebellions, which lead
to casualties, death penalties and out-migration. In 1693, for instance, dur-
ing the infamous ‘Jackson Conspiracy’, 27 soldiers seized the fort (killing
the Governor), and escaped from the island on a Company ship. As for the
slaves, the St Helena Records (StHR; Janisch 1885) mention dozens of
escapes, mostly by young slaves. A number of slave uprisings were re-
ported in the 1680s and 1690s, and slaves were trialled and brutally pun-
ished, often without proof or following a forced confession. As a result,
slaves seized every opportunity to flee the island, and the loss of manpower
was compensated by the importation of new labour, preferably from Mada-
gascar (more than one hundred slaves arrived from that island in 1716
alone).
In addition, illnesses and diseases ravaged the community in the first
half of the 18th century, and diseases like chicken pox spread on the island
like wildfire. Thirty planters died in the first half of 1718 alone and there
were large numbers of victims in the black population. Mortality rates in
this period were exceptionally high. An entry in the Consultations (Janisch
1885) from May 26, 1719 reads that “We usually decrease here among the
white people five in a hundred per annum – but in each of the two last
years not lesser than 10 per 100!!”. Census and shipping lists show that
there was a massive population turnover on St Helena, and the population
only stabilised from the 1740s onwards. Consequently, then, we can ex-
clude “founder effects”, simply because many of the founders did not stay
on the island long enough to leave an impact.
94 Daniel Schreier
5. Earlier evidence of St Helenian English
This leads me to the central question of the paper, namely how the earliest
forms of StHE can be identified and reconstructed sociolinguistically. The
sources used for this purpose are of different kinds, ranging from letters,
reports sent by island representatives to the London headquarters of the
East India Company, verbatim protocols of court cases, short stories, cas-
ual observations, etc. Of course, one must not make the mistake of taking
all these features at face value (even though I simply lack the space to enter
into a discussion of analytical problems here), but we will see that the fea-
tures reported yield a pattern nevertheless, which complements and sup-
ports the social and sociohistorical evidence presented.
The diachronic excursus begins with the most copious and reliable ac-
count of St Helena’s social history: the StHR (Janisch 1885). Though few
and far-between, they contain useful material, particularly when the direct
speech of white soldiers and planters is reported. Court cases are particu-
larly important here; the scribes had to render the exact wording because
the testimonials had to be sent with the general reports to the Company’s
headquarters in London, where punishment and indictment were assessed
and approved. An additional source of information comes from islanders’
letters sent to the UK, either to families or relatives in England or to the
East India Company directly. These specimens are rare, but they occasion-
ally contain interesting features. Consider for instance the following pas-
sages (emphasis added):
(2) Taylor complains that “Parrum’s cow is an unlawful beast for she hath bin
seen to leap over a wall above six foot high and over a ditch nine foot wide
and eight foot deep.” (StHR; January 21st, 1687)
(3) Mr. Coxe late Member of Council accused of employing the Company’s
blacks on Various and numerous occasions for his own benefit and advan-
tage – also “keeping one Black when ships were in the road continually
shooting of Patrages and Ginny hens for presents to commanders and when
shipps was not here to kill him wild goats which he presented as he pleased
and sent nor spent any of them to ye Govr” (StHR; April 26, 1688)
(4) I remember you was some years past very uneasy wth Mr. Beale (Letter
from Josiah Child; February 20, 1689)
(5) Complaint against Jonathan Mudge that he whipt Sarah Sinsemig’s son
while picking green tobacco on the Common. Mudge says yes he did be-
cause she and her son used to call to him as loud as they could in a jeering
way as he went by her house “Goe home Boy, Goe home Boy, Red cards is
trumps.” (StHR; April 19, 1697)
(6) Mr. Humphrey is come here and would faine extenuate ye crimes (Letter;
April 16, 1701)
(7) Edmonds answered what for running away I believe the Governor and Field
is agreed (StHR; September 29, 1702)
(8) Again last night he came drunk into my private room putting many imperti-
nent questions chiefly why did I not order a guard to wait on him ashoar be-
ing a gentleman, when I myself rather than to weaken our guards I won't use
none, or very seldom (StHR; October 27, 1703)
(9) Anthony the Prisoner of the Bar, Juba and he eat some Rice together about
Nine of the Clock, that same night, after they had eat the Rice, the Prisoner
bid them a good Night (StHR; 1762)
With the exception of the last sample, these passages come from docu-
ments produced between 1687 and 1703, thus very early on in settlement
history. They contain a number of insightful features:
– no plural –s with count and mass nouns (“a ditch nine foot wide”)
– past be levelling with pivot form was (“when shipps was not here”)
– bare root extension (i.e. usage of infinitives as preterits: “they had eat
the Rice”,1 also found in Thomas Goodwin’s report of the Jackson Con-
spiracy: “When they weighed anchor and set sail and run two leagues
off”)
– present tense be concord with is, particularly with plural NPs and con-
joined subjects (“Red cards is trumps”, “the Governor and Field is
agreed”)
– perfective be (“Mr. Humphrey is come here”)
– multiple negation (“I won’t use none”).
One cannot help but notice that (with the exception of perfective be and
present be regularisation) these features are widely attested in many (if not
all) non-standard varieties of British English. In fact, they are not in the
least regionally confined (such as no plural with mass nouns, or multiple
negation, or past be regularisation, the latter two representing “vernacular
roots” (Chambers 1995: 242), i.e. features that make an appearance in Eng-
lish varieties all over the world). If they are of any interpretative value for
96 Daniel Schreier
our search of the original inputs and the successive development of StHE,
then we can say that the manifestation of these features points to a working
or lower middle class background of the planters and soldiers, which sug-
gests that non-standard inputs had a strong impact on the development of
StHE.
Other insights come from newspapers, which flourished in the 19th cen-
tury; short stories, jokes, verbatim reports of conversation of “country fell-
ers” (usually for humorous purposes), articles and letters to the editor are
particularly promising in this context. As an illustration, let us consider
excerpts from the following letter to the editor, which originally appeared
in the first issue of The Advocate, Or St. Helena Weekly News (on May 15,
1851):
Mr. Hedditor, I is but a ould Yamstok, but Bil Jinkens wot is a Hinglishman
sais, as ou u is the proprest wun to rite to bout evree ting ; and Bill noes a
kow’s tale from a jak-ass as well as any man vot hever I seed; and ven he fus
cum’d to de hiland Bill knowed a yam from a tater drekly – by instink as he
kalls hit. So I takes Bills adwice, and if so be has I his rong, u musint be
hangry. Vot I wants to no is the rites of dis ere ; vy. they maks us pore fell-
ers pa for everee ting as day dus. – If ve vonts to go a fishen it ar 30 shillens
for a bote licens everee yare, an’ 3 poun for a passeg bote, an one poun for
everee tun if it ar a luggedge bote–wich don’t luke rite an jonnak; cos ve
mus git as vell as udder fokes vot is gut mor munny den ve is. An nou day
maks us pa werry hi for de vine vot ve drinks by risin de dooty, an it vornt
so werry little afore needer. Nou do it be werry fine for em to sa as ou day
vonts us not to drink, stil it don’t luke natril like not to drink nuthin.
Authorship and purpose of the document remain unclear. It could have
been a serious concern by a semiliterate Saint (who describes himself as
somebody “vot aint ad no heddykashen”) and who prefers to remain anony-
mous, signing the letter “A OULD YAMSTOCK”. Interestingly, there is no
reference to a Jinkens (or Jenkins?) family in the 1815 census (the StHR
mention Governor Robert Jenkins in the 1740s, but, having lived a century
earlier, he can’t have been the addressee of the first letter). Alternatively,
and perhaps the most plausible explanation, the document is a made-up
mock letter, produced for humorous purposes and general amusement (e.g.
to attract readership for a newly established newspaper). What is clear is
that it abounds in non-standard features; if these are in any way representa-
tive of 19th century StHE, then they provide a veritable treasure chest for
our present purpose (regardless of questions relating to authorship and
authenticity). To list just some of the phonological features that make an
appearance:
– hypercorrect /h/ (“Bil Jinkens wot is a Hinglishman”, “u musint be

hangry”)
– /h/ dropping (“dem fellers vot drinks so ‘ard”, “them as as, arnt no
better”)
– /v-w/ interchange, both towards /w/ (“I takes Bills adwice”) and /v/
(“a drop o’ vine for dinner”), sometimes in both directions and in the
same sentence (“the wimmen is so wishus ven day taks it into thare
heds”, “nou day maks us pa werry hi for de vine vot ve drinks”)
– consonant cluster reduction (“ven he fus cum’d”, “a drop for a fren”)
– jod dropping (“by risin de dooty”)
– TH stopping (“vitch is I tink werry clever”, “the rites of dis ere”)
– alveolar /n/ for –ing (“to git a onnest liven”, “keep fokes from
drinkin”)
– tense vowel in happY (“pa for everee ting”, “I dus pittee them”)
– STRUT for NURSE (“ven he fus cum’d”, “de moor vot day as to pa the
wuss it be for de ould man”)
It is quite striking that, in contrast to the samples from the StHR, many of
these features can be localised, namely to Southeast England and the
Greater London area (hypercorrect /h/, /h/ dropping, the /v-w/ interchange,
jod dropping, STRUT for NURSE, adopting Wells’ 1982 system of lexical
sets). The first hypotheses developed on social and historical information
are linguistically confirmed, and there is sociolinguistic evidence that non-
standard varieties of southeastern England provided a substantial input
here.
Finally, before turning to an evaluation and some first conclusions, we
need to look into reports of slave speech, which is particularly important
for the potential of pidginisation and/or creolisation on the island. First of
all, consider the trial against a slave by the name of Sattoe. A white planter
claimed that Sattoe had physically attacked and attempted to kill him, and
the court transcriber rendered the trial as follows: “Sattoe being examined
confesseth that upon his said master beating him and threatening to do it a
second time he took his knife and wounded him in the arm and legge. That
he received the knife from Rowland Mr. Swallows Black about ten days
before who said when your master beat or strike you then doe you beat him
again and kill with this knife” (StHR; November 3, 1679). This short sam-
ple contains two diagnostic features that point to possible creolisation
and/or substratum transfer: 3rd person singular present tense zero (“your
master beat or strike you”, and PRO drop (“you beat him again and kill Ø
with this knife”), which suggests that language contact on St Helena wit-
98 Daniel Schreier
nessed manifestations of pidginisation and creolisation (more samples are

located and discussed by Wright 2004), legitimising speculations that StHE
as such bears resemblances with Atlantic Creoles (Hancock 1991).
A much later, but equally insightful example comes from another arti-
cle, “The Secret of Mount Eternity”, which appeared in the Monthly Critic
and Flashlight in 1892. The first person narrative of a slave is rendered as
follows:
Ye see dis powd’-horn Massa? Well, one day that ere cork in de small end
done stuck fas’, and I ‘magines all day in my kop how I get him out. Not
like knock him in fear I split de horn, so I dinks I jus’ burn him out shua! So
I put my ramming rod in the fiah, and when he jus’ nice and read, I stick de
cold end of de rod inter de ground … I done forget dat de horn full ob pow-
der, and when de rod burn troo de cork, it go kerslick inter that powder. And
de powder explode … I clar to the Lord, Massa, it take three doctors to cut
it out, and it leave mark all around my eye you see. Surgeon-Major say
‘Jake, you berry long-headed man, and you head jus’ thick as it long; dat
what save you.’
This passage contains a number of (proto-)typical creole features, namely:
– completive done (“that ere cork in de small end done stuck fas’”, “I
done forget dat”)
– consonant cluster reduction (“stuck fas’”, “dat what save you”)
– TH stopping (“so I dinks”, “de rod burn troo de cork”)
– /v/ realised as [b] (“full ob powder”, “you berry long-headed man”)
– bare root extension (infinitive forms as preterits: “it take three doctors
to cut it out”, “I stick de cold end of the rod”, “and it leave mark”)
– copula absence (“you head Ø jus’ thick as it Ø long”, “when he Ø jus’
nice and read”)
– PRO drop (“Ø Not like knock him”)
Consequently, then, we have every reason to assume that the StHE feature
pool contained pidgin- and/or creole-like forms as well, and that these were
brought to the island rather than originating due to language contact (Eng-
lish, Malagasy, and African and/or Asian languages) on the island itself.
This claim is supported by the fact that there is just one reference to a slave
who did not speak English sufficiently enough to stand in court. Sattoe
defended himself in English, and so did dozens of other slaves who had to
face trial from the 1780s onwards. All the slaves who appeared in court
were proficient enough (in English) to answer the Governor’s and/or the
jury’s questions. There are reports of bi- and/or multilingualism (the StHR
report that some slaves spoke Portuguese with each other, which, needless
to say, made them highly suspicious), and we also read that, as late as in
1762, a bilingual (Malagasy-English) slave was hired as an interpreter on
ships bound to Madagascar. Even though some members of the community
were bi- and/or multilingual, there is no doubt that English served as a
lingua franca on the island, and that language shift to English was com-
pleted in the second half of the 19th century at the very latest: “The Inhabi-
tants, who are of European, Asiatic and African origin, and whose only
language is English, are peaceable and kind, and are as independent as
their white brethren, with whom they intermarry” (Grant 1881: 10–11).
6. (Some first) conclusion(s): Non-standardisation and St Helenian

English
This paper has presented some preliminary observations on the history and
development of StHE, with the aims of providing a socio-historical profile
of the community and reconstructing the sociolinguistic situation on the
island between the 1660s and the late 18th century, thus pinpointing the
origins and evolution of StHE. Generally speaking, the social, historical
and sociolinguistic findings complement each other well. Though sketchy
and far from representative, the samples presented here show that the StHE
feature pool was quite diverse, containing donor varieties from England,
continental Europe, Africa and Portuguese colonies in the Atlantic Ocean,
as well as the Indian subcontinent and Asia. The language samples, what-
ever their nature, are supported by the community’s social, sociodemo-
graphic and historical development. Consequently, then, a historical analy-
sis must necessarily integrate all these pieces of information; with due
caution, reports of direct speech, samples of dialect writing, etc. are cer-
tainly useful for historical linguistic methodology.
Second, and more important here, what does all this mean for ‘non-
standardisation’ in English? First of all, we note that a particularly strong
input to StHE came from non-standard, working-class southeastern British
English, which was quite likely the most influential donor variety of all.
Most of the planter families came from the English Southeast, bringing
non-standard forms of English with them. This variety (or perhaps: these
varieties) were highly influential in the StHE feature pool, not only be-
cause of the (comparatively) sizeable number of speakers, but also because
they served as a reference and target variety for members of the community
whose native language was not English (and who spoke English-based
100 Daniel Schreier
jargons, pidgins and/or creoles, or entirely different languages, e.g. French

or Portuguese, along with rudimentary English). Consequently, then, StHE
was non-standard to start out with, simply because it largely derives from
non-standard English varieties brought to the island by members of the
English working class. Additional processes of linguistic contact (with
other European, African and Asian languages) were likely to influence the
development of StHE (the extent of which needs to be examined in more
detail; see Schreier 2008) and to have a catalysing effect on non-
standardisation (via all kinds of substratum effects, pidginisation, perhaps
creolisation, etc.). These processes mirror what happened in other colonial
settings (e.g. in the Caribbean or in the Southern Hemisphere), even
though, needless to say, the local contact scenarios manifested themselves
differently in each case.
Non-standardisation, then, is explained historically by the fact that in
virtually all colonial possessions, the majority of the British founders – and
those that followed in their wake – were speakers of non-standard varieties,
and that the influence of the standard variety was small (simply due to the
fact that there were few speakers of Standard English among the permanent
populace). Dialect (and even more so: language) contact thus contributes
heavily to the continuing spread of non-standard varieties throughout the
English-speaking world. As for total numbers, the speakers of StHE may
be no more than the metaphorical drop in the ocean of the English-
speaking world; sociolinguistically, however, they represent what has hap-
pened and is happening in so many other places. The diversification and
spread of English goes hand in hand with non-standardisation, certainly in
the Outer and Expanding Circles.
Acknowledgments
My thanks go to Nicole Joho and Claudia Rathore, who read and com-
mented on an earlier draft of the manuscript.
Notes
1. Unless this spelling represents a past participle form after all, as Dave Britain
points out to me (p.c. February 9, 2006).
References
Bex, Tony and Richard J. Watts (eds.)

Bonfiglio, Thomas P.
2003 Race and the Rise of Standard American. Berlin: Mouton de
Gruyter.
Brooke, Thomas H.
1808 A History of the Island of St. Helena, from its Discovery by the Por-
tuguese to the Year 1806; to which is Added an Appendix. London:
Black, Parry and Kingsbury.
Chambers, Jack K.
1995 Sociolinguistic Theory: Linguistic Variation and Its Social Signifi-
cance. Oxford: Blackwell.
Cheshire, Jenny and Dieter Stein (eds.)
1997 Taming the Vernacular: From Dialect to Written Standard English.
London: Longman.
Crowley, Terry
2003 Standard English and the Politics of Language. London: Macmillan.
Crystal, David
2003 The Stories of English. London: Penguin.
Gargen, Henry
1665 A Description of the Island of St. Helena to whom itt may concerne:
By mee Henry Gargen from ye year 1661 to ye yeare of our Lord:
1665 [Manuscript].
Gosse, Philip
1938 St Helena, 1502–1938. London: Cassell & Co.
Graddol, David, Dick Leith and Joan Swann
1996 English: History, Diversity and Change. London: Open Univer-
sity/Routledge.
Grant, Benjamin
1881 A Few Notes on St Helena. St Helena: Grant.
Hancock, Ian
1991 St. Helena English. In Development and Structures of Creole Lan-
guages, Francis Byrne and Thom Huebner (eds.), 17–28. Amster-
dam/Philadelphia: Benjamins.
Honey, John
1997 Language is Power: The Story of Standard English and its Enemies.
London: Faber.
Janisch, Hudson R.
1885 Extracts from the St. Helena Records. St Helena: Grant.
102 Daniel Schreier
Kachru, Braj B.
1985 Standards, codification and sociolinguistic realism: The English
language in the outer circle. In English in the World: Teaching and
Learning the Language and Literatures, Randolph Quirk and Henry
G. Widdowson (eds.), 11–36. Cambridge: Cambridge University
Press.
Kachru, B.B.
1991 Liberation linguistics and the Quirk concern. English Today 25 (7,
1): 3–13.
Leith, Dick
1983 A Social History of English. London: Routledge and Kegan Paul.
Lippi-Green, Rosina
1997 English with an Accent. London: Routledge.
Melliss, John C.
1875 St. Helena: A Physical, Historical, and Topographical Description
of the Island, including its Geology, Fauna, Flora, and Meteorology.
London: Reeve.
1985 Authority in Language. London: Routledge.
Montgomery, Michael B., Janet M. Fuller and Sharon DeMarse
1993 ‘The black men has wives and Sweet harts [and third person plural -
s] Jest like the white men’: Evidence for verbal -s from written
documents on nineteenth-century African-American speech. Lan-
guage Variation and Change 5 (2): 335–357.
Mufwene, Salikoko S.
1996 The founder principle in creole genesis. Diachronica 13: 83–134.
Mufwene, Salikoko S.
2001 The Ecology of Language Evolution. Cambridge: Cambridge Univer-
sity Press.
Mühlhäusler, Peter
1997 Pidgin and Creole Linguistics (2nd edition). London: Battlebridge
Press.
Quirk, Randolph
1990 Language varieties and standard language. English Today 21 (6, 1):
3–10.
Rosenberger, Lukas Michael
2006 The Swiss English Hypothesis: The Linguistics of English in Swit-
zerland. Ph. D. thesis, English Department, University of Berne.
Sankoff, Gillian
2006 Age: Apparent time and real time. In Encyclopedia of Language and
Linguistics, Vol. 1, 2nd ed., Keith Brown (ed.), 110–116. Amsterdam:
Elsevier.
Schneider, Edgar W.
2001 Investigating variation and change in written documents. In The
Handbook of Language Variation and Change, J.K. Chambers, Peter
Trudgill and Natalie Schilling-Estes (eds.), 67–96. Oxford/Malden,
MA: Blackwell.
2003 The dynamics of New Englishes: From identity construction to dia-
lect birth. Language 79 (3): 233–281.
Schneider, Edgar, Bernd Kortmann, Kate Burridge, Rajend Mesthrie and Clive
Upton (eds.)
2004 A Handbook of Varieties of English. Vols. 1–2. Berlin: Mouton de
Gruyter.
Schreier, Daniel
2008 St Helenian English: Origins, Evolution and Variation. Amster-
dam/Philadelphia: Benjamins.
Trudgill, Peter
1972 Sex, covert prestige and linguistic change in urban British English.
Language in Society 1 (1): 179–195.
Watts, Richard J.
ment of Standard English 1300–1800, Wright, Laura (ed.), 29–48.
Wells, John
1982 Accents of English (3 volumes). Cambridge: Cambridge University
Press.
Wright, Laura (ed.)
2000 The Development of Standard English 1300–1800. Cambridge:
Wright, Laura
2004 Some data from slave speakers: The island of St Helena, 1695–1711.
Paper presented at the Westminster Creolistics Workshop: Dia-
chronic Studies and Theories of Creolisation, London, UK.
Chapter 6
From ‘standard’ to ‘nonstandard’ grammar.
New England in the days of Salem Witchcraft and
the Civil War
Adrian Pablé*
1. Scope and aim
Several grammatical features deemed unacceptable in present-day formal

Standard American English were taken to the early colonies by British
emigrants, who used them regardless of their social status or the linguistic
medium. The relegation of most of these traits to the level of ‘vulgar lan-
guage’, and, concomitantly, their banishment from cultivated speech, very
likely coincided with the early federalist period following the Revolution-
ary War and the War of 1812, when specific variants became subject to
‘focusing’ while their counterparts were assigned ‘dialectal’ status (see
e.g., Cooley 1992). The shift in social, i.e., diastratic, evaluation that some
of these features underwent in the course of the late eighteenth and early
nineteenth centuries has been commented on repeatedly in the scholarly
literature: George P. Krapp (1925: 261), for example, discussed the vari-
ants were and was in second person singular contexts, and presented con-
vincing evidence that the less general form you was constituted “the better
and finer usage” in eighteenth century America. Analogously, Krapp
(1925: 263) pointed out that adjectival them was formerly found in “repu-
table writings” and that ain’t was not thought “reprehensible in colloquial
style”. Alexander (1928: 148), in an article dedicated to the language of
colonial American English, expressed the notion mentioned above as fol-
lows: “Many so-called vulgarisms are found to have their roots far back in
the history of our language and to have graced the lips of distinguished
people in a past that is often not very remote”.
In the present article I will discuss a selection of grammatical features
and their distribution along linguistic contexts. These features are all at-
tested in early colonial New England as well as at the time of the Civil
War: however, none of them carried a social stigma in the former period,
while they clearly did so in the latter (as becomes clear from the comments
106 Adrian Pablé
on ‘provincialisms’ and ‘vulgarisms’ by early nineteenth century gram-

marians [e.g., Kirkham 1834]). The features under discussion are the fol-
lowing:
– Finite indicative be
– Nonstandard was
– Nonstandard verbal -s
– Unmarked present and past tense verbs
– don’t in third person singular contexts
– Levelling of participle forms
– Unmarked plurals
– them used pronominally and as an adjective
– Adjectives as modifiers of adjectives and verbs
– a-prefixing
– Subject relative pronoun deletion
My analysis relies on the scrutiny of selected texts drawn from two

sources: the Salem Witchcraft Papers, a collection of colonial court records
(Boyer and Nissenbaum 1977)1, and Civil War correspondence, in fact
letters exchanged between New England soldiers and the home front (Sil-
ber and Sievens 1996).2 I am aware of comparing data retrieved from two
different text genres, but, as I will specify below, the texts chosen for in-
vestigation are exclusively those with a ‘speech-reflecting’ quality.
2. New England’s nonstandard speech
Traditional New England folk speech, including many of its characteristic

grammatical features, had practically disappeared from the southern New
England states (Massachusetts, Connecticut, Rhode Island) by the early
twentieth century, while the largely rural northern states (New Hampshire,
Vermont, Maine) preserved it for a longer period. In his Word Geography
of the Eastern United States, Hans Kurath (1949: 5–9) attributed the de-
cline of New England’s folk speech to several interdependent factors, all at
work since the early colonial days: (1) the steady development of “an intel-
ligent and active middle class”, speakers of what Kurath calls “the common
speech”, whose forefathers – English farmers and craftsmen – had emi-
grated to New England as literate men; (2) compulsory school attendance,
even though rarely up to the age of fourteen; (3) protracted exposition to
literary English outside the school context: “The [early] New Englander
From ‘standard’ to ‘nonstandard’ grammar 107
read the Bible or had it read to him in his own home or that of a neighbor
until he knew much of it by heart” (Kurath 1949: 5). Bearing this in mind,
as well as the fact that church attendance and catechism were obligatory, it
may well be assumed that New Englanders grew up in a milieu at once
literate and speech centered. However, universal literacy did not lead to the
demise of New England folk speech; Kurath gives the following convinc-
ing explanation for this:
The small farmer, the farm tenant, the fisherman, the mountaineer learns his
language by ear. A person of this class reads little, if at all. He does his
shopping the year round in the same general store where he meets his like to
talk to. His occasional visits to […] the village or the county seat, have no
appreciable effect upon his habitual speech ways […] his pronunciation and
grammar are little affected by these contacts. In the local school he gets to
know the written language, but if he settles back into the simple life of his
forbears he will continue to speak much like his family and his neighbors of
an older generation … . (Kurath 1949: 8)
In the Linguistic Atlas of New England (Kurath et al. 1939–43) one can
catch a glimpse at what must have been New England’s traditional folk
speech around the late nineteenth century, as reflected in the language of
some octogenarian informants (in fact, those classified by the fieldworkers
as ‘old-fashioned’). Unfortunately, in Kurath’s Atlas relatively little space
was devoted to the regional folk grammar, the focus being clearly on pho-
netic and lexical variation. The present study thus intends to fill this void
by analysing primary evidence of New England’s (spoken) grammar pre-
dating the Atlas.
3. Sources and text genres relevant to the present study
3.1. Salem Witchcraft Papers
The importance of the Salem Witchcraft Papers (Boyer and Nissenbaum

1977) as a treasury for the study of the speech of colonial New England
was recognized early and has lead to the publication of several scholarly
articles (e.g., Alexander 1928; Rissanen 1997, 2003; Kytö 2004; Grund,
Kytö and Rissanen 2004).3 Of particular significance is the fact that the
Salem manuscripts were written 70 years after the first settlement of New
England, and hence offer an insight into a new regional variety spoken by
the first generation of genuine New Englanders:
108 Adrian Pablé
The records give us what is probably the best and most complete picture of
the popular language as spoken in New England at this period [...] The date
of the witchcraft trials, 1692, is sufficiently long after the first settlement of
New England for us to assume that the language indicated is really the
speech of American settlers and not merely that of recent arrivals from Eng-
land. This is made all the more certain because a good deal of the testimony
against the supposed witches was given by rather young people. (Alexander
1928: 390–391)
Alexander’s statement suggests that the history of New England English is
comparable to the evolution of Southern Hemisphere Englishes, to which
dialectologists have devoted considerable attention lately. Peter Trudgill,
who has written extensively on the formation of new dialects out of so-
called ‘tabula rasa’ situations, argues that a new unitary dialect resulting
from dialect mixture and levelling takes about 50 years to come into being,
i.e., two generations (Trudgill 2004: 23). Trudgill considers the children of
the children arriving with their parents at the time of the first settlements to
be the artificers of the new ‘homogeneous’ variety; these children, in fact,
will diverge from their parents’ speech patterns by converging to the pat-
terns of their peers, thereby establishing what variant will be chosen as the
new (categorical) form. What forms are ultimately selected will depend on
such factors as the demographic strength of a particular dialect variety and
the markedness of a variant (Trudgill 2004: 34–35). The situation in the
seventeenth century New England colonies mirrors what happened – two
centuries later – in Australia and New Zealand: in fact, Wakelin (1986)
convincingly argues in favor of a linguistic heterogeneity that must have
characterized Plimoth Plantation in the 1620s. By the time of the Salem
witchcraft trials in 1692, it can be assumed that the second and third gen-
eration colonists no longer spoke regional varieties of British English but a
veritable colonial variety, i.e., the earliest form of New England English.
The colonial records taken into account for the present study are (a)
transcripts of court examinations, in which speech is generally rendered in
dialogue format, and (b) testimony deposed by witnesses, which are written
down as indirect discourse, or mixed discourse (e.g., ‘He testifieth and
saith that on April 18 I did see …’). Overall I have chosen to analyze 155
such ‘speech-related’ texts from the Salem Witchcraft Papers (58 examina-
tions, 97 witness accounts), containing the sociolects of New Englanders of
very different educational backgrounds. In view of the ‘standard–
nonstandard’ framework adhered to in this article, I have focused on the
direct discourse of both educated Salemites, i.e., the magistrates who pre-
sided over the trials, and less educated ones, as taken down by the various
scribes.4
3.2. New England ‘vernacular’ letters
Written documents may, under certain circumstances, contain language

that is very close to the spoken, even if we must bear in mind that written
language is always one or more times removed from speech, regardless of a
writer’s educational background. As a matter of fact, the language con-
tained in some of the mid-nineteenth century New England letters in-
spected for the present study was found to digress considerably from what
must have been the conventional epistolary style of the time. These non-
standard linguistic elements are assumed to be the result of the intrusion of
speech habits upon the written language. In several of his publications,
dialectologist Michael Montgomery (e.g., 1997, 2005; Schneider and
Montgomery 2001) has classified these kinds of texts as “vernacular” or
“colloquial” documents, i.e., writings by individuals of lower social sta-
tions composed in the least self-conscious style. Montgomery (1997: 229)
describes three types of such individuals as well as the situations most
likely conducive to producing ‘vernacular’ language in writing:
If we can identify those individuals who were separated from loved ones
[‘lonelyhearts’], were in desperate straits and needed help [‘desperadoes’],
or were required to submit periodic reports [‘functionaries’], we may be on
the path to locating the letters of greatest linguistic interest […] these situa-
tions are compelling enough to motivate individuals to write for themselves,
to do their best in putting words to paper regardless of their level of literacy.
In other words, someone pleading for mercy or relief may well pay little at-
tention to the form (spelling, capitalization, grammar, etc.) of his or her writ-
ing, being more concerned with getting an unambiguous message across.
The extralinguistic situation underlying Civil War letters exchanged be-
tween Yankee Union soldiers and their families qualify as such ‘colloquial’
documents, as, addresser and addressee being often on intimate terms, they
fostered the recourse to informal, i.e., ‘close-to-oral’, language. Faced with
the prospect of either imminent death (the soldiers) or insolvency (the be-
loved ones remaining behind), we must assume that letter writers were not
particularly heedful of ‘correct’ writing; as transpires from the contents of
these letters, in fact, many of them were written hastily and under great
emotional strain. A caveat seems to be necessary at this point, however: we
110 Adrian Pablé
must always bear in mind that letters written by soldiers staying in a differ-
ent dialect area for a longer period (as was the case with New Englanders
stationed in the South), some of whom may have been in daily contact with
comrades in arms of different dialect backgrounds, are likely to contain
dialect ‘contaminations’.
For my mid-nineteenth century corpus I have inspected 126 letters ed-
ited by the historians Nina Silber and Mary Beth Sievens (1996): of these I
have selected 53 letters providing clues to New England’s contemporary
folk grammar. The presence of both unorthodox spellings and grammatical
features regarded as ‘nonstandard’ in those days served as the criteria un-
derlying text selection.5 A few of these letters contain apologetic remarks
on the part of the writer, which shows that he/she was conscious of not
adhering to linguistic and textual norms (italics added):
I shall hav to close for want of paper please excuse this bad righting and
right soon (Ellen Horton, of Chittenden/Vermont, writing to Edwin Horton,
1864)
Excuse all mistakes you know I write so seldom we all send you our love to
you (Sarah Fales, of Middletown/Rhode Island, writing to her son Edmund,
1862)
4. Grammatical features of late seventeenth and mid-nineteenth

century New England speech
4.1. Verb forms
4.1.1. Finite indicative be
In relation to New England, the verb form usually referred to in the litera-
ture as respectively ‘invariant be’ and ‘finite be’ functions as a substitute
for the suppletive verbs am, is and are (both copula and auxiliary), and
should be kept distinct from subjunctive be (even though the division is not
always that clear-cut, especially in colonial usage). This particular feature
was included in the Linguistic Atlas of England (see e.g., Atwood 1953:
27), however without any mention of its status in a crossdialectal perspec-
tive; in fact, Pablé and Dylewski (2007: 175) found that be in nineteenth
century New England folk speech differed considerably from be as recur-
rent in (present-day) African American Vernacular English (e.g., Bailey
and Bassett 1986) and other ethnic and regional varieties of American Eng-
lish. Again, this is confirmed in the data inspected here: the Salem Witch-
craft Papers feature a handful of tokens of the verb form at issue, with be
occurring preponderantly within interrogative clauses and in sentence-final
position, i.e., in positions not allowing verbal contraction (1–3), and not in
unstressed sentence-medial position (as is the case in African American
English):
(1) (Magistrate:) Mary you are accused of sundry acts of witchcraft: Tell me be
you a Witch? (Examination of Mary Black)6
(2) (Magistrate:) Where be those images, at your house? (Examination of Deliv-

erance Hobbs)
(3) (Clement Coldum:) She asked me to ride faster, I asked her why; she said the
woods were full of Devils, & said ther & there they be, but I could se none
(Clement Coldum vs. Elizabeth Hubbard)
I have found only two attestations of be in the Civil War letters (4–5). This
is unexpected because, on the one hand, the verb form was a prototypical
(and thus stereotypical) feature of nineteenth century New England folk
speech, as suggested by many contemporary literary portrayals and by the
evidence in the Linguistic Atlas of New England, but on the other hand it is
corroborating: in fact, Montgomery and Mishoe (1999: 261) propose that
occurrences of be in vernacular documents of the postcolonial period are
rare due to “pragmatic constraints”, such as being associated with spoken
interaction exclusively and therefore shunned in writing. Again, the two
instances of be occur in syntactically ‘strong’, i.e., non-contractable, posi-
tions (see Ihalainen 1991):
(4) I feel better about you where you be than I should if you were here and acted
as will cameron does… (Ira Morse, of Woodbury/Vermont, writing to her son
Franklin, 1862)
(5) …for i think that if you knew the particilars of things as they be you would
feel different about somethings than you now do (Hulda Morse, of Wood-
bury/Vermont, writing to her husband, 1864)
112 Adrian Pablé
4.1.2. Nonstandard was
Nonstandard was is a recurrent feature in both text collections inspected,

though still clearly the minority variant with respect to were. In the Salem
Witchcraft Papers plural was in nonstandard positions co-occurs with exis-
tential clauses (6), conjoined noun phrases (7), and, in a scattered fashion,
with simple noun phrases (8), including pronominal they (9):
(6) (Magistrate:) how many was there in all? (Examination of Richard Carrier)
(7) (Simon Willard, scribe:) the afflicted sayd Dan’ll Eames & Capt. Floyd was
upon that cloth when it was upon the table (Examination of Abigail Faulkner,
Sr.)
(8) (Simon Willard, scribe:) but Jaxon s’d he was at work at Cap. Wicoms of
Rowley last night he s’d these persons was not in their right mind (Examina-
tion of John Jackson, Sr.)
(9) (Magistrate:) was they to Sett up the divels Kingdome? (Examination of

Elizabeth Johnson)
In conjunction with singular you, on the other hand, was occurs regularly
within the discourse of the Salem magistrates; they alternate as a rule be-
tween you were and you was within one and the same interrogation (10):
(10) (Magistrate:) was you not at Mr Ballards house on thanksgiving day at night?
[…] were you Ever baptized? (Examination of Ann Foster, Mary Lacey Sr.,
and Mary Lacey, Jr., Second Version)
A noteworthy difference between the colonial records and the Civil War
letters is linked to the fact that in the latter corpus nonstandard was occurs
more frequently in conjunction with pronominal subjects, especially the
pronouns we and they (11–12); as might be expected, nonpronominal sub-
jects (i.e., zero, nominal) are equally found to attract nonstandard was in
the writings of mid-nineteenth century New Englanders (13–14), but not to
any greater extent than personal pronouns:
(11) They said that they was very short of coffee and salt and clothing (George
Bradley, Connecticut, writing to his friends at home, 1862)
(12) We dont Know whare we will be in the morning for when we was encamped
on miners hill they had us up every night (Reuben Thornton, of War-
wick/Rhode Island, writing to his friend George, 1863)
(13) Those that was drafted in Woodbury payd 300 (George Morse, of Wood-
bury/Vermont, writing to Franklin Morse, 1863)
(14) I think we have wrote enogh that you might know what our minds was (Ira
Morse, of Woodbury/Vermont, writing to Franklin Morse, 1864)
4.1.3. Nonstandard verbal -s
In the colonial records only three instances of nonstandard -s/-th were

found (15–16), notably in third person plural contexts in conjunction with
nonpronominal subjects; plural -s occurs once as part of a magistrate’s
discourse (15):
(15) (Magistrate:) how long hath your Mast’r & Mistris bin Witches? (Examina-
tion of Mary Warren)
(16) (Deliverance Hobbs:) There are a great many persons hurts us all (Examina-
tion of Deliverance Hobbs)
In the Civil War letters, the -s inflection in nonstandard contexts was found
to occur in nonpronominal environments – i.e., in relative clauses (17), one
of which in conjunction with a suppressed relative subject (as in example
16) – but also in pronominal ones (18):
(17) their is a good many comes into our lines every night (Edwin Horton, of Chit-
tenden/Vermont, writing to Ellen Horton, 1865)
(18) …and I myself has roat him a great many letters to send to you (Thomas
Nickerson, Rhode Island, writing to Ellen Bullock, 1862)
4.1.4. Unmarked present and past tense verbs
Uninflected third person singular present indicative forms are by no means

frequent in the Salem Witchcraft Papers. They are mostly encountered as
part of the contemporary ‘non-speech-based’ legalese, in particular in con-
114 Adrian Pablé
junction with periphrastic do. As far as full verbs are concerned, the zero
form and the inflected form can be found within the same record, typically
with reporting verbs. My impression is that the Salem scribes shifted arbi-
trarily between three variants in third person singular contexts when intro-
ducing reported discourse, namely zero, -s and -th (19–20):
(19) (Scribe unmentioned:) Elizabeth Clark who then was the wife of the aboves’d
Nathaniell Clarke doe testifie that … (Elizabeth Clark v. Susannah Martin)
(20) (Simon Willard, scribe:) He conffess that Yesterday he afflicted martha
Sprauge […] he Conffesses also that he and the Rest of his Comp’y did
daunce at Moses tylers House ... (Examination of Stephen Johnson)
Uninflected forms in third person singular contexts within ‘spoken’ dis-

course are also infrequent in the colonial documents: the most prominent
context attracting this particular type of unmarked verb, and especially the
auxiliary have, is the affirmative clause with a zero or non-adjacent subject
(21). I have found only two instances of uninflected present indicative non-
auxiliary verbs within ‘spoken’ discourse (22–23):
(21) (Joseph Hutchinson, landholder:) Abigaill Williams I have heard you spake
often of a booke that have bin offred to you (Joseph Hutchinson v. Abigail
Williams)
(22) (Elizabeth Symonds:) I have bin at the Docters but thay cannot give mee any
thing that doe mee any good (Elizabeth Symonds v. Sarah Wilds)
(23) (Mary Warren, servant maid:) I said Master, w’t make you say soe? (Exami-
nation of Mary Warren, May 12)
In third person singular contexts, the majority of unmarked verbs in the

Salem Witchcraft Papers are to be classified as past tense forms: the most
recurrent ones belong to the category of irregular verbs (e.g., come, run,
see, speak, break, shake, say) (24–25), while there are also scattered tokens
of ‘weak’ preterites (26). Obviously, the phenomenon at issue concerns all
person-number combinations:
(24) (Magistrate:) w’t the Divel Come to You did he not? (Examination of Eliza-
beth Johnson)
(25) (Abraham Wellman:) after this time when she see any person coming to milk
her she would run … (Abraham Wellman v. Sarah Cole)
(26) (Ann Putnam, Jr.:) I then saw that she was the very same woman that tould me
hir name was Mis Bradbery and she most greviously afflect and torment me
dureing the time of hir examination... (Ann Putnam, Jr. v. Mary Bradbury)
In the Civil War letters zero-marked third person singular verbs are rare
overall; it deserves notice that those with a present tense value express
either an iterative/habitual meaning (27–28)7, or stativity (29):
(27) It is a treat for Sarah to write she enjoy it much she will write soon (Sally
Fowler, of Danversport/Massachusetts, writing to Henry Fowler, 1862)
(28) I know that it take a long time for a letter to come but if you write often we
shall get them often … (Benjamin Morse/Vermont, writing to Rosina Morse,
1862)
(29) Forty Company in all. It look quite warlike here Cos. drilling all the while…
(John Norton, of Hartford/Connecticut, writing to Lissie Norton, 1861)
I also spotted two unmarked verbs with a past tense value, namely see and
run, both in a third person singular context (30–31):8
(30) I expect Hannah thinks it is a long time since she see her (Mattie Blanchard,
of Foster/Connecticut, writing to Caleb Blanchard, 1862)
(31) after getting off we continued up the river, when the celebrated “Hillins Ram”
run for us, striking our port quarter (W. H. Robert, Rhode Island, writing to
his wife, 1862)
4.1.5. Don’t in third person singular contexts
Only two instances of third person singular ‘do not’ are attested in the Sa-
lem Witchcraft Papers, in fact once in conjunction with a preposed pro-
noun subject (32) and once with a non-adjacent preposed pronoun subject
(33):
(32) (Robert Pike, scribe:) He farther say that he do not know any such pitt to be in
the place that he was sliding into (John Pressy v. Susannah Martin)
(33) (Joseph Putnam, scribe:) Why did she pinch the young woaman shee never
did nor dont know who did (Examinations of Sarah Good, Sarah Osborne, and
Tituba)
116 Adrian Pablé
In the Civil War letters nonstandard don’t occurs more frequently, which
deserves notice given the fact that the latter corpus is considerably smaller;
also, the linguistic contexts attracting don’t are different ones than in the
witchcraft records. Thus, we find the form under scrutiny in conjunction
with adjacent subjects, both pronominal (34) and nominal (35):
(34) if that dont disharten him put half Bushel of corn on his back and march all
day … (Charles E. Jewett, of Gilford/New Hampshire, writing to his brother
and sister, 1862)
(35) time dont seem as long there as it does here (Edwin Horton, of Chitten-
den/Vermont, writing to Ellen Horton, 1864)
4.1.6. Levelling of participle forms
In the colonial records, the phenomenon of levelling occurs with irregular

verbs displaying non-agreement between preterite and participle in present-
day Standard English: thus, the verbs write, tear, speak, break, shake, and
beat are variably treated by the colonists like verbs of the type ‘buy-
bought-bought’, e.g., ‘write-wrote-wrote’. Some of these levelled partici-
ples are part of the magistrates’ direct discourse (36–37):
(36) (Magistrate:) Why if you have not wrote in the book, yet tell me how far you
have gone? Have you not to do with familiar Spirits? (Examination of Bridget
Bishop, First Version)
(37) (Reverend Samuel Parris, scribe:) Mary Walcot said that her brother Jonathan
stroke her appearance & she saw that he had tore her coat in striking … (Ex-
amination of Bridget Bishop, First Version)
In the Civil War letters, wrote is the most prominent nonstandard participle
(38); this very participle was found to be recurrent throughout the New
England states at the time of the Linguistic Atlas of New England (see At-
wood 1953: 26). Other levelled past participles encountered in the vernacu-
lar letters inspected are fell, tore, drove and froze (39):
(38) I dont want you should feel hard towards me for what I have wrote for I mean
well … (Caleb Blanchard, of Foster/Connecticut, writing to his wife Mattie,
1863)
(39) …our Mizzen Mast shot away, rigging dammaged, and boats tore in Pieces,
but most fortunately, only two killed … (W.H. Robert, of Rhode Island, writ-
ing to his wife, 1862)
4.2. Nouns, pronouns and modifiers
4.2.1. Unmarked plurals
In the Salem Witchcraft Papers absence of the plural marker -s is a phe-

nomenon confined to the following nouns: year, month, day, night, mile,
rod and pole (40–41). The scribes are sometimes found to alternate be-
tween marked and unmarked forms within one and the same document;
based on this, it is tempting to think that zero-marked plurals may in fact
have been due to scribal negligence. However, the elevated number of to-
kens found suggests that such variants as year and years were rather in free
variation within colonial discourse:
(40) (Magistrate:) how maney year is it since you had familliarity with the divil?
(Examination of Mary Lacey Sr. and Jr., Ann Foster, Richard & Andrew Car-
rier)
(41) (Scribe unmentioned:) …that he would kil 10 folck in boston before next six
day … (Susannah Sheldon v. Sarah Proctor et al.)
I have spotted only two unmarked plural nouns in the Civil War letters
(42–43), even though from the Linguistic Atlas of New England (Kurath
1939–44: M83) it appears that at least the plural noun year was still com-
mon among the elderly ‘noncultured’ informants in the 1930s:
(42) Tomorrow you will be 23 year old (Sally Fowler, of Danversport / Massachu-
setts, writing to her husband, 1862)
(43) I doant know what they will do when they come to hav an addition to there
family and they will ere the corse of seven month if nothing happenes (Ellen
Horton, of Chittenden/Vermont, writing to her husband, 1864)
118 Adrian Pablé
4.2.2. The demonstrative pronoun them used pronominally and as an

adjective
In the colonial records there are scattered instances of demonstrative them

used either as an adjective (44) or as a pronoun (45), the latter especially in
contexts involving a relative clause (e.g., ‘here are them that accuse you’):
(44) (Simon Willard, scribe:) Mary Warrin: was asked if this was one of them men
she saw the other night (Examination of John Jackson, Jr.)
(45) (Scribe unmentioned:) & shee said shee Came to see what Lies them were that
we raisd of hur…and as soon as Ever she s’d Rachall Herd them words: she
run out of Dors (James Fuller, Jr., v. Rachel Clenton)
In the Civil War letters, them used as an adjective and a pronoun each oc-
cur once (46–47). It is, however, highly likely that the feature at issue was
a frequent one in the contemporary folk speech, judging from the fact that
its status is that of a vernacular ‘angloversal’:
(46) I saw them Troops from Boston that were attacked in Baltimore shook hands
with a great many of them … (John Norton, of Hartford/Connecticut, writing
to Lissie Norton, 1861)
(47) and those who counsel it are no better than Jeff Davis and all his friends,
thems [them is] my sentiments … (Edward Hall, of Exeter/New Hampshire,
writing to Susan Hall, 1863)
4.2.3. Adjectives as modifiers of adjectives and verbs
A recurrent pattern in the witchcraft records involves the use of an adjec-

tive as a modifier of either an adjective (48) or, less prominently, of a verb
(49):
(48) (Mercy Lewis:) mr Burroughs caried me up to an exceeding high mountain

and shewed me all the kingdoms of the earth (Mercy Lewis v. George
Burroughs)
(49) (Simon Willard, scribe:) she s’d she would for god was a god of truth: & she
presently spoke very hoars (Examination of Jane Lilly and Mary Colson)
In the Civil War letters the use of adjectives as adverbs parallels that of the
colonial records (50–51):
(50) …and I hope that there will not be an other such a thing to occur…for it cre-
ates a dredful bad feeling amoung us (Thomas Nickerson, of Rhode Island,
writing to Ellen Bullock, 1862)
(51) Emma is so large now she helps considerable (L. W. White, of Royal-
ston/Massachusetts, writing to her husband Stephen White, 1864)
4.3. Syntactic structures
4.3.1. a-prefixing
In the Salem Witchcraft Papers the a-prefix appears in the same contexts as
the ones described by Walt Wolfram (1980) for twentieth century Appala-
chian English, namely in conjunction with progressives in which the auxil-
iary is either overtly realized or omitted (as illustrated in example 52).
Verbs of perception (especially see and hear) also attract the a-prefix on -
ing forms without an overt form of be (53).9 Besides appearing on -ing
forms functioning as a complement of perception verbs, the a-prefix also
occurs with other types of adverbial constructions (as in example 54). It
deserves notice that one instance found in the colonial records may violate
the phonological constraint described by Wolfram (1980: 125–126) for
Appalachian English, namely that the a-prefix only gets attached to mono-
syllabic verbs and those plurisyllabic verbs stressed on the first syllable;
the latter condition may not have been fulfilled in the case of the verb to
discourse (55):
(52) (Samuel Wilkins:) ...as i was a weaveing the yarn broak exceeding fast: and as
I was a tying a thread I had a stroak on my hand … awhile after as I was once
in the woods and a goeing hom … (Samuel Wilkins v. John Willard)
(53) (James Carr:) I hard sumting a coming to me againe (James Carr v. Mary
Bradbury)
(54) (Samuel Barton:)…saith that I being at Thomas putnams a helping to tend the
aflickted folks I heard them talking … (Samuel Barton and John Houghton for
Elizabeth Proctor)
120 Adrian Pablé
(55) (Priscilla Chub:) sum time last winter I was a discoursing with abigail hobs
her wicked cariges and disobedience to hir father and Mother (Priscilla Chub
v. Abigail Hobbs)
In the Civil War letters, a-prefixing is a phenomenon confined to monosyl-

labic verbs. It is found with progressives whose auxiliary is overtly real-
ized, and especially in conjunction with go having a future meaning (56);
in the colonial records, in turn, a-going was found to be a main verb exclu-
sively. I have also spotted instances of a-prefixed ing- forms that lack the
auxiliary and with an adverbial function (as in example 54 above); notably,
in my nineteenth century example the competing (‘standard’) variant ap-
pears in one and the same sentence (57):
(56) whether there is agoing to be any thing more to it I do not know … (Edwin
Horton, of Chittenden/Vermont, writing to his wife, 1863)
(57) I am up stairs in the bed room a riting and celia is out in the other room riting
a letter to you for orlando (George Morse, of Woodbury/Vermont, writing to
Franklin Morse, 1862)
4.3.2. Subject relative pronoun deletion
In the Salem Witchcraft Papers subject relative marker deletion is primar-

ily a phenomenon affecting existential clauses as well as it-cleft sentences
(58–59); semantically weak verbs (60) and structures of the type ‘NP-
relative clause-VP’ (61) also attract the structure at issue – though clearly
less frequently so:
(58) (Magistrate:) we now further enquire: here is one person saith you brought
her the book (Examination of Ann Pudeator)
(59) (Abigail Faulkner, Sr.:) she s’d yes but it is the devill dos it in my shape (Ex-
amination of Abigail Faulkner Sr.)
(60) (Nehemiah Abbot:) I had a kow was so weak and lam that she could not go
with out the halp of thre or fouer men … (Nehemiah Abbot v. Elizabeth How)
(61) (Scribe unmentioned:) The rags being put into water, two of the aforenamed
persons were in dreadful fits almost choaked, and the other was violently run-
ning … (Examination of Candy)
As already noted with regard to the colonial documents, the omission of

the subject relative pronoun in the Civil War letters is most prominent in
the context of existentials (62). One example also involves a semantically
weak verb (63). Notably, constructions of the type ‘two of the aforenamed
persons were in dreadful fits almost choaked’ (see example 61 above) do
not occur:
(62) Their has been a great number of convertions I attended one eve their were a
number spoke and told what the Lord had done for them (Eunice Snow, of
Greenwich/Connecticut, writing to her son Henry, 1864)
(63) We had one ball took the legs of three boys at one time (George Sargent, of
Manchester/New Hampshire, writing to his brother, 1861)
5. Comparing the two historical varieties under scrutiny
The feature discussion of the preceding sections suggests that the spoken
grammar of New England English was fundamentally the same in the sev-
enteenth and the nineteenth centuries: the number of shared grammatical
traits presented in the article corroborate this hypothesis. What did change
in the course of the late eighteenth/early nineteenth century was the social
evaluation – from positive or neutral to negative – underlying these gram-
matical forms and structures. The fact that the Salem magistrates suppos-
edly used features highly stigmatized in our days constitutes evidence in
favor of the sociolinguistic development outlined above. The counterargu-
ment that the forms and structures used by the magistrates could be attrib-
utable to scribal error is unconvincing, on political-historical grounds: in
fact, the court’s clerks were employed by the official authorities, who in
turn were keen to cleanse the colony once and for all of ‘God’s foes’; keep-
ing in mind the sacred cause of these trials, it is sensible to assume that the
Salem scribes must have made every effort to portray the state officials
acting as examiners favorably – among other things, by avoiding putting in
their mouths language traits negatively connoted at the time.
It has also transpired on several occasions that the linguistic contexts at-
tracting the nowadays stigmatized variant are not congruent when colonial
and postcolonial data are compared: hence, in nineteenth century New
England folk speech the range of contexts allowing the nonstandard variant
seems to have been wider (with respect to the colonial variety): for in-
stance, nominal subjects (in particular conjoined nouns) no longer attract
122 Adrian Pablé
plural was to a higher extent than pronominal subjects, and nonstandard

don’t also occurs in conjunction with single nominal subjects. In the seven-
teenth century, collocations like they was and we was were clearly minority
variants, while, to my knowledge, phrases combining a noun subject with
indicative don’t (e.g., ‘time do not’) have not been attested so far in Early
Modern documents. In a similar vein, Bailey and Ross (1988: 199–200)
confirm the results obtained from scrutinizing the Salem Witchcraft Pa-
pers, namely that in Early Modern English the -s/-th marker was disfavored
in the context of a personal pronoun other than he/she/it (e.g., ‘I takes’),
which was no longer the case in postcolonial varieties of nonstandard
American English. On the other hand, it turns out that narrowing of linguis-
tic contexts was also involved in the passage from colonial New England to
postcolonial New England English: thus, subject relative marker deletion
applied to a wider range of contexts in colonial English than in nineteenth
century New England folk speech. In fact, colonial constructions of the
type ‘two of the aforenamed persons were in dreadful fits almost choaked’
(see example 61) are certainly not characteristic of nineteenth and early
twentieth century spoken/colloquial language, possibly owing to a lack of
transparency as to which verb is embedded and which is not.
It deserves mentioning that some of the features found in the colonial
records are attested in several nineteenth century literary portrayals of the
rustic New England dialect, but do not occur in the Civil War letters inves-
tigated: the absence of specific oral features from ‘colloquial’ documents
can be explained as follows: (a) the form is avoided in writing because it is
overtly stigmatized; (b) the form is primarily used in conversation, which is
why pragmatic constraints apply. One such feature, besides finite indica-
tive be, is and in absolute clauses, which occurs occasionally in the Salem
Witchcraft Papers (64–65), and of which we have ample attestations in
nineteenth century dialect literature:
(64) (Magistrate:) it is very awfull to all to see these agonies & you an old pro-
fessor thus charged with contracting with the devil … (Examination of Re-
becca Nurse)
(65) (Samuel Wilkins:) awhile after as I was once in the woods and a goeing hom
& a little boy with me. I thought I must run and I said … (Samuel Wilkins v.
John Willard)
Last but not least, it must not be forgotten that in spite of the corroborating
evidence suggesting that there is a common grammar underlying colonial
and postcolonial New England speech, some grammatical features are

found to have been characteristic of only one period: hence, the preterites
seen and done, which did not enter what Mencken (1948: 353–354) refers
to as the ‘common speech’ until the late eighteenth century, occur repeat-
edly in the Civil War letters, but are in fact absent from the Papers. In turn,
the Salem records feature a considerable number of structures that became
obsolete in the course of the eighteenth century, which is why they no
longer appear in nineteenth century writings (i.e., neither in vernacular
letters nor in fictional portrayals): among them are be as a perfect auxiliary
(e.g., ‘you are become a tormentor’), the analytic possessive (‘Martha Car-
rier her malice’), the transitive use of to discourse (meaning ‘converse
with’), the topicalization of pronoun objects (‘it he did himself’), whether
as a mere sign of interrogation introducing a direct question (‘Whether you
did not know that it was the Devils book when you signed?’).
In his Dissertations on the English Language, dating from 1789, Noah
Webster (1967: 384, 389) praised the “New England common people” for
still using “ancient English phrases” and for having an “idiom [that] is
purely Saxon or English”. On the other hand, he conceded that they “often
make false concord”. The present article has for the most part considered
precisely the latter phenomena in order to show that ‘false concord’ was by
no means a marker of low education among early New Englanders, but
rather reflected ‘ancient use’. This, the prescriptivist Webster was not will-
ing to admit.
Notes
*The author would like to thank Radoslaw Dylewski, Miriam Locher and Jürg
Strässler for their valuable comments made on the first draft of this paper.
1. For the present paper I have availed myself of the electronic version of Boyer
and Nissenbaum (1977). For critical comments on Boyer and Nissenbaum’s
edition, see Grund, Kytö, and Rissanen (2004).
2. As the two corpora are of unequal size, I have not worked within a quantitative
frame. Thus, the fact that a feature occurs, for instance, with a higher fre-
quency in my colonial corpus than in my Civil War corpus does not necessar-
ily mean that this very feature had become less prominent in nineteenth cen-
tury New England folk speech.
3. See Grund (2007) for a critical view on the Salem Witchcraft Papers as a
source for studying spoken seventeenth century American English.
124 Adrian Pablé
4. Whenever I cite direct discourse as evidence, I will identify the ‘speaker’ and
try to supply information about the social status of the latter (e.g., ‘magis-
trate’). As far as the words rendered as indirect discourse are concerned, I will
limit myself to mentioning the fact that it is a scribe reporting somebody else’s
words, and, if possible, I will identify the scribe.
5. Obviously, I have also come across letters featuring unorthodox spellings but
no grammatical ‘mistakes’.
6. I have seen fit to highlight the key words or expressions in the examples.
7. Wright (2001: 251–53) equally found such uninflected present indicative
forms (other than periphrastic do) expressing habitual/iterative aspect in Early
Modern English records.
8. As far as verb stems used as preterites are concerned, the Civil War letters are
hardly representative of the contemporary folk speech of New England. In
fact, based on E. B. Atwood’s (1953: 5–25) analysis of preterites in the Lin-
guistic Atlas of New England, the unmarked past forms come, see, give, sit and
begin (among other ones) were still relatively common in the speech of the
‘noncultured’ informants born in the latter half of the nineteenth century.
9. Concerning the latter category, Wolfram (1980: 111) distinguishes between
cases featuring embedded sentences lacking the wh relative pronoun + be
(e.g., ‘I saw the boys a-playin’ basketball but I didn’t see those that was a-
playin’ football’) and post-nominal -ing forms that are verb complements, i.e.,
having an adverbial function (e.g., ‘I saw the boys a-playin’ basketball but I
didn’t see them steal anything’).
References
Alexander, Henry
1928 The language of the Salem witchcraft trials. American Speech 3:
390–400.
Atwood, E. Bagby
1953 A Survey of Verb Forms in the Eastern United States. University of
Ann Arbor: Michigan Press.
Bailey, Guy and Marvin Bassett
1986 Invariant be in the Lower South. In Language Variety in the South:
Perspectives in Black and White, Michael Montgomery and Guy Bai-
ley (eds.), 158–179. Tuscaloosa: University of Alabama Press.
Bailey, Guy and Garry Ross
1988 The shape of the superstrate: Morphosyntactic features of Ship Eng-
lish. English World-Wide 9 (2): 193–212.
Boyer, Paul and Stephen Nissenbaum. (eds.)

1977 The Salem Witchcraft Papers. Verbatim Transcriptions of the Court
Records. 3 Vols. New York: Da Capo Press. Available at:
http://etext.virginia.edu/salem/witchcraft/texts/transcripts.html
Cooley, Marianne
1992 Emerging standard and subdialectal variation in Early American
English. Diachronica 9 (2): 167–187.
Grund, Peter
2007 From tongue to text: The transmission of the Salem witchcraft ex-
amination records. American Speech 82 (2): 119–150.
Grund, Peter, Merja Kytö and Matti Rissanen
2004 Editing the Salem witchcraft records: An exploration of a linguistic
treasury. American Speech 79 (2): 146–166.
Ihalainen, Ossi
1991 On grammatical diffusion in Somerset folk speech. In Dialects of
English: Studies in Grammatical Variation, Peter Trudgill and Jack
Chambers (eds.), 104–119. London/New York: Longman.
Kirkham, Samuel
1834 English Grammar in Familiar Lectures. Baltimore: Plaskett. Repr.
Delmar: Scholars’ Facsimiles and Reprints, 1989.
Krapp, George P.
1925 The English Language in America. 2 vols. New York: Frederick
Ungar Publishing Company.
Kurath, Hans et al. (eds.)
1939–43 The Linguistic Atlas of New England. 5 vols. Providence: Brown
University.
Kurath, Hans
1949 A Word Geography of the Eastern United States. Ann Arbor: Uni-
versity of Michigan Press.
Kytö, Merja
2004 The emergence of American English: Evidence from seventeenth-
century records in New England. In Legacies of Colonial English,
Raymond Hickey (ed.), 121–157. Cambridge: Cambridge University
Press.
Mencken, Henry Louis
1948 The American Language: Supplement II. New York: Knopf.
Montgomery, Michael
1997 A tale of two Georges: The language of Irish Indian traders in Colo-
nial North America. In Focus on Ireland, Jeffrey Kallen (ed.), 227–
254. Amsterdam and Philadelphia: John Benjamins.
2005 Voices of my ancestors: A personal search for the language of the
Scotch-Irish. American Speech 80 (4): 341–365.
126 Adrian Pablé
Montgomery, Michael and Margaret Mishoe

1999 He bes took up with a Yankee girl and moved up there to New York:
The verb bes in the Carolinas and its history. American Speech 74
(3): 240–281.
Pablé, Adrian and Radoslaw Dylewski
2007 Invariant be in New England folk speech: Colonial and postcolonial
evidence. American Speech 82 (2): 151–184.
Rissanen, Matti
1997 Salem witchcraft trials as evidence of Early American English. In
Language in Time and Space, Herbert Ramisch and Kenneth Wynne
(eds.), 183–192. Stuttgart: Franz Steiner Verlag.
Rissanen, Matti
2003 Salem Witchcraft Papers as evidence of Early American English.
English Linguistics 20 (1): 84–114.
Schneider, Edgar and Michael Montgomery
2001 On the trail of early nonstandard grammar: An electronic corpus of
Southern U. S. Antebellum overseers’ letters. American Speech 76:
388–410.
Silver, Nina and Mary Beth Sievens (eds.)
1996 Yankee Correspondence: Civil War Letters between New England
Soldiers and the Home Front. Charlottesville: The University Press
of Virginia.
Trudgill, Peter
2004 New-Dialect Formation. The Inevitability of Colonial Englishes.
Edinburgh: Edinburgh University Press.
Wakelin, Martyn
1986 English on the Mayflower. English Today 8: 30–33.
Webster, Noah
1789 Dissertations on the English Language. Repr. Menston: The Scolar
Press Limited, 1967.
Wolfram, Walt.
1980 A-prefixing in Appalachian English. In Locating Language in Time
and Space, William Labov (ed.), 107–142. New York: Academic
Press.
Wright, Laura
2001 Third-person singular present tense -s, -th, and zero, 1575–1648.
American Speech 76 (3): 236–259.
Chapter 7
The rise of prescriptive grammars on English in the
18th century
Miriam A. Locher
1. Introduction
This chapter deals with the dramatic increase in the number of publications
on English grammar in the second half of the 18th century in England.
These texts have been discussed in connection with the process of lan-
guage standardization since they propagate a normative, prescriptive view
of language (cf., e.g., Milroy and Milroy 1991). The questions I will focus
on are the following:
– Why is there an increase in the number of grammars of the English lan-

guage from the middle of the 18th century onwards? and
– How are these texts to be understood in their socio-historical context?
One can of course ask ‘why should we care about prescriptive grammars?’,
‘why should we study them’ or ‘why should we ask the questions previ-
ously raised’? The answers to these questions are, of course, positive ones
in that it makes a great deal of sense to study this era and these texts for the
following reasons: (1) If we are dealing with sociolinguistics, we first of all
have to study language in its context, and investigate the factors that might
have influenced the creation of such texts. The social history of the time
has to be taken into account: We profit from knowledge about the school
situation, about the literary texts of the time, about earlier grammars, about
the printing situation, etc. in order to develop an understanding of the
meaning of the texts. I intend to point out a number of such socio-historical
factors in this chapter. (2) The grammars of the 18th century are important
with respect to many of the ideological ideas on language, that are still
present in today’s discourse, and the process of language standardization,
which, of course, did not end in the 18th century. Finally, (3) to understand
today’s ongoing language processes better, it is important to embed our
investigations in historical studies. Some have called this field ‘historical
128 Miriam A. Locher
sociolinguistics’ others have named it ‘socio-historical linguistics’, as re-

flected in the title of the online journal edited by Tieken-Boon van Ostade
in the Netherlands, that actually combines the two terms. I believe that we
can gain much from such investigations, no matter what term we use.
This chapter is organized as follows: In section 2, I will give a brief his-
torical sketch to contextualize the situation before the 18th century. In Sec-
tion 3, I will describe the increase in the number of grammars of the Eng-
lish language in the 18th century in more detail and in Section 4 I will point
to some characteristics of such grammars in order to demonstrate how
these texts were constructed. Then I will proceed to answer the question
‘Why in the 18th century?’ and will illustrate my findings with examples
from the grammars (Section 5). To conclude I will offer an analysis of a
grammar from 1784 by a lesser known grammarian called John Fell, be-
cause he displays a surprisingly modern attitude to the study of the English
language that might be unexpected for a prescriptive grammarian (Section
6).
2. Before the 18th century: A brief historical sketch
The situation of grammar writing from the renaissance until just prior to
the 18th century can only be presented very briefly here. In general, we can
say that, while there was a wealth of textual material in English, there was
not yet what we would understand as a standardized system of orthography
or an established version of English that was perceived as ‘standard Eng-
lish’. In addition, many scientific and scholarly texts were written in Latin
rather than English, there was no official school system yet, nor was there a
systematic teaching of the English language as such. Overall, we can say
that, in comparison with the 18th century, there were only a number of trea-
tises on the English language from 1580 onwards, some of which were still
in Latin.
Examples of such 16th and 17th century grammars of the English lan-
guage are the following:
William Bullokar 1586 Pamphlet for Grammar
Paul Greaves 1594 Grammatica Anglicana
Charles Butler 1633 The English Grammar
Ben Jonson 1640 English Grammar
John Wallis 1653 Grammatica Linguae Anglicanae
John Wharton 1654 An English Grammar
The rise of prescriptive grammars on English in the 18th century 129
For example, John Wallis’ Grammatica Linguae Anglicanae, which was

written in Latin, was highly influential for the writers of the next century
as well.
3. The increase in the number of grammars in the 18th century
As mentioned above, there is an enormous increase in grammar writing

from the 1750s onwards. Ian Michael (1997: 41) reports for the last three
decades of the 18th century that 99 new grammars were published during
that time. If you count the new editions and print-runs of previously pub-
lished grammars as well, the total is an astonishing 345. To visualize this
increase in the number of grammars in the second half of the 18th century, I
have reproduced a graph by Michael (1987: 12) that shows the number of
publications in intervals of five years. Figure (1) demonstrates that the
Figure 1. The increase in the number of grammars on the English language per five
year slots, according to Michael (1987: 12) (shading added; reprinted
with the permission of Cambridge University Press)
number of publications increases steadily from the middle of the 18th cen-
tury onwards. What one can also see is that this process did not stop by any
means at the end of the 18th century, a point I will return to later.
4. Material: The 18th century grammars consulted
In order to study the grammars, we have to go to the original primary

sources, which I accessed in the form of the “English Linguistics 1500–
1800” microfiches, edited by Alston in 1974. For this analysis the gram-
mars I have chosen from this corpus are all from the second half of the 18th
century. The authors are:
Fisher, Ann 1750 A new grammar.
Priestley, Joseph 1761 The rudiments of English grammar.
Lowth, Robert 1762 A short introduction to English grammar.
Fenning, Daniel 1771 A new grammar of the English language.
Fell, John 1784 An essay towards an English grammar.
Murray, Lindley 1795 English grammar, adapted to the different
classes of learners.
The grammars by Fisher, Priestley, Lowth, Fenning and Murray were suc-
cessful grammars of their times, while Fell’s was less well known (Alston
1974).
4.1. Some characteristics of these grammars
Before we move to some explanations for the increase in numbers, I will

mention a few characteristics of the grammars I consulted to give the
reader an idea of how these texts were constructed.
– In the title pages or prefaces it is usually mentioned that the grammars

are written for young men, ladies, children, and/or schools or foreign-
ers. The authors often mention that English is not taught at schools and
that their book is intended to fill this gap.
– The grammars have a clearly ‘prescriptive’, some also a ‘proscriptive’
stance. This means that they not only list what should be done, but some
also describe what linguistic behavior should be avoided by giving ex-
amples of bad English.
– By analogy, a distinction is being made between ‘proper’, ‘correct’, and

‘polite’ versus ‘improper’, ‘low’ and ‘incorrect’ use of English.
– Some authors attempt to move away from the Latin example of gram-
mar writing by inventing English terminology. Ann Fisher (1750), for
example, proposes the term ‘helping verb’ instead of ‘auxiliary verb’.
– In terms of text genre, there are two main types: Namely a “Question &
Answer” text, or, alternatively, the grammars are written as an informa-
tive “essay”, that is as a factual non-dialogic text.
– The grammarians comment on the way in which their books should be
used, and usually recommend learning the grammar rules by heart.
Some also offer exercises for the students.
– In addition, it should be mentioned that the grammarians rely heavily on
each others’ work. There is, in other words, a significant amount of
copying going on, although some authors, like Lindley Murray, avoid
reproaches of plagiarism by referring to themselves as ‘compilers’
rather than as authors of the grammars (cf. Tieken-Boon van Ostade
1996; Steadman-Jones 2003; Terry 2003).
These characteristics are also the ones reported in the literature on pre-
scriptive grammars in general. I will present some examples for the pre-
scriptive nature of the texts, as this is one of their most striking features.
The examples are taken from Robert Lowth’s influential A short introduc-
tion to English grammar from 1762. Robert Lowth was a renowned Oxford
Professor of Poetry, well-known for his expertise in Hebrew. In 1777 he
became Bishop of London.
What we can see in Figure (2) are two fairly typical pages of Lowth’s
Grammar. Lowth gives the description of the grammar rules in larger print
and then comments on the main text in footnotes. It is mainly in these foot-
notes that we find evaluative comments, and good and bad examples of
English. The arrows are added and point to the sources that Lowth uses:
Clarendon, Dryden, Pope, Tillotson.
To illustrate this usage, let us have a closer look at footnote 7, reprinted
in Example (1):
(1) “He caused all persons, whom he knew had, or he thought might have, spo-
ken to him, to be apprehended.” Clarendon, Vol. III, p. 618.8 It ought to be
who, the Nominative Case to had; not whom as it were the Objective Case
governed by knew. (Lowth 1762: 97, emphasis in bold added)
This it ought to be is a quite typical formulation used by Lowth and can be

read as an indication of his prescriptive stance.
96 A Short Introduction to English Grammar 97

Sentences are Simple, or Com- are the Nominative7 Cases; and an-
pounded. swer to the question who, or what?
A Simple Sentence hath in it as, “Who is loved? Thomas.” And
but one Subject, and one Finite the Verb agrees with the Nomina-
Verb; that is, a Verb in the Indi- tive Case in number and person8;
cative, Imperative, or Subjunctive
7
Mode. “He caused all persons, whom he knew
had, or he thought might have, spoken to him,
A Phrase is two or more words
to be apprehended.” Clarendon, Vol. III, p.
rightly put together in order to 618.8 It ought to be who, the Nominative
make a part of a Sentence; and Case to had; not whom as it were the Ob-
sometimes making a whole Sen- jective Case governed by knew.
tence. “Scotland and Thee did each in other live.”
Dryden, Poems, Vol. II p. 220.
It ought to be Thou.
The most common PHRASES
8
used in simple Sentences are as fol- “But Thou false Arcite never shall obtain
lows: Thy bad pretence. Dryden, Fables.
“That Thou might fortune to thy side engage.”
1st Phrase: The Substantive be- Prior.
fore a Verb Active, Passive, or Neu- It ought to be shalt, mightest. The mistake
ter; when it is said what thing seems to be owing to the compounding of Thou
is, does, or is done; as, “I am;” and You as equivalent in every respect; where-
as one is Singular, the other Plural. See
“Thou writest;” “Thomas is above, p. 48. “Great pains has [have] been
loved:” where I, Thou, Thomas taken.” Pope, P.S. to the Odyssey. “I have
considered, what have [hath] been said on
both sides in this controversy.” Tillotson, Vol.
I. Serm. 27.
Figure 2. Lowth, A short introduction to English grammar, 1762, pages 96-97

(arrows added)
Lowth uses a canon of mainly restoration and early 18th century writers as
his source for good and bad examples. As Table 1, compiled by Tieken-
Boon van Ostade (1997: 452), shows, Lowth does not shy away from criti-
cizing even the language of the bible. Swift, to name just one of his other
sources, is mentioned 39 times, and is criticized 39 times. Lowth thus sets
himself up as an authority over authors of literary work.
Table 1. Numbers of quotations from Lowth’s “best writers” as well as from the
Bible (reproduced from Tieken-Boon van Ostade 1997: 452; reprinted
with the permission of the Société Neophilologique)
total main text notes criticised
Bible 44 10 34 21
Old testament 19 6 13 6
New testament 25 [sic.] 4 22 15
Swift 39 - 39 39
Addison 22 4 18 16
Dryden 20 4 16 16
Pope 18 2 16 14
Clarendon 17 - 17 17
Milton 16 1 15 15
Prior 14 - 14 14
Tillotson 12 - 12 12
Atterbury 6 - 6 6
Bolingbroke 6 - 6 6
Gay 1 - 1 1
5. Why in the 18th century?: Suggested reasons for the increase in the
number of prescriptive grammars
Let me now move to answering the question ‘why did this increase in the
number of grammars occur in the second half of the 18th century’? I pro-
pose the following four reasons:
(I) There is a new market situation with so-called ‘social climbers’,
(II) a process of language standardization is involved,
(III) as well as the notion of politeness,
(IV) and finally, new developments in the printing trade play an impor-
tant role.
It has to be stressed that these reasons are interrelated and thus closely
connected. In what follows I will discuss them one by one.
5.1. New market situation: ‘Social climbers’
In the 18th century we have an increase in population numbers from

roughly 7 to 12 million people (Belanger 1982: 18–19; ‘United Kingdom’,
Encyclopaedia Britannica 2007). The middle classes grew and aspired to
climb the social ladder. They are referred to as social climbers.
Apart from amassing more wealth, these people wanted to better their
situation by investing into ‘polite’ language – a notion that I will return to
later. With respect to the grammars, Susan Fitzmaurice (1998: 315) claims
that “[w]hat is striking about these texts is that they seemed to promise
social advancement”. Correct language was therefore seen as a commodity
that could be bought and “marketed” (Fitzmaurice 1998: 325). Tieken-
Boon van Ostade (2000: 33) supports this argument by saying that the
booksellers and publishers had an interest in exactly this market to make a
financial profit: “The booksellers’ concern was not the language but the
market”. For example, one of the leading publishers, Robert Dodsley, not
only published Samuel Johnson’s famous Dictionary in 1755, but also
Lowth’s grammar in 1762 (Tieken-Boon van Ostade 2000, 2003).
5.2. The process of ‘standardization’
The 18th century is not only the century of population increase, the middle
classes and social climbers, but it is also closely associated with the proc-
ess of language ‘standardization’ in general. Even as early as 1697, Daniel
Defoe called for a language ‘Academy’ in analogy to the Académie Fran-
çaise. This English Academy, however, never became a reality. And Swift,
in his 1712 Proposal for Correcting, Improving, and Ascertaining the Eng-
lish Tongue also argued for improving the current language situation.
In the course of the 18th century the idea that English was deteriorating
and that it was past its prime is also reflected in the grammars which try to
‘fix’ proper language usage. This process is called ‘codification’ in the
literature on standardization (cf., e.g., Haugen 1966). Several ideologies
have been found to play a prime role in this process, as discussed, for ex-
ample by Milroy and Milroy (1991). Watts (2000) mentions, among others,
the
– Language and nationality myth: The idea that Britain needs one lan-
guage that unifies it and that can be transported to the colonies;
– The myth of superiority: The idea that a particular variety of English is

better than all the others;
– The myth of the perfect language: The idea that a language can reach a
state of perfection. In the grammars this is usually seen to have been the
case at the turn of the 17th to the 18th century; and
– The myth of the undesirability of change: This means that if change is
happening, it is perceived as being for the worse.
The process of standardization is a fascinating topic that would have de-

served a chapter in its own right. Here I cannot go much further into detail,
but I will briefly illustrate the ‘myth of superiority’ with an example from
Hugh Jones (1724). In his An accidence to the English tongue, Jones is
quite clear about which English needs to be fixed. He claims that:
(2) Out of this Confusion of English may we collect 5 principal Dialects and
Tones.
1. The Northern Dialect, which we may call Yorkshire.
2. The Southern, or Sussex Speech.
3. The Eastern, or Suffolk Speech.
4. The Western, which we may call Bristol language.
5. The Proper, or London Language. (Jones 1724: 13)
What we can see here is a process of ‘selection’, that is elevating one vari-
ety at the cost of the others. This process also eventually helped to bring
about a stigmatization of those speakers who did not speak the ‘proper’
language (cf., e.g., Milroy 1999, 2000). As I have reported previously,
Lowth, like many of the other grammarians in the second half of the 18th
century, uses a canon of written, literary sources to create the distinction
between good and bad English.
5.3. The notion of ‘politeness’
One aspect that is closely linked to the process of standardization is the

notion of politeness. Several researchers (Stein 1994; Fitzmaurice 1998;
Watts 1990, 2000, 2002) have studied this aspect and the following explo-
rations are based on their work. We can summarize three processes that are
of importance with respect to politeness in the 18th century:
(I) Two poles emerge between ‘correct’ / ‘proper’ versus ‘incorrect’ /

‘low’ / ’vulgar’ language usage.
(II) There is the creation of a moral distinction between people who speak
politely and ‘with propriety’ and those who do not.
(III) It is believed that it is possible to learn the ways of ‘polite society’.
I will now look at these three propositions in turn. A vital aspect of the
study of politeness is the historical development of this concept, a point
also stressed in my current research on politeness in contemporary English
(cf., e.g., Locher 2006). In the present context this means that we have to
look at the change in meaning the word ‘politeness’ experienced in the 18th
century. Stein (1994: 8) claims that before the 18th century, politeness re-
ferred to “a social ideal, the polite urban, metropolitan gentleman, well-
versed in the art of ‘polite’ conversation, a man about town”. By the sec-
ond half the 18th century, Stein continues, a new notion of ‘politeness’ had
developed, one that is closely linked to prescriptivism, in that two poles
between ‘correct’ / ‘proper’ language spoken by the so-called ‘polite soci-
ety’ and ‘incorrect’ / ‘low’ / ‘vulgar’ language usage are created. To illus-
trate this consider the following three examples from Lowth’s grammar. In
Example (3), Lowth describes the usage of an eight days as obsolete, vul-
gar and improper:
(3) About “an eight days:” that is, a space of eight days. Luke ix. 28. But the
expression is obsolete, or at least vulgar; and we may add likewise im-
proper: … (Lowth 1762: 20)
Lowth objects to the usage of “an eight days”, i.e., the combination of the
singular article with a plural noun, on the grounds that days “has not been
reduced by use and convenience into one collective and compact idea”
such as a hundred or a thousand. In Example (4), Lowth describes an
agreement mismatch as an enormous solecism, a serious blunder:
(4) You was, the Second Person Plural of the Pronoun places in agreement with
the First or Third person Singular of the Verb, is an enormous Solecism: and
yet Authors of the first rank have inadvertently fallen into it. (Lowth 1762:
48)
And in (5) Lowth censures the omission of relative pronouns most severly:
(5) The Construction is hazardous, and hardly justifiable, even in Poetry.

(Lowth 1762: 137)
What we can see from these examples is a creation of a notion of ‘correct’,

‘proper’ and ‘polite’ language versus ‘incorrect’, ‘improper’ and ‘impolite’
language usage.
The second aspect linked to this development is that a moral distinction
started to be made between people who speak politely and ‘with propriety’
and those who do not. This is illustrated once more with an extract from
Hugh Jones’ text. He claims that:
(6) We should aim at an elegant and fluent Style; gliding like a smooth River,
and not running violently like a rapid Torrent. Our Language affords us
Choice of Words, and Variety of Expression; in which we should imitate the
Learned and Polite, the Correct and Pure, without jingling Terms, harsh or
obsolete, vulgar or unbecoming Words, ungrateful to the Ear, difficult in
Sound, or offensive to Modesty, good Manners, or good Sense. (Jones 1724:
62, emphasis added)
Jones thus claims that the Learned and Polite people are also Correct and
Pure and that they should be taken as a model. A similar view is expressed
by Lindley Murray, an American who lived and published in England most
of his life. He argued in the following way in 1795:
(7) That persons who think confusedly, should express themselves obscurely, is
not to be wondered at; for embarrassed, obscure, and feeble sentences, are
generally, if not always, the result of embarrassed, obscure, and feeble
thought. (Murray 1795: 183)
Murray proposes that a connection is made between a person’s mental

capacity and the way he or she speaks. Finally, in (8) Lowth describes the
so-called mismatching of verb forms as follows:
(8) In the rest it seems wholly inexcusable. The absurdity of it will be plainly
perceived in the example of some of these Verbs, which Custom has not yet
so perverted. We should be immediately shocked at I have knew, I have
saw, I have gave, &c: but our ears are grown familiar with I have wrote, I
have drank, I have bore, &c. which are altogether as barbarous.” (Lowth
1762: 89–90, emphasis in bold added)
That Lowth labels these mismatchings as inexcusable, perverted, and bar-

barous and argues that one should be shocked at hearing them is significant
because these words reflect on the speaker’s character. The way a person
talked began to be seen as an index of his or her social worth as well. What
we witness, then, in the latter half of the 18th century is, to put it in Richard
Watts’ words, that
language use became a marker of socio-political affiliation in Britain, and
has continued to serve the function of social, political and educational de-
marcation till the present day. (Watts 1990: 299)
This brings me to the final point connected to politeness. The social climb-
ers mentioned earlier had an interest in acquiring ‘proper’ language. This
‘proper’ language is what they were given to believe they would find in the
grammars. They wanted to avoid appearing as if they belonged to the
‘lower’ classes, and they hoped that they could better their social situation
by acquiring ‘proper’ English. The texts clearly indicated that you could
learn to speak properly, as exemplified in Example (9), taken from
Lowth’s preface:
(9) A Grammatical Study of our own Language makes no part of the ordinary
method of instruction which we pass thro’ in our childhood; and it is very
seldom that we apply ourselves to it afterward. And yet the want of it will
not be effectually supplied by any other advantages whatsoever. Much
practice in the polite world, and a general acquaintance with the best au-
thors are good helps, but alone will hardly be sufficient: we have writers,
who have enjoyed these advantages in their full extent, and yet cannot be
recommended as models of an accurate style. (Lowth 1762: vii)
In his preface, Lowth first points out that a lack in proper language educa-
tion cannot be supplied by any other advantages whatsoever. In other
words, no matter how wealthy you have become, your language will still
give you away – unless you study. In the continuation of the example,
Lowth admits that even writers who had practice in the polite world still
make mistakes, presumably because they had no proper training in the Eng-
lish language. At the same time, however, Lowth implies that the whole
endeavor of the grammar he has written is meant to educate those who do
not have access to this polite world, and that studying the English language
is a necessity in order to become refined and in command of an accurate
style. The middle classes were thus encouraged in their struggle for social
betterment and were led to believe that language played a major role in
such an advancement.
5.4. New developments in the ‘printing’ trade
Finally, the last point suggested as a factor that facilitated the increase in
the number of grammars in the 18th century is the development in the print-
ing trade more generally. According to Belanger (1982), at the end of the
17th century there were virtually no magazines or newspapers and publish-
ing was only possible in London, Oxford and Cambridge due to the so-
called ‘Licensing Act’, which expired in 1695. If we compare this situation
to the one at the end of the 18th century, we find that a dramatic shift has
taken place. There are now local newspapers and periodicals, even though
the national newspapers are still based and thriving in London, and there
are pamphlets and books available for purchase. In addition, the copyright
situation had changed dramatically over the same period of time, in that
there was no longer any perpetual copyright. This meant that provincial
publishing received an enormous impetus, that cheap editions of the clas-
sics could now be produced, and that unaltered reprinting of established
bestsellers, such as the grammars, had become possible everywhere.
This changed situation in the printing trade, in combination with the
other factors previously mentioned, helps to explain the large numbers of
re-editions and reprinting of prescriptive grammars. Lowth’s grammar, for
example, sold 34,000 copies by 1781 (Mandelbrote 2004). Since the mar-
ket was clearly there, this might also go a long way towards explaining that
the increase in the number of new grammars continued into the 19th cen-
tury.
5.5. Summary: Suggested reasons for the increase in the number of

prescriptive grammars
To summarize so far, we can say that there are several interrelated factors
which can explain the increase in the number of grammars of the English
language from the middle of the 18th century. The grammars have to be
seen in the light of the process of ‘standardization’, which elevated one
particular English variety over others. This process can be seen in connec-
tion with the notion of ‘politeness’ that had developed away from describ-
ing polished, and refined behavior in general, to pointing out ‘correct’ lan-
guage usage. The middle classes believed that, by acquiring ‘polite’
language, they would also get access to ‘polite’ society. The grammar
books, in addition to spelling books and dictionaries, were perceived by the
social climbers as a help on their way to achieving this aim. There was, in
other words, a market for these books, and one that was considerable due
to the increase in population in the 18th century. Finally, and not unimpor-
tantly, the new developments in the printing trade made it possible to cater
for this market and to produce books all over the country.
6. John Fell’s Essay towards an English grammar (1784)
To demonstrate the topicality of 18th century grammars for today’s lin-

guists, I have chosen examples from the grammar by John Fell, a minister,
schoolteacher and classical scholar (Gordon 2004). This grammar was
published in 1784 and contains a preface written by the editor, that is “gen-
erally attributed to Richard Sharp, a pupil of Fell’s” (Alston 1974: I, 380).
It is not a particularly successful grammar in that it was only published
once, but I have chosen it because there are interesting thoughts on the
English language as such that are worth being discussed. Let me start by
giving a number of details on this work.
6.1. Characteristics
According to the editor in the preface, John Fell’s grammar is written and
published with the aim of “instructing youth”, for those who “are already
acquainted with polite literature” and for “foreigners” (xiii). It consists of a
preface, a discussion of the parts of speech, a chapter on syntax and a text
on auxiliary verbs. Fell’s grammar is thus less complete than others’ since
he leaves out a discussion of the spelling system or prosody (cf. e.g., Ann
Fisher’s grammar). The text is written in an ‘essay’ rather than a question
and answer style. Fell does not give ‘bad examples’ of English but uses
‘good writing’ for illustration. Among the sources of good writing that he
uses, Fell lists the Holy Scriptures, Shakespeare, Milton, Sidney, Locke,
Addison, Dryden and Pope (xv). In addition, Fell explicitly refers to the
grammars by Lowth, Johnson and Priestley. He often discusses different
points of view on a grammar issue, and asks rhetorical questions before
presenting his own conclusions. He thus engages in a discourse with the
other grammarians (cf. Watts [1999] for the notion of a ‘discourse commu-
nity’ with respect to the grammarians). What I want to focus on here are
the contradictory statements about language in the editor’s Preface (v–xiv)
but also in Fell’s main text, because they show us competing forces at
work.
6.2. Striking contradictions
On the one hand, the Preface clearly describes the quite common idea that
there is a pure and genuine English, as can be seen in Example (10).
(10) During the last thirty or forty years …. there are but a few [writers] that
deserve the praise of having expressed themselves in a pure and genuine
strain of English. (Fell 1784: vi, emphasis added)
In (11), the editor identifies the age of best writing at the beginning of his
century:
(11) I fear we cannot find in our later writers so rich, so appropriated a diver-
sity of expression as we can in those that flourished at the beginning of
this century. (Fell 1784: x, emphasis added)
And in (12), he shows prescriptive and evaluative tendencies, when he

speaks of disgusting language that is due to affectation. This term was, at
the time, perceived to be the opposite of ‘polite’ language (Watts 2002:
159).
(12) The alterations in our language here taken notice of, are certainly not for
the better: they give the phraseology a disgusting air of study and formal-
ity: they have their source in affectation, not in taste; … (Fell 1784: ix,
emphasis added)
Examples (10) to (12) are very much in line with the prescriptive stance I
have illustrated in previous sections.
The contradictions to this position arise in quotations such as the fol-
lowing. In (13), a fairly modern idea of a non-prescriptive, even descriptive
approach to a grammarian’s work is pursued by the editor:
(13) It matters not what causes these customs and fashions owe their birth to;
the moment they become general, they are laws of the language; and a
grammarian can only remonstrate, how much soever he disapprove. From
his opinions and precepts an appeal may always be made to the tribunal of
use, as to the supreme authority and last resort: for all language is merely
arbitrary. (Fell 1784: xii–xiii, emphasis added)
This modern attitude is especially aptly summarized in Fell’s comments on

neuter and passive verbs in (14):
(14) … the Writer will by no means presume either to resist or censure, intend-
ing no more than to state the facts, frequent among our best writers, just
as they are, … Whether such conduct would be a real improvement, or a
diminution [sic.] of our language, the learned public must determine, with
whom is the undoubted right of decision. (Fell 1784: 114, emphasis
added)
Here Fell seems to be influenced by the science tradition thriving in the

18th century, which has also been called the Age of Reason and Enlighten-
ment. He presents himself as somebody who collects facts and presents
them just as they are. This attitude can also be found in Joseph Priestley’s
work, The rudiments of English grammar from 1761, and precedes the
ideas of the 19th century grammarians who put forward the idea that lin-
guistics should be descriptive rather than prescriptive (e.g., Max Müller
1861; Bloomfield 1933; discussed in Milroy and Milroy 1999: 6–7). At the
same time, however, Fell leaves judgement to the learned public, which
immediately contradicts a descriptive approach as such.
What is the solution to these apparent contradictions between the quota-
tions (13) and (14) and the ones in examples (10) to (12)? I suggest that the
writer of the preface and Fell himself do not actually solve the contradic-
tions, but that the solution the editor proposes in the preface merely adds a
further angle to the discussion as can be seen in a further extract. In one
and the same Example ([15]–[17]) we can see that he wants the grammar to
be both prescriptive as well as descriptive. Just before Example (15), the
editor has mentioned the efforts of the Académie Française to save French
from deterioration. He goes on to say:
(15) [B]ut the republic of letters is a true republic, in its disregard to the arbi-
trary decrees of usurped authority. Perhaps such an institution would do
still less with us. Our critics are allowed to petition, but not to command:
and why should their power be enlarged? (Fell 1784: xi)
The editor is thus not in favor of an English Academy, and implies once
more that authority lies with custom, that is with the language users. The
role of critics can therefore only be to suggest changes, but not to com-
mand them. He then continues by saying that:
(16) The laws of our speech, like the laws of our country, should breathe a
spirit of liberty: they should check licentiousness, without restraining free-
dom. (Fell 1784: xi)
In other words, the editor – and I assume Fell too (cf. example [14]) – is in
favor of regulating language usage after all, but only to a certain degree
and without constraining a person’s freedom. In addition, the editor feels
strongly that the English language in its current stage is deteriorating. He
says that:
(17) The most effectual method of preserving our language from decay, and
preventing a total disregard to the Saxon part of it, is to bring about a
revolution in our present mode of education. (Fell 1784: xi)
The solution to prevent this deterioration from continuing, in the editor’s

view, is to revolutionize the education system.
6.3. Striking contradictions: Summary
Summarizing this discourse, we can see two contradictory ideals. On the

one hand, a ‘scientific’ approach is described that is coupled with an idea
that language changes, that ‘use’ is the ultimate authority, and that the
grammarian is only a compiler of what he can observe in the (primary)
texts, an attitude reflecting the era’s scientific concerns with empiricism.
On the other hand, the text demonstrates a different ideological approach
which propagates a belief in the existence of a ‘best’ language, coupled
with a notion that the Saxon heritage must be preserved. The time of best
writing is linked to the writers of the early 18th century. This conservative
view stands in opposition to the implied recognition that language changes
through use. Such contradictory views are not reconciled in the text, but an
emphasis on the importance of education is added.
In doing this, the text also reflects a concern of many of the numerous
grammar publications during the 18th century, namely the feeling that the
English language has been neglected in the teaching of young people and
that it should deserve more attention in the schools. While ‘English’ had
already been taught in schools before the 18th century, never before was
there a need for instruction expressed this clearly and frequently. This can
also be seen in the numerous private schools that opened in the 18th century
to provide an education in the English language, some of which only lasted
a couple of years. Attending school in general, however, only became com-
pulsory towards the end of the 19th century (cf. Michael 1987; ‘United
Kingdom’, ‘Education’ in Encyclopaedia Britannica).
The 18th century is a fascinating era for linguists to study an ever-

increasing awareness of the English language as witnessed in the printed
material of the time. The grammars that are produced in such enormous
numbers from the middle of the century onwards deserve being studied in
more detail. Among others, Tieken-Boon van Ostade and her team are cur-
rently pursuing this interest in the research project on ‘The codifiers and
the English language: tracing the norms of Standard English’ (2004). In
this chapter I discussed a few reasons for the increase in the numbers of
publications on English. In addition, I suggest that Fell’s grammar serves
as an intriguing example to show that many of the issues that linguists deal
with today, such as prescriptivism versus descriptivism exemplified in the
controversy around John Honey’s contribution (e.g., 1997), the ongoing
question of the development and nature of standard English, and the con-
nection of these issues to education were already discussed in the 18th cen-
tury. The brief analysis of Fell’s Grammar shows that a text of a compara-
tively unknown grammarian and his editor can serve unexpectedly as a rich
source for modern-day linguists who are interested in these subjects.
Acknowledgments
My heartfelt ‘thank you’ go to the following friends and colleagues who gave feed-
back to an early version of this paper that I presented as a lecture: Anne-Françoise
Baer-Bösch, Jachin Baumgartner, Mathias Kimmich, Danièle Klapproth, Nicole
Nyffenegger, Lukas Rosenberger, Philippe Schweighauser, Jürg Strässler, Julia
Straub, Franz Andres Morrissey and Richard Watts.
References
1. Primary sources: Grammars

(ordered by year of publication; all published by Scolar Press, Menston, 1974)
William Bullokar 1586 Pamphlet for Grammar
Paul Greaves 1594 Grammatica Anglicana
Charles Butler 1633 The English Grammar
Ben Jonson 1640 English Grammar
John Wallis 1653 Grammatica Linguae Anglicanae
John Wharton 1654 An English Grammar
Jones, Hugh 1724 An Accidence to the English Tongue

Fisher, Ann 1750 A New Grammar
Priestley, Joseph 1761 The Rudiments of English Grammar
Lowth, Robert 1762 A Short Introduction to English Grammar
Fenning, Daniel 1771 A New Grammar of the English Language
Fell, John 1784 An Essay towards an English Grammar
Murray, Lindley 1795 English Grammar, Adapted to the Different
Classes of Learners
2. Sources mentioned
Alston, Robin C. (ed.)
1974 English Linguistics 1500–1800 (A Collection of Facsimile Reprints).
Menston: The Scolar Press.
Belanger, Terry
1982 Publishers and writers in eighteenth-century England. In Books and
their Readers in Eighteenth-Century England, Isabel Rivers (ed.), 5–
25. Leicester: Leicester University Press.
Bloomfield, Leonard
1933 Language. New York: Holt, Rinehart and Winston.
Defoe, Daniel
1697 An Essay upon Projects. Menston: Scolar Press, 1969.
Education, history of
2007 In Encyclopaedia Britannica. Retrieved March 25, 2007, from Ency-
clopaedia Britannica Online: http://search.eb.com/eb/article-
9105951.
Fitzmaurice, Susan
1998 The commerce of language in the pursuit of politeness in eighteenth-
century England. English Studies 78: 309–328.
Gordon, Alexander
2004 Fell, John (1735–1797). In Oxford Dictionary of National Biogra-
phy. From the Earliest Times to the Year 2000, Henry Colin Gray
Matthew and Brian Harrison (eds.), 264. Oxford: Oxford University
Press.
Haugen, Einar
1966 Dialect, language, nation. American Anthropologist 68: 922–935.
Honey, John
1997 Language is Power: The Story of Standard English and its Enemies.
London: Faber and Faber.
Locher, Miriam A.
2006 Polite behavior within relational work: The discursive approach to
politeness. Multilingua 25 (3): 249–267.
Mandelbrote, Scott
2004 Lowth, Robert (1710–1787). In Oxford Dictionary of National Biog-
raphy. From the Earliest Times to the Year 2000, Henry Colin Gray
Matthew and Brian Harrison (eds.), 613–616. Oxford: Oxford Uni-
versity Press.
Michael, Ian
1987 The Teaching of English, from the Sixteenth Century to 1870. Cam-
bridge: Cambridge University Press.
1997 The hyperactive production of English grammars in the nineteenth
century: A speculative bibliography. Publishing History 41: 23–61.
Milroy, James
1999 The consequences of standardisation in descriptive linguistics. In
Standard English: The Widening Debate, Richard J. Watts and Tony
Bex (eds.), 16–39. London: Routledge.
2000 Historical description of the ideology of the standard language. In
The Development of Standard English 1300–1800: Theories, De-
scriptions, Conflicts, Laura Wright (ed.), 11–28. Cambridge: Cam-
1991 Authority in Language (3rd edition). London: Routledge.
Müller, Max
1861 Lectures on the Science of Language, Delivered at the Royal Institu-
tion of Great Britain in 1861. London: Longman, Green, Longman,
and Roberts.
Steadman-Jones, Richard
2003 Lone travellers: The construction of originality and plagiarism in
colonial grammars of the late eighteenth and early nineteenth centu-
ries. In Plagiarism in Early Modern England, Paulina Kewes (ed.),
181–200. London: Palgrave Macmillan.
Stein, Dieter
1994 Sorting out the variants: Standardization and social factors in the
English language 1600–1800. In Towards a Standard English 1600–
1800, Dieter Stein and Ingrid Tieken-Boon van Ostade (eds.), 1–17.
Berlin and New York: Mouton de Gruyter.
Swift, Jonathan
1712 Proposal for Correcting, Improving, and Ascertaining the English
Tongue. Menston: Scolar Press, 1969.
Terry, Richard
2003 'In pleasing memory of all he stole': Plagiarism and literary detrac-
tion, 1747–1785. In Plagiarism in Early Modern England, Paulina
Kewes (ed.), 181–200. London: Palgrave Macmillan.
Tieken-Boon van Ostade, Ingrid

1996 Lindley Murray and the concept of plagiarism. In Two Hundred
Years of Lindley Murray, Ingrid Tieken-Boon van Ostade (ed.), 81–
96. Münster: Nodus Publikationen.
1997 Lowth's corpus of prescriptivism. In To Explain the Present: Studies
in the Changing English Language in Honour of Matti Rissanen,
Terttu Nevalainen and Leena Kahlas-Tarkka (eds.), 451–463. Hel-
sinki: Société Neophilologique.
2000 Robert Dodsley and the genesis of Lowth's Short Introduction to
English Grammar. Historiographica Linguistica 27: 21–36.
2003 'Tom's grammar': The genesis of Lowth's Short Introduction to Eng-
lish Grammar revisited. In The Teaching of English in the Eighteenth
and Nineteenth Centuries, Essays for Ian Michael on his 88th Birth-
day. Special issue of Paradigm 2 (7), F. Austen and C. Stray (eds.),
36–45.
2004 The Codifiers and the English Language: Tracing the Norms of
Standard English. Retrieved 13 February 2007, from
http://www.lucl.leidenuniv.nl/index.php3?c=126
United, Kingdom
2007 In Encyclopaedia Britannica. Retrieved March 25, 2007, from Ency-
clopaedia Britannica Online: http://search.eb.com/eb/article-44883.
Watts, Richard J.
1990 The role of early grammar writers in creating a linguistic tradition. In
Sprachtheorie und Theories der Sprachwissenschaft: Geschichte und
Perspektiven. Festschrift für Rudolf Engler zum 60. Geburtstag, Ri-
carda Liver, Iwar Werlen and Peter Wunderli (eds.), 299–315. Tü-
bingen: Gunter Narr.
Press.
2002 From polite language to educated language: The re-emergence of an
ideology. In Alternative Histories of English, Richard J. Watts and
Peter Trudgill (eds.), 155–172. London: Routledge.
Chapter 8
Lest the situation deteriorates – A study of lest as
trigger of the inflectional subjunctive
Anita Auer
1. Introduction
“Lest the situation deteriorates Orrrr: Lest the situation deteriorate” serves
as the title of a correspondence on the conjunction lest as a possible trigger
of the inflectional subjunctive, which I came across in a language forum on
the World Wide Web (Online discussion forum. Retrieved 29 May 2007,
from http://forum.wordreference.com/showthread.php?t=64969). In this
first posting, the anonymous writer, who reports to be a native of England,
raises the question of whether the conjunction lest requires the following
verb form to be in the indicative or the subjunctive mood. S/he notes that
the French equivalent of lest – de peur que – is followed by the subjunctive
and thus wonders whether “Lest the situation deteriorate” may be “some
bizarre leftover of the English subjunctive”. Another contributor (living in
Canada and 61 years old in 2005) confirms that the phrase is a subjunctive
and also voices the opinion that s/he thinks that it is the correct form and
that “there’s life in it yet”. Finally, a third contributor (native of Italy, 21
years old in 2006) states that s/he was taught “that lest is usually followed
by should” and then raises the question of whether “Lest the situation
should deteriorate” is correct.
I am aware that it is a rather unconventional way to begin an academic
paper with the presentation of a language forum discussion from the world
wide web. The discussion in the online forum does however highlight the
timeliness of the topic, the inflectional subjunctive and in particular that
the conjunction lest may be a trigger of the subjunctive. It is generally
agreed that the decline of the subjunctive as an inflectionally marked verb
form started in ME times (cf. Görlach 2001: 122; Traugott 1972: 149).
Statements such as “the subjunctive is dying”, “it is moribund except in a
few easily specified cases” (Fowler 1965: 595) and “[t]he subjunctive, like
whom, survives – on life support” (Redfern 2001: 65) are fairly common.
In recent years, several studies have been carried out that aimed at shed-
150 Anita Auer
ding light on the precarious subjunctive situation. These corpus-based stud-

ies investigated the development of the subjunctive form in a range of con-
structions in British English as well as other varieties of English (for stud-
ies on the subjunctive in adverbial clauses see Peters 1998; Moessner
2002a, 2002b, 2006; Auer and González-Díaz 2005; Auer 2006; and Grund
and Walker 2006; for studies on the mandative subjunctive see Haegeman
1986; Övergaard 1995; Hundt 1998, forthc.; Peters 1998; Serpollet 2001;
Fillbrandt 2006; Moessner 2007; Leech et al. forthc.). As I cannot summa-
rise the outcome of every single study here, suffice it to say for now that
the fate of the subjunctive is not as straightforward as the statements above
suggest and the form “is still enjoying an active retirement” (Peters 1998:
101).
The main focus of this paper is the role of the conjunction lest1 in the
development of the inflectional subjunctive, more specifically from Early
Modern English to the twentieth century. Even though the use of the sub-
junctive form started declining in ME times, Barber (1997: 173) notes
about the subjunctive in Early Modern English that it
is particularly frequent in clauses introduced by if, and is also quite common
in ones introduced by though, unless, except ‘unless’, whether, and till. [...]
The subjunctive is also found in clauses expressing purpose [...]
which can be introduced by the conjunction lest. As regards the level of
formality in which the inflectional subjunctive was used in Early Modern
English, Barber (1997: 173) claims that “it was part of everyday familiar
speech, even among lower-class characters.” According to Charleston
(1941: 139), “the subjunctive mood has almost completely decayed away”;
“[i]n the early 18th century, however, the subjunctive mood form is more
tenacious of life.” She then lists different kinds of clauses in which the
inflectional subjunctive was used in the beginning of the eighteenth cen-
tury. One of these environments are “negative clauses of purpose (intro-
duced by lest, for fear [that], etc.)”, which are illustrated with the exam-
ples “... they had best moderate their ambition in time, lest infamy or death
attend their success” and “... while thou hast power left thee, avoid the
tempting evil, lest thy grand enemy... return to the assault...” (Charleston
1941: 140). Similarly, Görlach (2001) notes the following:
The few remaining subjunctive forms (If he be there, God save the queen)
were distributed as they are in PDE, though slightly more frequent in the
18th century, mainly in wishes and hypothetical conditional clauses (intro-
duced by if, lest, tho’, altho’, albeit, unless, except, whether, however, that,
A study of lest as trigger of the inflectional subjunctive 151
according to Kirkby 1746:89), and largely confined to formal registers;

some phrases have become fossilized (Heaven forbid). (Görlach 2001: 122)
The conjunction lest was thus considered a trigger of the inflectional sub-
junctive in the eighteenth century. In the nineteenth century, conjunctions
such as although, unless and if (and possibly also lest) still required the
subjunctive after them if the language user tried to be formally correct.
Görlach (1999: 83) notes in a footnote that “Savage (1833: 94–5) was quite
outspoken in considering If it is so and Although he is there / does so and
Would to God that it was so as improprieties”. Even though individuals
proposed these notions of correctness, as regards the inflectional subjunc-
tive, “PDE conditions had largely been reached by 1900” (Görlach 1996:
83). In the twentieth century, Quirk et al. (1985: 158) exemplify clauses of
condition or negative purpose introduced by the conjunction lest with “The
President must reject this proposal, lest it cause strife and violence. <for-
mal>” and then state that “[t]his lest construction is restricted to very for-
mal usage in BrE, but it is more current in AmE”. In the example given lest
is followed by the inflectional subjunctive. According to Quirk et al.
(1985: 1108), “[a]rchaic lest tends to have a modal auxiliary or (esp) in
AmE the present subjunctive”. Similarly, Huddleston and Pullum (2002)
note that
lest, which belongs to formal style, is the only preposition where the sub-
junctive is the preferred construction, though specialised should, [...], is also
readily used. Ordinary declaratives are possible, but comparatively rare.
(Huddleston and Pullum 2002: 1000)
What Quirk et al. and Huddleston and Pullum agree on is that lest is con-
fined to formal contexts and that it mostly requires the subjunctive form or
modal auxiliaries. Their accounts also differ slightly: (1) Quirk et al.
(1985) state that the subjunctive is especially used in American English, a
fact that is no longer mentioned in Huddleston and Pullum’s grammar
(2002), which was published almost twenty years later. (2) Where Quirk et
al. (1985) note that lest can require modal auxiliaries, Huddleston and Pul-
lum (2002) have already narrowed modal auxiliaries down to should.
Even though the inflectional subjunctive has been on the decline since
the Middle English period, a type of construction in which the subjunctive
form has been claimed to have survived is in adverbial clauses introduced
by a range of conjunctions, one of which is the conjunction lest. In the
second half of the twentieth century, the lest-subjunctive construction ap-
peared to be more common in American than in British English (Quirk et
al. 1985: 1108). Also, according to the statements above, the construction
152 Anita Auer
used to be part of everyday familiar speech, even in lower classes, in Early

Modern English but has been associated with formal style from the Late
Modern English period onwards. As none of these claims are supported by
empirical data, the fate of lest and the subjunctive will be investigated
here.
The paper will be divided into two parts: The first part will trace the ac-
tual usage of the conjunction lest as a trigger of the inflectional subjunctive
– an investigation which will be based on diachronic and synchronic cor-
pora. The second part of the paper will be concerned with meta-linguistic
comments on the lest-subjunctive construction. The comment on the lan-
guage forum about the subjunctive being the correct form as well as Gör-
lach’s quotation (1996: 83) of Savage (1833: 94–95) suggest that the sub-
junctive is strongly associated with correct and proper language use. I will
thus investigate comments by grammarians and language-guardians from
the Early Modern English period onwards to see whether (a) lest has al-
ways been considered a trigger of the inflectional subjunctive, and (b)
these comments mirror actual language usage.
2. The development of the lest-subjunctive construction from

1570–1994
2.1. Methodological preliminaries
The corpora used in this investigation are the Helsinki Corpus of English
Texts (HC), A Representative Corpus of Historical English Registers 3
(ARCHER-3)2, the Lancaster-Oslo/Bergen Corpus (LOB), The Freiburg -
LOB Corpus of British English (FLOB), the Brown Corpus of Standard
American English (BROWN), the Freiburg-Brown Corpus of American
English (FROWN) and the British National Corpus (BNC). The main focus
is on the development of the lest-subjunctive construction in British Eng-
lish. However, in order to verify the claim by Quirk et al. that the construc-
tion is more common in American English, I will also look at some
twentieth-century American data, i.e. BROWN and FROWN.
This corpus study is restricted to adverbial clauses that are introduced
by the conjunction lest (and its variant spellings last(e), leste, les, lesse,
leest, lyst, lyste, leist least(e) according to the OED online) and the 3rd per-
son singular present tense. Even though the subjunctive-indicative contrast
is still given in Early Modern English (indicative: thou lookest and he
looketh; subjunctive: thou look and he look), in the periods following,
which will also be investigated in this paper, the distinction only survives
in the 3rd person singular present tense (except for the verb to be). As indi-
cated in the language forum discussion at the beginning of the paper and,
more importantly, by the linguists’ comments above, the subordinator lest
can either be followed by the inflectional subjunctive (Lest the situation
deteriorate), modal auxiliaries (Lest the situation should deteriorate), or
the indicative (Lest the situation deteriorates). Which of the three forms
the conjunction lest prefers will be investigated in the following section.
2.2. The lest-subjunctive construction – a corpus study
Table 1 presents the results of the conjunction lest followed by the three-
way distribution of the inflectional subjunctive, modal auxiliaries, and the
indicative from 1570 to 1949.3
Table 1. The diachronic development of lest-constructions and the three-way dis-
tribution (HC and ARCHER)
3rd p. 1570- 1640- 1700- 1750- 1800- 1850- 1900-
sing.
1640 1710 1749 1799 1849 1899 1949
Subjunc- 11 4 - - - - -
tive (73.3%) (44.4%)
Mod. 4 4 3 2 1 3 -
Aux. (26.7%) (44.4%) (100%) (100%) (100%) (100%)
Indica- - 1 - - - - -
tive (11.2%)
Total 15 9 3 2 1 3 0
The conjunction lest does not occur very frequently followed by 3rd person
singular present tense verbs.4 The data do however provide some indica-
tions: The conjunction lest was predominantly followed by the inflectional
subjunctive in the Early Modern English period, in particular in 1570 to
1640, taking up 73.3%. In 1640–1710 the use of the lest-subjunctive con-
struction had already dropped to 44.4%, which then equals the frequency
of the lest-modal auxiliary construction. From 1700 onwards we do no
longer come across a single instance of the lest-subjunctive construction.
Between 1700 and 1899 lest is exclusively followed by modal auxiliaries –
a result that does not agree with the statements by Charleston (1941: 139)
and Görlach (2001: 122) that lest triggers the inflectional subjunctive in the
154 Anita Auer
eighteenth century. As for lest-indicative constructions, I only came across

one instance in 1640–1710, which shows that this type of construction has
already been rare in Early Modern English and not only in present-day
English (cf. Huddleston and Pullum 2002: 1000). The corpus does not con-
tain any examples of the conjunction lest (3rd person singular present tense)
in the period 1900–1949 at all. The results in Table 1 indicate that the con-
junction lest contributed to the alleged death struggle of the subjunctive
rather than to its survival.
As regards Barber’s claim (1997: 173) that the inflectional subjunctive
“was part of everyday familiar speech, even among lower-class characters”
in Early Modern English, only 1 out of 15 lest-subjunctive constructions
has been attested between 1570 and 1710 that may be considered an exem-
plification of everyday speech. It occurs in the play A Chaste Maid in
Cheapside (1630) by Thomas Middleton, and in particular in the speech of
the Second Promoter, an informer, who may not necessarily belong to the
higher circles of society (see example 1).
(1) Either will serue, This Butcher shall kisse Newgate, lesse he turne vp the
Bottome of the Pocket of his Apron, You goe to seeke him? (HC 1630 Th.
Middleton, A Chaste Maid in Cheapside)
The other 14 examples containing the lest-subjunctive construction are

found in the following text types: Bible (3 instances), philosophy (3 in-
stances), educational treatises (2 instances), fiction (2 instances), hand-
books (2 instances), science (1 instance), and biography (1 instance). All
these text types are close to the formal end on a formal–informal contin-
uum, which suggests that the construction under discussion can already be
associated with formal style in the Early Modern English period.
A study of the development of the inflectional subjunctive in adverbial
clauses5 (introduced by a range of conjunctions, including lest) from 1570
onwards showed that the decline of the form was not so straightforward
(Auer 2006: 45). In fact, after a fairly rapid decline from 1570 to 1700 the
data show an increase in usage in the second half of the eighteenth century
and the first half of the nineteenth century. Thereafter we can observe fur-
ther decline of the subjunctive form in adverbial clauses. A comparison
between the development of the inflectional subjunctive following all con-
junctions and the conjunction lest only is provided in Figure 1 below.
The results show very clearly that lest did certainly not contribute to the
brief revival of the inflectional subjunctive in the Late Modern English
period. Auer (2006) ascribes this “blip in the diachronic development of
the subjunctive” to the influence of prescriptive grammars published in the

Late Modern English period which advocated the use of the inflectional
subjunctive. Whether the grammarians also recommended lest as a trigger
of the subjunctive form will be looked at in section 3.
80
70
60
50
% 40
30
20
10
-
1570-1640 1640-1710 1700-1749 1750-1799 1800-1849 1850-1899 1900-1949
all conjunctions lest
Figure 1. The development of the inflectional subjunctive – all conjunctions versus

lest (based on the Helsinki Corpus and ARCHER-3)
Was the era of the lest-subjunctive construction over by 1700 or were there
signs of a revival after 1950? After all, the construction has not been con-
sidered extinct by linguists in the twentieth and twenty-first centuries. Ta-
ble 2 provides results on the lest-construction and the three-way distribu-
tion in British English corpora from 1961 (LOB) and the 1990s (FLOB) as
well as comparative American corpora.
Table 2. Lest and the three-way distribution in British and American English
LOB FLOB BROWN FROWN
(BrE; (BrE; (AmE;1961) (AmE;
1961) 1990s) 1990s)
Subjunctive - 1 8 5
Modal Auxil- - 2 3 1
iaries
Indicative - - - -
TOTAL 0 3 11 6
The data clearly show that lest (in the 3rd person singular present tense) has
a very low frequency in all four corpora, even more so in British English
data (LOB/FLOB) than in American English data (BROWN/FROWN).6
156 Anita Auer
The results thus confirm the claims made by Quirk et al. (1985: 158; 1108)
that (a) the use of lest is archaic, (b) the construction “is more current in
AmE”, both in 1961 and in the 1990s, and (c) in AmE lest is mainly fol-
lowed by the present subjunctive.7
The only example of the lest-subjunctive construction in FLOB (BrE) is
taken from a novel and is thus categorised as general fiction (see example
2).
(2) She didn’t say this out loud, having become, she believed, the world champion
walker-on-eggs, always alert lest a chair come flying across the room aimed at
her head. (FLOB, K06)
Note that the BROWN and FROWN lest-subjunctive constructions, as

illustrated in (3), occur in a wide range of textual categories, namely (A)
reportage, (B) editorial, (F) popular lore, (G) Belles Lettres, biographies,
essays, (H) miscellaneous, (J) science, (K) general fiction, (N) adventure
and western, and (P) romance and love story. The only category that con-
tained more than 1 (AmE) lest-subjunctive example is (K) general fiction
with 3 out of 13 examples.
(3) I did not sleep much that night, which I spent struggling against the Kaiser,
dodging his submarines and holding him back in the trenches lest he storm
Paris. (FROWN, G39)
Even though the occurrence of lest in LOB/FLOB and BROWN/FROWN

is rather low, the data do show that lest was used with a subjunctive and
modal auxiliaries in British English in the 1990s. Considering that there is
no occurrence of lest in LOB (1961) but in FLOB (1990s), this might indi-
cate that the lest-construction experiences a revival. In order to confirm or
reject this very tentative assumption, I will investigate the BNC which does
not only focus on selected years but contains examples from more years
from 1960 to 1994.
The BNC search returned 411 lest occurrences out of which 185 lest-
constructions (3rd person singular present tense) can be investigated with
regard to the paradigmatic variation. Figure 2 shows that in the period
1960–1994 the lest-construction occurred 59.4% (110 instances) with the
inflectional subjunctive, 35.7% (66) with modal auxiliaries and 4.9% (9)
with the indicative form. Strang (1970: 209) claims that “[…] the subjunc-
tive was from the 15c, as it is now, largely a function of be.” This is how-
ever not the case with respect to lest-constructions as the BNC data show:
44.5% (49 instances) of the examples contain the verb be (see example 4)
whereas 55.5% (61) contain a lexical verb (see example 5).
Modal
Auxiliaries
35.7%
Inflectional
Subjunctive
59.4%
Indicative
4.9%
Figure 2. Lest and the three-way distribution in the BNC (N=185)
(4) Lest it be thought I am attributing vile motives to such people, it is noticeable

that when non-card communions are held, attendance is considerably dimin-
ished. (BNC B1J 1373)
(5) That premium should not grow too large lest it bring increasingly intolerable
cohabitation rules with it -- which is why the basic level of child benefit should
be raised first. (BNC ECB 1096)
Huddleston and Pullum’s (2002: 1000) remark that should is an alternative

form to the inflectional subjunctive in lest-constructions can be supported
as 66 instances contain 72.7% (48) should, followed by 12.1% (8) might,
7.6% (5) would, 6.1% (4) may, and 1.5% (1) could.
When the occurrence of the lest-constructions is divided into sub-
periods, the following development can be observed: The results depicted
in Figure 3 clearly show that lest-constructions followed by the inflectional
subjunctive experienced a revival taking up 86.3% (95 instances) during
the period 1985 to 1994. Considering this sudden increase, the question
arises in which text domains the construction occurs and whether lest is
confined to formal contexts, as Quirk et al. (1985) and Huddleston and
Pullum (2002) argue. The results show that 99.1% (109 instances) occur in
written texts and only 1 instance was found in spoken material. A subdivi-
sion in terms of text domain shows the results presented in Table 3.
158 Anita Auer
100
90 86.3
80
70
60
% 50
40
30
20
10
5.5 5.5
0
1960-1974 1975-1984 1985-1994
Figure 3. Diachronic development of lest and the subjunctive (N=110)8
Table 3. Lest and the subjunctive in different text domains

Text domain lest & subjunctive
(N=110)
Written Arts 4.5% (5)
Belief and thought 11.8% (13)
Imaginative 32.7% (36)
World affairs 15.5% (17)
Commerce and finance 7.3% (8)
Social science 16.4% (18)
Leisure 5.5% (6)
Applied science 3.6% (4)
Natural and pure sciences 1.8% (2)
Spoken Education / Informative 0.9% (1)
The construction under investigation is most frequently found in so-called

imaginative texts, i.e. 32.7% (36 instances). Other text domains in which
lest and the subjunctive occur fairly frequently are social science with
16.4% (18), world affairs with 15.5% (17), and belief and thought with
11.8% (13). In the BNC the category imaginative texts includes novels,
short stories, poems, and drama scripts.9 As Lee (2001: 53) points out, the
sub-genre information, i.e. whether a text is poetry or a short story, “is not
recorded in the file headers or in any documentation associated with the
BNC” and this makes it difficult or even impossible for the researcher to
retrieve more detailed information. This is certainly the case with this
study as I am unable to find out whether the examples can be categorised
as formal or informal texts. The text domains social science, i.e. academic
prose, world affairs, and belief and thought are categorised as informative
written texts (Lee 2001: 50), which indicates that these texts are more
likely to be situated on the formal end of the continuum.
As regards the distribution according to gender (in written texts only),
men use the lest-subjunctive-construction slightly more frequently than
women; men use it 39.4% (43 instances) and women use it 31.2% (34 in-
stances).10
An analysis of the distribution according to age shows that the use of
lest-subjunctive-constructions is largely confined to writers aged 35 to 60+
(see Table 4 below).
Table 4. The distribution of lest and the subjunctive according to age
Unknown 0-14 15-24 25-34 35-44 45-59 60+
66.1% (72) - - 1.8% (2) 11% (12) 10.1% (11) 11% (12)
As age and gender of several writers is labelled ‘unknown’, a correlation

between these two variables amounts to rather low figures, as can be seen
in Table 5.
Table 5. The distribution of the lest-subjunctive-construction according to age and
gender
25-34 35-44 45-59 60+
Male 2.7% (1) 24.3% (9) 16.2% (6) 18.9% (7)
Female 2.7% (1) 8.1% (3) 10.8% (4) 13.6%(5)
Unknown - - 2.7% (1) -
The results show that the construction is used most frequently by men aged
35–44, followed by men aged 60+ and then by men aged 45–59. As regards
women’s use of the lest-subjunctive construction, it was most frequently
used by women aged 60+, followed by the age range 45–59 and then 35–
44. As the numbers are so low, the results would not be considered statisti-
cally significant. The data do however reveal that men and women over the
age of 35 use the construction in written language. This result may reflect
the sampling choices made by the BNC compilers, i.e. the sample of, for
instance, imaginative texts written by people under 35 would presumably
be rather small.
160 Anita Auer
This corpus-based study of the lest-subjunctive construction showed

that the construction was in use in the Early Modern English period but
then disappeared. After a hibernation period of 250 years, we were able to
observe a slow increase starting in the 1960s and then an extremely sharp
rise between 1985 and 1994.
3. Lest and the subjunctive – a match made in (grammar) heaven?
One of the comments made in the online language forum was that the sub-
junctive was the correct form following lest. The use of the term correct
suggests that there are/were formal rules that a language user should be
aware of. I will not focus on the issue of correctness here but investigate
meta-linguistic comments by grammarians and language-guardians from
Early Modern English onwards to find out (a) whether the combination of
the conjunction lest and subjunctive is mentioned at all, and if so, (b)
whether these comments can shed some light on the historical development
of the construction.
As Dons (2004: 222) points out, “[t]he government of conjunctions
used to be more important in Early Modern English than it is today, be-
cause the subjunctive was much more frequent.” This is also reflected in
Early Modern English grammars which contain comments “in more or less
detail on various conjunctions and their moods”. The usage data from the
Early Modern English period (Table 1) clearly showed that the conjunction
lest was followed by the inflectional subjunctive, which raises the question
of whether grammarians in the Early Modern English period have also
pointed out the government of lest in their works. According to Dons
(2004: 225), who analysed a wide range of Early Modern English gram-
mars, “[t]he conjunctions if and although are mentioned most frequently,
while unless, provided, ere, before, and in case occur only once or twice.”
Moreover, Table 43 in Dons’s (2004: 224) monograph contains a list of all
the conjunctions mentioned in the grammars, which does however not con-
tain the conjunction lest. Note that not all grammarians in the Early Mod-
ern English period accepted the subjunctive as a mood,11 which is why
some grammars do not contain accounts of the subjunctive, e.g. Greaves
(1594), Jonson (1640), and Wallis (1653), according to Dons (2004: 106).
To illustrate selected accounts of grammarians who do consider the sub-
junctive a mood, I will first have a look at Poole (1646), who notes that
[t]he Subjunctive mood hath some such conjunctions as these joyned with it,
when if, since that, because: as, when I did love, if I live, since that you have
heard, because he was sick. (Poole 1646: 10)
Poole points out that the subjunctive is linked with certain conjunctions,
one of which is if, according to the examples he provides. He does however
not describe the meaning of the subjunctive mood and from his examples it
appears that Poole does not recognise the form of the subjunctive either.
Lye in the Childs Delight (1671) describes the subjunctive mood as fol-
lows:
The Subjunctive, which is join-ed with, or depends upon another Verb in the
same sentence. When it is used in wishing, ‘tis called the Optative Mood;
but the Potential, when it hath the signs may, can, might, would, &c. I may,
or can turn. (Lye 1671: 120)
Following this account, Lye only provides examples of the potential to
illustrate the subjunctive mood, i.e. examples containing auxiliary verbs. I
did not come across an inflectional subjunctive example, let alone an ex-
ample with lest or a comment on lest being a trigger of the subjunctive.
The subjunctive account in Miège’s English Grammar (1688) serves as
the final illustration from the Early Modern English period. Miège (1688)
notes the following:
The Subjunctive Mood is much the same in English with the Indicative; but
that it is used with Conjunctions before it, from whence it has got the Name
of Subjunctive. ‘Tis also called Optative, because in it we make Wishes; and
likewise Potential, for that the Verb May and Might are used in this Mood.
Thus, with the Conjunction If, Though, or Provided, the Present Tense is
Conjugated in the following manner. (Miège 1688: 54)
Miège points out that the subjunctive is used with conjunctions such as if,
though, and provided. Following the description of the subjunctive, he
illustrates the mood with the inflectional subjunctive (If I love, If thou
lovest, If he love, etc.) and modal auxiliaries, i.e. the potential mood (That I
may love, That thou mayst love, That he may love, etc.).
The selected accounts from the Early Modern English period showed
that those grammarians who recognised the subjunctive as a mood did not
describe it in a uniform fashion. It is indicated that conjunctions serve as
triggers of the subjunctive but the conjunction lest is not explicitly listed.
This evidence suggests that the use of the lest-subjunctive construction in
the Early Modern English period could not have been triggered by gram-
marians’ comments.
162 Anita Auer
According to the corpus data, in the eighteenth century there were no

lest-subjunctive constructions but only lest-modal auxiliary constructions.
To find out whether grammarians in the eighteenth century commented on
either of these constructions, I investigated a range of grammars that repre-
sent both the pedagogic and philosophical grammar writing tradition (cf.
Auer 2004). The grammars investigated from the first half of the eighteenth
century did not contain comments on the conjunction lest and its relation-
ship to the subjunctive and/or modal auxiliaries. Accounts containing ei-
ther a description or an example of lest and the subjunctive are contained
in Table 6 below.
Table 6. Lest and subjunctive accounts in eighteenth-century grammars of English

Grammarian Year
Bayly 1758 p. 105 – The first English Writers were more accu-
(5th ed.) rate than the Modern in distinguishing the Subjunc-
tive from the Indicative: With if, unless, except, lest,
that, though, if so be, until – they always used the
Verb in the second and third Person Singular with-
out any Change.
Ash 1761 p. 82 – The Conjunctions if, though, except, &c.
Institutes implying a manifest Doubt or Uncertainty, require
the subjunctive Form of the Verb: as, “Tho’ he slay
me, yet will I trust in him. I will not let thee go ex-
cept thou bless me. Kiss the Son lest he be angry. If
he but speak the Word. See thou do it not.”
White 1761 p. 10 – This Mood is called by us the Subjunctive,
because there are certain words, in our language, to
which it is generally subjoined, or which it generally
follows.
These words are, before, ere, except, however, if,
lest, so, tho’, till or until, whatsoever or whatever,
whether, whosoever or whoever; and perhaps a few
more.
p. 117 – The Subjunctive Mood, in its Present
Tense, is also annext to Lest.
“Bid him bring his pow’r before sun-rising, lest his
Son George fall into the blind cave of eternal night.”
SHAKESPEAR.
“Go, lest the haughty partner of my sway with jeal-
ous eye thy close access survey.” POPE.
“This let him know, lest, wilfully transgressing, he
pretend surprisal; unadmonish’d, unforewarn’d.”
MILTON.
“Reason he made right; but bid her well beware, and
still erect, lest, by some fair-appearing good sur-
priz’d, she dictate false, and misinform the will.”
IDEM.
Lowth 1762 pp. 142–143 – Lest; and that with a Negative follow-
ing it; and if with but following it; necessarily re-
quire the Subjunctive Mode: Examples; “Let him
that standeth, take heed, lest he fall.” I Cor. x. 12.
“Take heed, that thou speak not to Jacob.” Gen. “If
he do but touch the hills, they shall smoke.” Ps. civ.
32.
Ward 1767 p. 127 – Of the Conjunctions and Indefinites which
require a Verb in the Subjunctive Mood.
To express uncertain States, except, lest, so, before,
ere, till, if, howsoever, though, although, unless,
with who, - and whatsoe’er, and whether, the Sub-
junctive Mood prefer.
The Subjunctive Mood is frequently used with the
Conjunctions and Indefinites mentioned in the Rule;
as, [...] Let us sacrifice to the Lord, LEST he FALL
upon us with Pestilence. Ibid. [Old Testament] [...]
Shaw 1778 pp. 105–106 – Rule V. The conjunctions if, though,
except, lest, before, ere, till, howsoever, unless,
whether, with the indefinites whosoever and whatso-
ever, frequently govern a subjunctive mood, when
the sense is doubtful or uncertain; as, [...]
Rule VI. The conjunctions lest and that annexed to a
command preceding, and if with but following it,
govern a subjunctive mood; as, Let him that standeth
take heed lest he fall. See that thou do it not. If he do
but touch the hills, they smoke.
Murray 1795 pp. 128–129 – Lest and that, annexed to a command
preceding, and if with but following it, necessarily
require the subjunctive mode; as, “Let him that
standeth take heed lest he fall;” “Take heed that thou
speak not to Jacob;” “If he do but touch the hills,
they shall smoke.”
Gardiner 1799 p. 90 – The conjunctions lest, that, and till, require
sometimes the subjunctive: as, “Love not sleep lest
thou come to poverty;” “Take heed that thou speak
not.”
164 Anita Auer
Bayly (1758: 105) lists the conjunction lest as a trigger of the subjunctive
whereas Ash (1761: 82) is the first grammarian to illustrate the lest-
subjunctive-construction (Kiss the Son lest he be angry). In the same year,
White (1761: 117) lists lest as a trigger of the subjunctive and provides
examples from renowned writers such as Shakespeare (1564–1616), Milton
(1608–1674) and Pope (1688–1744). It is noteworthy that the remaining
lest examples I came across in eighteenth-century grammars are taken from
the Bible, both the Old and the New Testament. In the 1762 edition of
Lowth’s grammar he illustrates the construction with “Let him that
standeth, take heed, lest he fall”, which is taken from the first epistle of
Paul the Apostle to the Corinthians (New Testament). This exact example
and examples with other conjunctions that were first used in Lowth’s
grammar can also be found in Shaw (1778) and Murray (1795).12 Lowth
(1762) commented on the conjunctions lest and that and their governing
the subjunctive mood separately, an approach that was also adopted by
Shaw (1778), Murray (1795), and Gardiner (1799). All four of these
grammarians emphasise that lest necessarily requires the subjunctive,
which is illustrated with the inflectional subjunctive and not with modal
auxiliaries. Even though these grammars emphasise the lest-subjunctive
construction and the account is formulated as a rule to be observed, this is
not mirrored in actual usage as the corpus results showed.
An investigation of meta-linguistic comments from the nineteenth cen-
tury led to the following results: Most grammarians do not mention lest in
conjunction with the subjunctive, as exemplified by Leigh (1840: 86), who
states that the subjunctive “is preceded by a conjunction, expressed or un-
derstood”. Other grammarians such as Sweet (1892) describe the conjunc-
tion lest as expressing negative purpose or avoidance as in “they took away
the knife lest he should cut himself”. Lest is here followed by the modal
auxiliary should rather than the inflectional subjunctive. Those nineteenth-
century grammarians who did comment on the government between certain
conjunctions and the subjunctive appeared to somehow challenge their
eighteenth-century colleagues on this topic. Cobbett (1818) for instance
states that
Bishop Lowth, and on his authority, Lindley Murray, have said, that some
conjunctions have a government of verbs; that is to say, make them or force
them to be in the subjunctive mode. And then these gentlemen mention par-
ticularly the conjunctions, if, though, unless, and some others. But (and
these gentlemen allow it) the verbs that follow these conjunctions are not
always in the subjunctive mode; and, the using of that mode must depend,
not upon the conjunction, but upon the sense of the whole sentence. (Cob-
bett 1818: 146–147)
Similarly, Foster and Foster (1858) note that
Our earlier grammarians laid it down that ‘some conjunctions require the in-
dicative, and some the subjunctive, mood after them;’ and, whether in obe-
dience of them, or from some more remote cause which we have not pene-
trated. Scotch writers almost invariably use the subjunctive with the
conjunctions if, lest, although, and whether, whatever the sense may be.
Subsequent grammarians have with much more accuracy decided, that when
a matter is contingent and future, the subjunctive is required; but the indica-
tive, if the thing is in itself certain, whatever the dubiety of the speaker con-
cerning it. (Foster and Foster 1858: 239)
Earle (1898: 131) argued along the same lines that “[t]he frequent connec-
tion between certain conjunctions and the subjunctive must not induce us
to think as if they caused the Mood, for indeed a little observation will
show us that this relation is by no means constant.”
This survey of meta-linguistic comments during the Early and Late
Modern English periods revealed that between the publication of the earli-
est English grammars and the end of the eighteenth century, grammarians
recognised a connection between certain conjunctions and the use of the
inflectional subjunctive. The lest-subjunctive construction in particular was
however only pointed out and strongly advocated by grammarians in the
second half of the eighteenth century. Nineteenth-century grammarians
distanced themselves from such claims by emphasising that it is not the
conjunction but the meaning of the entire sentence that decides which
mood should be used.
4. Conclusion
It was the aim of this paper to trace the development of the conjunction lest
as a trigger of the inflectional subjunctive. An investigation of diachronic
as well as synchronic corpora has shown that the construction was still
used in Early Modern English, then disappeared for 250 years, and it has
experienced an enormous revival between 1985–1994 (the end date of the
study). The analysis of meta-linguistic comments by grammarians and lan-
guage-guardians exhibited that only eighteenth-century grammarians were
particularly concerned with emphasising that lest necessarily required the
inflectional subjunctive. In the period before, i.e. in Early Modern English,
166 Anita Auer
grammarians still struggled to come to terms with the concept of mood in

the English language, and in the period following the eighteenth century,
grammarians were concerned with distancing themselves from claims made
by eighteenth-century prescriptivists and emphasised their descriptive ap-
proach. For all that, this development in the grammar writing tradition was
not reflected by actual language usage.
One important question remains to be answered and this concerns the
revival of the lest-subjunctive construction in British English between the
1960s and 1994 (and possibly beyond). The increase may be explained by
looking at the recent development of the mandative subjunctive in British
English. Övergaard (1995), Serpollet (2001), and Leech et al. (forthc.) are
all concerned with the development of the mandative subjunctive, i.e. the
subjunctive in subordinate that-clauses, the main clause of which expresses
recommendation, demand, proposal, resolution, intention (cf. Quirk et al.
1985: 156–157), in American and British English in the twentieth century.
Throughout the twentieth century, the inflectionally marked subjunctive
increased (at different times) in both varieties of English – the rise first
occurred in American English and was then observed in British English.
The much later occurring increase of the subjunctive form in British Eng-
lish is considered to have been influenced by American English, in particu-
lar through the mass media after World War II, according to Övergaard
(1995: 89). Serpollet (2001: 531) also argues that even though the Ameri-
can data show that the increase of the mandative subjunctive has slowed
down, it appears to have had an influence on British English. The increase
of the lest-subjunctive construction does in fact resemble the rise of the
mandative subjunctive in British English as the comparison of frequencies
in LOB (12.6%) and FLOB (38.3%) in Leech et al. shows (forthc.: Ch. 3;
cf. Serpollet 2001). Note also that the lest-subjunctive construction gained
ground in American English as the BROWN and FROWN data showed in
Table 2 before it reappeared in British English. I will here conclude that
the lest-subjunctive construction has clearly made a revival which strongly
resembles the recent development of the mandative subjunctive. As for the
fate of the inflectional subjunctive, the form appears to have taken up a
long-neglected activity in its retirement (cf. Peters 1998: 101).
Acknowledgements
This paper was written in the context of the NWO-project The Codifiers
and the English Language: Tracing the Norms of Standard English (Uni-
versity of Leiden, The Netherlands). I would like to thank Marianne Hundt,

Lilo Moessner, Karlijn Navest, and Michiel de Vaan for comments and
suggestions on an earlier version of this paper.
Notes
1. Lest is usually classified as an adverbial subordinator that expresses negative

purpose, which Kortmann (1997: 86) describes as “not-p is an intended result
or consequence of q that is yet to be achieved”. The subordinator developed
from the Old English phrasal connective þy læs (þe), which is “a combination
of the instrumental form of the demonstrative þæt, the adverb læs ‘less’ and
the optional subordinating particle þe” into monomorphemic lest through the
process of univerbation (López-Couso 2007: 12).
2. ARCHER-3 = A Representative Corpus of Historical English Registers 3.
1990-1993/2002/2007. Compiled under the supervision of Douglas Biber and
Edward Finegan at Northern Arizona University, University of Southern Cali-
fornia, University of Freiburg, University of Heidelberg, University of Hel-
sinki, Uppsala University, University of Michigan, and University of Man-
chester.
3. Note that the Helsinki corpus compilers subdivided the data into 70-year time
spans (1570–1640; 1640–1710) whereas the ARCHER data consists of 50-
year time spans, i.e. starting with 1700–1749 up to 1900–1949. As the sizes of
the corpora differ, a comparison of frequencies will mostly be based on per-
centage figures rather than absolute numbers.
4. López-Couso (2007) investigated the distribution of lest with respect to its
functions of (a) expressing negative purpose and (b) serving as a complemen-
tiser from Old English to the present day (850–1990) based on the Helsinki
Corpus and ARCHER-2. López-Couso’s results are not restricted to a certain
person and tense (as is the case in the current study). She (López-Couso 2007:
14) notes that there is an “overall low frequency of occurrence (only 322 in-
stances in circa 3.4 million words ...” and that “the connective lest has been at-
tested in all periods of the history of English.”
5. This study is based on the Helsinki Corpus and ARCHER-3.
6. As regards the general occurrence of lest in British and American English,
Algeo (2006: 202) notes the following: “In the mid twentieth century, lest was
apparently 5 times more frequent in American English than in British, the ratio
in the Brown and LOB corpora being 17:3. CIC [Cambridge International
Corpus], now shows lest to be actually more frequent in British use than in
American. It has 53.7 iptmw [instances per ten million words] in British texts,
mainly fiction, and 32.6 in American texts, mainly academic.”
168 Anita Auer
7. Peters (1998: 97) investigated the mandative subjunctive and the subjunctive
in adverbial clauses in BROWN, LOB, and the Australian ACE corpus, the
latter of which contained 3 lest-subjunctive examples.
8. The publication date of three examples (2.7%) was unknown, which is why the
percentage figures do not add up to 100%.
9. Rohdenburg and Schlüter (forthc.) investigated the subjunctive use following
lest in selected British and American newspapers: “The results confirm that
the subjunctive is virtually obligatory in this context in AmE, and that BrE has
already caught up to a considerable extent.” This is shown “by the clear differ-
ence between the earlier (1990–1992) and later (2001–2004) years of The
Times, which have 58% and 77% of subjunctives, respectively.”
10. Note that 22% (24 instances) of the writers are unknown and 7.4% (8 in-
stances) are labelled as ‘mixed’.
11. For a description of the situation in the eighteenth century see Auer (2004).
12. It was shown that Murray leant heavily on Lowth’s work when writing his
grammar (Vorlat 1959; Tieken-Boon van Ostade 1996).
References
Algeo, John
2006 British or American English? A Handbook of Word and Grammar
Patterns. Cambridge: Cambridge University Press.
Ash, John
1761 Grammatical Institutes, or an Easy Introduction to Dr. Lowth’s
English Grammar: Designed for the Use of Schools, and to Lead
Young Gentlemen and Ladies into the Knowledge of the First Prin-
ciples of the English Language. London.
Auer, Anita
2004 The treatment of the subjunctive in eighteenth-century grammars of
English. Paradigm 2 (8): 3–18.
2006 Precept and practice: The influence of prescriptivism on the English
subjunctive. In Syntax, Style and Grammatical Norms: English from
1500–2000, Christiane Dalton-Puffer, Dieter Kastovsky, Nikolaus
Ritt and Herbert Schendl (eds.), 33–53. (Linguistic Insights.) Frank-
furt/Bern: Peter Lang.
Auer, Anita and Victorina González-Díaz
2005 Eighteenth-century prescriptivism in English: A Re-evaluation of its
effects on actual language usage. Multilingua 24 (4): 317–341.
Barber, Charles Laurence
1997 Early Modern English. Edinburgh: Edinburgh University Press.
Bayly, Anselm
1758 An Introduction to Languages, Literary and Philosophical. London.
Charleston, Britta Marian
1941 Studies on the Syntax of the English Verb. Bern: Francke.
Cobbett, William
1818 A Grammar of the English Language. New York.
Dons, Ute
2004 Descriptive Adequacy of Early Modern English Grammars. Berlin;
New York: Mouton de Gruyter.
Earle, John
1898 A Simple Grammar of English Now in Use. New York: G.P. Put-
nam’s Sons.
Early English Books Online (EEBO)
Available at: http://eebo.chadwyck.com/home
Fillbrandt, Eva-Liisa
2006 The development of the mandative subjunctive in the Early Modern
English Period. Trames 10 (60/55), 2: 135–151.
Foster, Alexander and Margaret E. Foster
1858 Points in English Grammar. The London Review 10: 223–244. [Re-
printed in English Language and Language-Teaching 1800–1865,
Vol. 1, Roy Harris (ed.), 223–244 (1995). London: Routledge/
Thoemmes.]
Fowler, Henry Watson
1965 A Dictionary of Modern English Usage. New York: Oxford Univer-
sity Press.
Gardiner, Jane
1799 The Young Ladies’ English Grammar; Adapted to the Different
Classes of Learners. York.
Görlach, Manfred
1999 English in Nineteenth-century England: An Introduction. Cam-
bridge: Cambridge University Press.
2001 Eighteenth-century English. Heidelberg: C. Winter.
Greaves, Paul
1969 Reprint. Grammatica Anglicana. Menston: The Scholar Press Lim-
ited. Original edition, Canterbury: Legatt, 1594.
Grund, Peter and Terry Walker
2006 The subjunctive in adverbial clauses in nineteenth-century English.
In Nineteenth-century English. Stability and Change, Merja Kytö,
Mats Rydén and Erik Smitterberg (eds.), 89–109. Cambridge: Cam-
Haegeman, Liliane
1986 The present subjunctive in contemporary British English. Studia
170 Anita Auer
Huddleston, Rodney and Geoffrey K. Pullum

Hundt, Marianne
1998 It is important that this study (should) be based on the analysis of
parallel corpora: On the use of the mandative subjunctive in four ma-
jor varieties of English. In The Major Varieties of English. Papers
from MAVEN 97, Växjö 20–22 November 1997, Hans Lindquest,
Staffan Klintborg, Magnus Levin and Maria Estling (eds.), 159–175.
(Acta Wexionensia Humaniora 1.) Växjö University.
forthc. Colonial lag, colonial innovation, or simply language change? In One
Language, Two Grammars? Differences between British and Ameri-
can English, Günther Rohdenburg and Julia Schlüter (eds.). (Studies
in English Language.) Cambridge: Cambridge University Press.
Jonson, Ben
1972 The English Grammar. Menston: The Scholar Press Limited. Origi-
nal edition, London: Bishop, 1640.
Kortmann, Bernd
1997 Adverbial Subordination. A Typology and History of Adverbial Sub-
ordinators Based on European Languages. (Empirical Approaches
to Language Typology, 18). Berlin; New York: Mouton de Gruyter.
Lee, David Y.W.
2001 Genres, registers, text types, domains, and styles: Clarifying the con-
cepts and navigating a path though the BNC jungle. Language
Learning & Technology 5 (3): 37–72.
Leech, Geoffrey, Christian Mair, Marianne Hundt and Nicholas Smith
forthc. Changes in Contemporary English. Cambridge: Cambridge Univer-
sity Press.
Leigh, Percival
1840 The Comic English Grammar; A New and Facetious Introduction to
the English Tongue. London: R. Bentley.
López-Couso, María José
2007 Adverbial connectives within and beyond adverbial subordination:
The history of lest. In Connectives in the History of English, Ursula
Lenker and Anneli-Meurman-Solin (eds.). (Current Issues in Linguis-
tic Theory.), 11–29. Amsterdam: Benjamins.
Lowth, Robert
1762 A Short Introduction to English Grammar. London.
Lye, Thomas
1671 The Childs delight together with an English Grammar. London: S.
Simmons. [Retrieved from Early English Books Online]
Miège, Guy
1688 The English grammar, or, The Grounds and Genius of the English
Tongue. London: J. Redmayne. [Retrieved from Early English Books
Online]
Middleton, Thomas
1630 A Chast Mayd in Cheape-Side. A pleasant Conceited Comedy Neuer
Before Printed. As It Hath Beene Often Acted at the Swan on the
Banke-side, by the Lady Elizabeth Her Seruants. London. [Retrieved
from Early English Books Online]
Moessner, Lilo
2002a The contribution of Scots to the development of the subjunctive in
standard English. Paper presented at the ICEHL 2002 in Glasgow.
2002b Who used the subjunctive in the 17th Century? In Language – Con-
text and Cognition, Sybil Scholz and Ute Breidenbach (eds.), 227–
235. München: Langenscheidt-Longman.
2006 The subjunctive in Early Modern English adverbial clauses. In Cor-
pora and the History of English. Papers Dedicated to Manfred Mar-
kus on the Occasion of His Sixty-Fifth Birthday, Christian Mair and
Reinhard Heuberger (eds.), 249–263. Heidelberg: Winter.
2007 The mandative subjunctive in Middle English. In Studies in Middle
English Forms and Meanings, Gabriella Mazzon (ed.), 209–226.
Frankfurt am Main: Peter Lang.
Murray, Lindley
1795 English Grammar, Adapted to the Different Classes of Learners.
York.
The Oxford English Dictionary
1989 (2nd Ed.). Oxford: Oxford University Press. [Available at:
www.oed.com].
Övergaard, Gerd
1995 The Mandative Subjunctive in American and British English in the
20th Century. Stockholm: Almqvist & Wiksell.
Peters, Pam
1998 The survival of the subjunctive. Evidence of its use in Australia and
elsewhere. English World-Wide 19 (1): 87–103.
Poole, Joshua
1646 The English Accidence, or, A Short, Plaine, and Easie Way for the
More Speedy Attaining to the Latine Tongue, by the Help of the Eng-
lish. London: Printed by R.C. for Henry Seile and Richard Lownes.
[Retrieved from Early English Books Online]
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik
1985 A Comprehensive Grammar of the English Language. London:
Longman.
172 Anita Auer
Redfern, Richard K.
2001 Can the English language take care of itself? A dialogue. The English
Journal 90 (4): 60–66.
Rohdenburg, Günther and Julia Schlüter
forth. Chapter 19: New departures. In One Language, Two Grammars?
Differences between British and American English, Günther Roh-
denburg and Julia Schlüter (eds.). (Studies in English Language.)
Savage, W.H.
1833 The Vulgarisms and Improprieties of the English Language. London:
T.S. Porter.
Serpollet, Noëlle
2001 The mandative subjunctive in British English seems to be alive and
kicking… Is this due to the influence of American English? In Pro-
ceedings of the Corpus Linguistics 2001 Conference. Technical Pa-
pers, Vol. 13, Paul Rayson, Andrew Wilson, Tony McEnery, Andrew
Hardie and Shereen Khoja (eds.), 531–542. Lancaster: University
Centre for Computer Corpus Research on Language.
Shaw, John
1778 A Methodological English Grammar. York and Manchester.
Strang, Barbara
1970 A History of English. London: Methuen.
Sweet, Henry
1892 A New English Grammar Logical and Historical. Part I – Introduc-
tion, Phonology, and Accidence. Oxford: Clarendon Press.
Tieken-Boon van Ostade, Ingrid
1996 Lindley Murray and the concept of plagiarism? In Two Hundred
Years of Lindley Murray, Ingrid Tieken-Boon van Ostade (ed.), 81–
96. Münster: Nodus Publikationen.
Traugott, Elisabeth Closs
1972 A History of English Syntax. A Transformational Approach to the
History of English Sentence Structure. New York: Holt, Rinehart and
Winston.
Vorlat, Emma
1959 The sources of Lindley Murray’s ‘The English Grammar’. Leuvense
Bijdragen 48: 108–125.
Wallis, John
1969 Reprint. Grammatica Linguae Anglicanae. Menston: The Scholar
Press Limited. [Original edition, Oxford: Lichfield, 1653.]
Ward, William
1767 A Grammar of the English Language in Two Treatises. York.
White, James
1761 The English Verb. London.
Chapter 9
The BBC Advisory Committee on Spoken English
or How (not) to construct a ‘standard’
pronunciation
Jürg Rainer Schwyter
1. Introduction
The BBC started life as the British Broadcasting Company (note: Com-
pany, not yet Corporation) in 1922. Its “influence”, the social historian Asa
Briggs wrote, “both on education and the pronunciation of ‘standard Eng-
lish’ has been noticed by almost all the people who have described its
work” (1995: 222). In order to understand how the BBC had gained such a
uniquely powerful position I will examine – with the help of linguistic
examples, historical BBC documents and articles from the BBC weekly
magazine Radio Times – the language policy of the BBC Advisory Com-
mittee on Spoken English 1926–1939, especially its (ultimately failed)
attempt at fixing and diffusing a uniform standard of pronunciation; I will
also try to give some impressions of the public’s reaction to that policy.
Although my main focus will be on the early days of BBC language policy,
as “early days are crucial ones in either individual existence or corporate
organisation” (Reith 1924: 24), it will also become clear that some aspects
of the Advisory Committee’s work can still be felt today, for example with
respect to the pronunciation of proper names as well as in terms of public
notions of “correctness”, whatever changes in actual language policy the
BBC has introduced in more recent years.
2. The foundation of the BBC Advisory Committee on Spoken

English
From very early on the BBC was meant to be a tool not only to entertain, in
the most positive sense, but also to educate “the great multitude”, namely
by “carry[ing] into the greatest possible number of homes everything that
is best in every department of human knowledge” (Reith 1924: 34). “It
176 Jürg Rainer Schwyter
followed naturally”, as Briggs has pointed out, “that genuine differences of

opinion would be expressed about what constituted ‘the best’”; but John
Reith, the BBC’s first managing director, tried to solve this problem by
creating a “network of advisory committees which drew upon the services
of experts in various fields” (Briggs 1995: 218–219). One such committee
was the Advisory Committee on Spoken English. Its formation in 1926 was
considered absolutely essential as in the words of Reith summarised in the
minutes of a preliminary meeting held on 25 June,
it was becoming increasingly apparent that large numbers of words in com-
mon use were pronounced by educated people in entirely different ways, and
in many cases it was practically impossible to say which was best. Further,
in addition to the particular question of vowel sounds or accents in various
words, there arose such general questions as the final and medial “r” in
southern English and various common usages by educated people which in
some quarters might be regarded as being as wrong and unfortunate as the
practices of (say) the Cockney.
The Broadcasting Company felt that they should secure the most expert
advice to the intent that, if possible, a standard form of pronunciation for
doubtful words should be settled [...] and a ruling given as to whether such
modern customs of educated people as those mentioned above should be ac-
cepted as justified and authorised, or whether the Broadcasting Company
should by their powerful example endeavour to stem modern tendencies to
inaccurate and slurred speech. (R6 / 201 / 1)
The above quotation is revealing for a number of reasons. On the one hand,
the apparent impossibility of clearly identifying the “best” pronunciation –
note, here it is again, “the best” – shows that, in the 1920s, Reith was sim-
ply confronted with the same problem already stated, towards the end of
the 19th century, by Alexander J. Ellis and Henry Sweet, namely that
“even among educated London speakers” there is quite considerable varia-
tion “from individual to individual, and more markedly from generation to
generation” (quoted by Mugglestone 2003: 259, 261).1 On the other hand,
we not only witness here an expression of Reith’s firm belief in the possi-
bility as well as necessity of fixing, of standardising pronunciation, but
also an example of what Jean Aitchison (2001: 13) has called a “vintage
year” view of language; or as Gimson (1994: 80) put it a little more techni-
cally, “within RP, those habits of pronunciation that are most firmly estab-
lished tend to be regarded as ‘correct’, whilst innovation tends to be stig-
matized”.
The BBC Advisory Committee on Spoken English 177
Two additional points, however, should be made here. First, “the public
image” of the BBC was very much one “drawn from upper-class or upper
middle-class life” (Briggs 1995: 167). The broadcasters, “mostly young
men”, were in “a large portion [...] University undergraduates” who, before
the microphone, should speak “good English and without affectation”
(Reith 1924: 37, 51, 162). Therefore, the most appropriate medium for the
BBC, it was agreed, was the Public School Pronunciation – eventually
renamed Received Pronunciation or RP – as this accent “would convey a
suitable sense of sobriety, impartiality, and impersonality” (McArthur
1992: 110).2
The second point is this: it may be asked whether in fact radio and TV
can influence people’s speech behaviour and, if so, to what extent. Labov
and Harris (1986: 20), for example, have claimed that “linguistic traits are
not transmitted across group boundaries simply by exposure to other dia-
lects in the mass media or [even] in schools” – with the exception of a few
self-conscious corrections, sometimes hypercorrections, in formal styles.
Trudgill (1986: 39–41) explains why: as accommodation seems the most
likely explanation for the spread of linguistic features from speaker to
speaker, that is the conscious or unconscious convergence of a speaker to
the speech of his/her interlocutors, face-to-face contact is obviously a pre-
requisite for linguistic diffusion to take place. It is thus clear that “the elec-
tronic media are not very instrumental in the diffusion of linguistic innova-
tions, in spite of widespread popular notions to the contrary”, a fact
supported by “the geographical patterns associated with linguistic diffu-
sion”:
Were nationwide radio and television the major source of this diffusion,
then the whole of Britain would be influenced by a particular innovation si-
multaneously. This of course is not what happens: London-based innova-
tions reach Norwich before they reach Sheffield, and Sheffield before they
reach Newcastle. (Trudgill 1986: 40)
By contrast, neither the broader public nor Reith and his contemporaries at
the BBC seem to have doubted the Company’s mission and success.
Though Reith admitted that it was impossible to “compute in concrete
terms” the influence broadcasting would have on people’s pronunciation,
he insisted that “there is an influence, and a great one”: “children in par-
ticular have acquired the habit of copying the announcer’s articulation; this
has been observed by their teachers” (1924: 183, 162).
The front page of the Advisory Committee’s minute book not only re-
peats the brief mission statement, but also illustrates what a most distin-
guished body this actually was:3
The British Broadcasting Company, recognising their responsibility in
setting a standard of spoken English, have appointed the following to act as
an Advisory Committee: –
Dr. Robert Bridges [Poet Laureate since 1913 and a founder of the Soci-
ety for Pure English; Bridges became chairman of the Advisory Committee];
Mr. Logan Pearsall Smith [a naturalised American literary scholar and
essayist representing, on the Committee, the Society for Pure English, which
he co-founded with Bridges and others];
Mr. G. Bernard Shaw [the Irish playwright and critic, whose keen interest
in phonetics and spelling reform was well known];
Mr. Daniel Jones [Professor of Phonetics at University College London
and compiler of the English Pronouncing Dictionary, first published in
1917];
Sir Johnston Forbes-Robertson [a well-known actor and theatre manager,
who was “one of the most distinguished speakers of the British stage” (Ra-
dio Times, 16 July 1926)];
Mr. A. [Arthur] Lloyd James [a Welsh phonetician at the School of Ori-
ental and African Studies and former pupil and colleague of Jones at UCL;
Lloyd James had advised the BBC before and became the Advisory Com-
mittee’s honorary secretary]. (R6 / 201 / 1)
The Committee officially met for the first time on 5 July 1926. Two points
stand out from that meeting. First, the Committee put itself up as the
BBC’s absolute authority in matters of what is referred to as “doubtful
words”: it “will decide upon the term to be adopted [...] by announcers and
other officials of the Company” (R6 / 201 / 1).
Secondly, a number of “general principles” of pronunciation were
agreed so that rules for announcers could be drawn up; among them were
the following:
Vowel sounds The Chairman demonstrated that it is possible to give all
in unaccented vowels in unaccented syllables a flavour of their original
syllables. character [...] and that indeed the matter was merely one of
good or bad articulation, e.g. the slovenly speaker uses a
single sound (represented by “eh”) for all vowels in unac-
cented syllables. He says parehdy, parehsite, Julieh, Ehphe-
lieh, where the speaker with good articulation says parody,
parasite, Julia and Ophelia.
It was agreed that the announcers should attempt the differen-

tiation of such vowel sounds, but without becoming stilted.
The “h” in After some discussion it was decided that speakers of South-
which, why, ern English would find difficulty in pronouncing which,
whale, etc. whale, why, white, etc. with the aspirated “w” [...], although
differentiation between which and witch, whale and wail, etc.
should be recommended in order to avoid homophones.
Untrilled final The possibility of pronouncing the “r” in fire, tower, sure,
“r” and “r” hour, poor, etc. without trilling, was demonstrated. It was felt,
between vow- however, that Southern English speakers [...] would have
els. considerable trouble in pronouncing an untrilled “r”. If that
were so, it was agreed that the untrilled “r” should be treated
as a separate vowel, though not syllabic, e.g. tired to be pro-
nounced tired, not tahd [i.e. no smoothing either4]. It was
agreed that an attempt should be made to give the letter some
sound value, however slight.
Foreign It was agreed [...] that foreign words in common use should
words. be Englished and that where their sounds approximated to
English sounds the original sounds be respected, e.g. chauf-
fer, and in proper names, Shoobert, but Mose-art, Reams
(Rheims).
(Minutes, 5 July 1926; R6 / 201 / 1; emphasis in the original

document)
The Committee further decided not to use the phonetic alphabet for the
transcription of their decisions, in order to make communication to the
Press – and therefore the general public – as easy and straightforward as
possible (8 November 1926; R6 / 201 / 1). All the Committee’s rulings
were, as a matter of routine, published in the Radio Times, whose letter
pages at the time developed into a veritable forum for language prescriptiv-
ists and purists.
From a linguistic point of view, it is quite remarkable that most of the
pronunciation features drawn up in the Committee’s very first list of rec-
ommendations are not only “old friends” going back to the 19th century,5
but they still caused bouts of fierce controversy in the early 20th century –
and beyond – and I therefore would like briefly to look at them in turn.
There are dozens of letters addressed to the editor of the Radio Times
complaining about the pronunciation of vowels in unstressed syllables (that
is the shwa or even its elision), about the pronunciation of aspirated “wh”,
of “r” in all its forms from postvocalic to linking and intrusive, and about
the smoothing of triphthongs. Here is an example of one such letter:
Announcers’ English.
Dear Sir, – The announcers seem to find a lot of difficulty with the letters
“r” and “h”. Why should a word ending in “er” be pronounced as if it ended
in “aw” or “ah”? Why put in an “r” when it isn’t there? Such vulgarisms as
“Indiar” and “Australiar rand Africa” are painful. Why say “modden” when
the word is “modern”? [...] We also heard that the King had been “weeled”
in a bath-chair, when – I suppose – he announcer meant “wheeled”. Also, is
it the British Empire or the British Empiah? If the B.B.C. pronunciations in
the above instances are the correct ones, then I apologize for my ignorance.
If they are not correct, surely listeners are justified in saying so. – F. W. E.
Wagner, Castletown Road, West Kensington. (Radio Times, 28 June 1929)
The “wh” aspiration question, incidentally, triggered months of contro-
versy in the Radio Times, with quotations ranging from the New English
Dictionary (that is, the Oxford English Dictionary) to Anglo-Saxon. With
respect to this feature Gimson (1984: 47) observed that “the phoneme /V/
as in white, although characterized as obsolescent by phoneticians a hun-
dred years ago, was nevertheless often recommended as appropriate in
more formal styles.”
It is also interesting to note that, as late as 1981, Burchfield, in his BBC
Guide, recommended to “avoid the intrusive r”, at least “in the formal pres-
entation of the news or of other scripted speech” (1981: 10), even though
this is a well-known feature (be it categorical or variable) of non-rhotic
accents (Wells 1982: 222–227). Hughes, Trudgill, and Watt (2005: 46)
equally consider intrusive /r/ as “very much a part of RP”, but they add that
“careful speakers and speakers of adoptive RP may [...] avoid it”, whereas
Trudgill (2002: 174) lists intrusive /r/ among a number of “features which
used not to be RP and now are RP”.
Concerning the treatment of foreign words, Loyd James told his audi-
ence in a radio talk, broadcast on 29 November 1926:
Foreign words are a source of anxiety, and here the committee has one pol-
icy, which is, for better or worse, to anglicise as many as possible. Whether
a foreign word has lived long enough among us to be given papers of natu-
ralisation is for the committee to decide. (R6 / 196 / 1)
It seems, to give just one example, that the word garage was given its UK
passport eventually. While on 5 July 1926 “It was [...] decided not to angli-
cise the word ‘garage’ yet”, the minutes of the Committee’s tenth meeting,
held on 20 November 1930, record: “GARAGE: The committee substituted

GÁRREDGE for their previous recommendation.” (R6 / 201 /1)
Finally, as for giving the postvocalic /r/ “some sound value” and the un-
stressed vowels “a flavour of their original character”, it was not until 1934
– more than eight years after the first meeting – that the Committee finally
realised that
Southern announcers cannot treat the r sound in the Northern manner, and
very few English born speakers give to the unaccented vowels the flavour
that Mr. Bridges recommended. But the B.B.C. very definitely concerns it-
self with checking ultra-modern tendencies in the language [...]. (Minutes,
Enclosure A; 20 September 1934; R6 / 201 / 2)
This is, in fact, another important point about the working mechanisms of
the Committee. In spite of the fact that “in many cases the recommenda-
tions of the Committee involve a dogmatic decision on points which might
legitimately be argued, but to prevent confusion uniformity has to be at-
tempted” (Radio Times, 16 July 1926), the Committee actually increasingly
sought the public’s views and, not infrequently, took some of the criticism
on board, and from the later 1920s onwards showed an increasing willing-
ness to change some of its earlier decisions.
3. A “listening” BBC?
In 1930 the Committee even went so far as to regard all its decisions on
pronunciation as “provisional and be published as such in ‘The Radio
Times’, being given as much publicity as possible”. Then “these provi-
sional decisions should be reviewed at the following meeting, when they
would be either confirmed as final, or altered in the light of criticism re-
ceived” (17 January 1930; R6 / 201 /1).
I should like to illustrate this new, pragmatic “listening BBC” with two
brief examples. The first concerns place names. The Committee decided
early on to produce a pronouncing dictionary of English place names that
would be “of service to general learning, and of practical use to the
B.B.C.” (8 November 1926; R6 / 201 /1). In the spring of 1927, “a card
index of some nine hundred place names, with their pronunciations, had
been compiled from information received from listeners in the country” (26
May 1927; R6 / 201 /1). But the place name consultation process undoubt-
edly reached its climax in 1929 with a full-page appeal in the Radio Times
by Lloyd James. He wrote that “several hundreds” of letters had already
been received about place names with unfamiliar pronunciations, “but

more are required” (Radio Times, 3 May 1929).
The reactions were overwhelming, and three weeks later, the Radio
Times had to publish the following statement: “Mr. Lloyd James has re-
ceived so many letters in reply to his request for place-names of strange
pronunciation that he is unable to reply to them all individually”. It contin-
ued on a lighter note with respect to the BBC’s station and transmitter in
Daventry (Northamptonshire):
A listener has taken exception to our spelling the local pronunciation of
“Daventry” as “Daintry”. This, she says, should be “Dane-tree”. Daventry
was the scene of the last stand of the Danes; and the town’s crest is a “Dane
under a tree”. We thank her for this information, which was new to us. The
next time we see a Dane under a tree anywhere, we shall think of Daventry.
(Both Sides of the Microphone; Radio Times, 24 May 1929)
Although the main work on place names was of course done by the Com-
mittee (particularly a three-member sub-committee specifically set up for
the purpose; it consisted of Daniel Jones, Lloyd James and J. C. Stobart,
the Director of the Education Department (R6 / 201 / 1)), the input from
listeners should not be underestimated. The result was impressive. A pam-
phlet, entitled Recommendations to Announcers Regarding the Pronuncia-
tion of Some English Place Names, published in 1930, contained “some
1500 English names, a few Manx names, and one or two Welsh names
from border counties” (Broadcast English II, 1930: 5).
The English place name booklet also introduced two important novel-
ties. First, although it continued to be the case that “one pronunciation only
should be recommended for Announcers”, there now are, in a separate
column, the “accepted local pronunciations which are still in general use”.
Secondly, it was decided – for the first time – “that pronunciations should
be given in phonetic script, but that a popular notation should also be
given” (Minutes, 25 July 1929; R6 / 201 /1).
As for Daventry, therefore, both pronunciations were listed (in phonetic
script and modified spelling): first the pronunciation recommended to an-
nouncers!czu?msqh .d@Âuudmsqx, followed by the local (and historical)
!cdhmsqh.c@Âxmsqx (Broadcast English II, 1930: 36).
The second example I would like to give relates to the pronunciation of
the word ski. The Norwegian and Swedish pronunciation, where <sk> be-
fore front close vowels is pronounced /R/, was common in English at the
time (see Jones 1924, 1937: s.v., who in both editions gives /Ri:/ first, fol-
lowed by the less common alternative /ski:/ in brackets). However, there
was a constant stream of letters objecting to the pronunciation /Ri:/. Here is

one from 1927:
“Ski” or “Shee”?
Dear Sir, – Would it not be as well if the B.B.C. Advisory Committee on
Spoken English gave attention to that vexed word “ski”? When broadcasting
some time ago from the Plymouth Station, I was informed by the Announcer
that, so far as he was aware, the accepted pronunciation was “shee”. If the
Norwegian pronunciation be adopted, then presumably the Norwegian plu-
ral, without the final “s” should also be used: which I venture to think would
ultimately lead to confusion. In any case, since the word is almost univer-
sally called “ski” on the Continent, where the sport actually takes place, is it
not somewhat absurd for us in England to adopt the sound used by the mi-
nority in Scandinavia? [...] F. McDermott, Tregoose, St Columb, Cornwall.
(Radio Times, 16 September 1927)
The ski question – the word was borrowed into English from Norwegian –
is directly linked to the treatment of loan words in English and, therefore,
the question, as Lloyd James put it, whether “a foreign word has lived long
enough among us to be given papers of naturalisation”, i.e. to be “Eng-
lished”; that, he said, was “for the Committee to decide” (see above, §2).
Of course there are no objective criteria of “long” or even “common use”,
the condition given in the 1926 “general principles”. According to the Ox-
ford English Dictionary (s.v.) there is an isolated early use of ski in English
dating from 1755, but all the other early attestations are late 19th and early
20th century. But, after another three years of uncertainty, the Committee
did decide to listen to its critics: on 17 January 1930 the pronunciation
“skee” was adopted (R6 / 199 / 2), and on 7 February 1930 the Advisory
Committee’s new findings were published in the Radio Times. This, how-
ever, was not the end of it. The complaints continued – now arguing for
reinstatement of the pronunciation /Ri:/ – and so did the discussion among
Committee members. On 20 September 1934, the Committee “listened”
again and therefore accepted “SHEE to replace SKI”, that is, to “de-
naturalize” the loanword, so to speak, and thus to reverse the previous re-
verse decision (R6 / 201 / 2).
In spite of such apparent “openness” to suggestions and criticism, the
BBC still was, in many respects, frightfully upper-class, public school and
Oxbridge. In the letter pages of the Radio Times (12 February 1932) a lis-
tener (the playwright and biographer St John Ervine) asked the meaning of
the word nesophile, which he had heard in a broadcast but could not find in
any of the dictionaries. To which the broadcasters who had used the word
replied, obviously tongue-in-cheek but nonetheless rather high-handedly:
“We apologize with horrid humility for having gone one better than the
best dictionary. Our only excuse is a classical education. We derive the
word, which means simply an ‘island lover’ from the Greek noun nēsos ‘an
island’, and the verb phileō ‘I love’. It is one which we have always used
under the impression that it was supported by tradition.” (In the actual re-
ply, the Greek words are not transliterated, but in their original Greek
script.)
4. Achievements, changes and suspension
Excesses like the above – patronising though they were – should not, how-
ever, tarnish the larger picture. In its first dozen or so meetings, the BBC
Advisory Committee on Spoken English achieved the following:
– it drew up general pronunciation guidelines for announcers and news-

readers;
– it drew up whole lists of words whose pronunciation was uncertain and
made recommendations or suggestions for these;
– it implemented the compilation of a card index for the pronunciation of
place names;
– it decided to publish its findings and, if necessary, revise them.
The last two points – as we have already seen in the case of English place
names – eventually led to the publication of a series of enormously popular
pamphlets, entitled Broadcast English:
– I: Recommendations to Announcers Regarding Certain Words of

Doubtful Pronunciation (1st edn 1928; 2nd edn 1932; 3rd edn
1935; a fourth edition was planned in the course of 1938 but never
published (Minutes, 28 June 1938; R6 / 201 / 2))
– II: Recommendations to Announcers Regarding the Pronunciation of
Some English Place Names (1930)
– III: [...] Scottish Place Names (1932)
– IV: [...] Welsh Place Names (1934)
– V: [...] Northern Irish Place Names (1935)
– VI: [...] Foreign Place Names (1937)
– VII: [...] British Family Names and Titles (1939)
The minutes of 29 November 1928, that is immediately after the publica-

tion of Broadcast English I, record that “the attitude of the Press was, on
the whole, favourable, and that the pamphlet had aroused considerable
interest throughout the country” (R6 / 201 / 1). The pronunciations con-
tained in the 2nd edition of Broadcast English I, by the way, were recorded
and published by the Linguaphone Institute.6
On the surface, the Advisory Committee on Spoken English continued
its work – in a fashion broadly similar to that outlined here – until its sus-
pension at the beginning of the Second World War in 1939. Below the
surface, however, there were various difficulties, conflicts and crises; as a
result of these, two important and wide-ranging sets of changes in policy
and structure were implemented over the years.
First, a kind of widening-up process: the accession of new members to
the Committee came in two “waves”, the first of which was the result of
the Committee’s first serious crisis in 1929, triggered mainly by the prob-
lem of irregular attendance of some members and decisions taken by ma-
jority voting in meetings where half of the members may have been ab-
sent.7 The second, bigger “wave” of enlargement took place in 1934 and
led to the creation of a permanent specialist sub-committee whose paid
members – Professors Daniel Jones, Lloyd James and Henry Cecil Ken-
nedy Wyld, since 1920 Merton professor of English language at Oxford
University, and Mr Harold Orton, a dialectologist from King’s College,
Newcastle, and former pupil of Wyld (ODNB 2004–6) – would make rec-
ommendations before the word-lists were put before the full Committee.
By October 1938 the main Committee consisted of 18 non-specialist mem-
bers, including representatives of various learned societies such as The
British Academy or The Royal Society of Literature, plus the four mem-
bers of the specialist sub-committee (for details, see Appendix).
The second set of changes is of a more linguistic nature, namely the
general introduction of the International Phonetic Alphabet (or IPA) in
1934, first used tentatively, as we saw, in the English place names booklet.
As Lloyd James pointed out:
The early recommendation to record pronunciations only in the modified
spelling system proved unsatisfactory, and a strictly phonetic pronunciation
(using the International Phonetic Alphabet) was adopted in addition. (Min-
utes of the first meeting of the reconstituted Committee, 20 September 1934;
R6 / 201 / 2)
This change is clearly connected with the creation of the specialist sub-
committee; the presence of more linguists, as well as the Committee’s in-
creased contacts with academics from overseas,8 not only helped to push
through sensible measures like the use of the IPA, but also “ensured that
the strict prescriptivism expressed by Reith [...] was to some extent miti-
gated” (McArthur 1992: 110).
Finally, in 1937, we see the suspension of regular publications of the
Committee’s decisions, decisions often reached by majority voting as, fre-
quently, several pronunciations were found to be “equally good”. A state-
ment by R. C. Norman, the Chairman of Governors, to the main Committee
reads:
The Corporation [...] now proposes that, since the public persistently misun-
derstands its motive in publishing a list of pronunciations recommended for
the use of Announcers, it should no longer necessarily publish them in the
“Radio Times” and in the daily Press as a matter of routine. (Attached to the
minutes of 29 January 1937; R6 / 201 / 2)
Instead, the BBC would only “give private instruction to announcers”
based on the specialist members’ reports and comments by the main Com-
mittee.
At the outbreak of the Second World War the Advisory Committee on
Spoken English was formally suspended, although Lloyd James had “made
a strong request that [it] should continue its work” (18 December 1939; R6
/ 196 / 11). Tellingly, it was not reactivated in 1945.9 Instead, the BBC-
internal Pronunciation Unit was eventually to emerge in the 1940s, with
Daniel Jones as Chief Pronunciation Advisor, a role he kept until his death
in 1967. The Unit’s much reduced responsibility was and still is consistent
and “accurate” pronunciation – rather than the promotion of a particular
accent or even a single pronunciation of common words with “rival forms”
– of all types of proper names or phrases, from any language, that BBC
staff need to say before the microphone (Catherine Sangster, BBC Pronun-
ciation Research Unit, 18 November 2004, private communication).
5. Conclusion
So, what can we conclude from all this? First, we noticed a strict and dog-
matic prescriptivism; the original intention was not only to achieve consis-
tency among BBC announcers and newsreaders but also to educate the
public through notions of what was – or was not – “good and correct Eng-
lish”; in short, to find and define the “best” pronunciation, to fix and dif-
fuse it, and thus create a uniform standard.
Then, more and more we saw a kind of “listening BBC”, which re-
garded its decisions on pronunciation as largely provisional until a proper
feedback from an ever-larger circle of committee members, advisers and
the public was received. Additionally, there was a slow but steady trend
towards what could be called “linguistic professionalization”: the weight
clearly shifted towards the specialist sub-committee; alternative pronuncia-
tions were admitted and eventually found to be “equally good”; the IPA
was used routinely and the dogmatic press releases of the Committee’s
rulings were stopped.
Was it all a failure then? Yes and no. Yes, in the sense that the goals the
Advisory Committee set out in its early mission statements – to fix and
thus standardise spoken English by finding the “best”, “most correct” pro-
nunciation of “doubtful words” and then, by the BBC’s “powerful exam-
ple, endeavour to stem modern tendencies to inaccurate and slurred
speech” – simply could not be attained; phonology, as Lesley Milroy
(1999: 173) has observed, is “particularly resistant to standardisation”. To
give just one example, unstressed vowels: the Advisory Committee could
neither stop “the loss of a post-tonic secondary stress in words such as
territory, adversary, ceremony, with a consequent weakening of the vowel
to /?/ and its frequent elision” (Gimson 1984: 47)10 nor prevent the change
from /H/ to /?/ in weak syllables, particularly after /l/ and /r/, as in angrily
or merrily (Gimson 1994: 83, 99–100).11 Language variation and change
are undeniable facts, as is the focused (rather than fixed) nature of even a
reference accent such as RP (Smith 1996: 65–66) – something the Advi-
sory Committee only eventually came to terms with.
It was not a failure in two other senses, however. First, the Advisory
Committee played an important role in the emergence of a kind of “broad-
cast English” or “broadcast style” which – though allowing for some vari-
ability – nonetheless conveyed and still conveys a sense of “objectivity”
and “authority” going far beyond the UK, particularly when it comes to
news broadcasts. The Advisory Committee’s influence can be felt up to
this day (most clearly when it comes to the generally respectful treatment
of domestic as well as foreign place and personal names). In this respect,
the early and pioneering work of the BBC in defining a style appropriate
for broadcasting may be seen as somewhat parallel to the influence of
printing on the written language, though the analogy is of course only a
superficial one and therefore should not be pushed too far.
And no, it was also not a failure in that it raised awareness of language
issues among the population. True, every variation in pronunciation trig-
gered and still triggers a flood of letters by the “language mavens” (as Ste-
ven Pinker has famously called them), but at the same time, the various
discrepancies in pronunciation – be it between two newsreaders, be it be-
tween a newsreader’s pronunciation and our own – has made us think
about notions such as “standard”, RP, and “correctness” more than ever
before. And this, I believe, is the other lasting legacy of the BBC Advisory
Committee on Spoken English.12
Appendix
The Reconstituted Main Committee (31 October 1938; R6 / 196 / 10):
(Biographical information, taken from the ODNB 2004–6 unless stated otherwise,
is provided for only those members not already introduced in §§2 and 4).
Professor George Gordon (Chairman) [a literary scholar, President of

Magdalen College, Oxford and Vice-Chancellor of Oxford University];
Lady Cynthia Asquith [writer of memoirs and diaries; admired at the
time primarily for her “unique blend of intelligence and playfulness”, social
background and beauty; friendships with D. H. Lawrence and the playwright
J. M. Barrie (ODNB 2004–6)];
Lord David Cecil [a literary biographer and critic, former fellow of
Wadham College, Oxford];
Sir Kenneth Clark [art historian, director of the National Gallery and
surveyor of the king’s pictures];
Alistair Cooke [BBC Film Critic from 1934–1937, he was suggested by
Lloyd James as he is “a very brilliant man [...] and very familiar with mod-
ern educated American usage” (Lloyd James to Reith, 31 March 1935; R6 /
196 / 7); Cooke’s weekly “Letter from America” could be heard on the BBC
until just before his death in March 2004];
Professor Julian S[orell] Huxley [zoologist, formerly professor of zo-
ology at King’s College, London; Secretary of the Zoological Society];
F[rank] L[aurence] Lucas [author and classical scholar, fellow of King’s
College, Cambridge];
P[ercy] H[ugh] B[everly] Lyon [Headmaster of Rugby School (Who
Was Who 1991)];
Miss Rose Macaulay [novelist and essayist, Macaulay was also the au-
thor of Catchwords and Claptrap (1926), “which reflected the pleasure she
derived from the English language and her insistence on verbal precision”
(ODNB 2004–6)];
Sir Edward Marsh (Representing Royal Society of Literature) [re-
tired civil servant, patron of the arts, chairman of the Contemporary Art So-
ciety and council member of the Royal Society of Literature];
Emeritus Professor Sir H[erbert] J[ohn] C[lifford] Grierson [literary

critic and scholar, formerly Professor of English at the universities of Aber-
deen and Edinburgh];
S[amuel] Ratcliffe [a journalist and lecturer, suggested by Shaw as he is
“very sensitive to shades of pronunciation, and he does a lot of lecturing in
America and comes up against all the differences between spoken English
and Spoken American” (19 June 1934; R6 / 196 / 5)];
Dr. I[vor] A[rmstrong] Richards [lecturer in English at Cambridge
University, later professor at Harvard, well-known to linguists for his col-
laboration with C. K. Ogden on The Meaning of Meaning (1923) and, in the
1930s, on “Basic English”];
George Bernard Shaw [see §2];
Logan Pearsall Smith (Representing Society for Pure English) [see
§2];
Dr. W[alter] W[ilson] Greg (Representing British Academy) [inde-
pendent literary scholar and biographer, fellow of the British Academy];
The Rev. Canon H[arold] Costley-White (Representing English As-
sociation) [when nominated by the English Association, Costley-White was
Headmaster of Westminster School; he later became Dean of Gloucester
(Who Was Who 1979)];
Sir Kenneth R[alph] Barnes (Representing Royal Academy of Dra-
matic Art) [head of the Royal Academy of Dramatic Art, knighted in 1938
“at the insistence of George Bernard Shaw, a firm friend and generous bene-
factor” (ODNB 2004–6)].
And the four members of the permanent specialist sub-committee:

Professor Daniel Jones [see §2];
Mr. A. Lloyd James [see §2];
Professor H. C. K. Wyld [see §4];
Mr. Harold Orton [see §4].
Notes
1. Gimson (1994: 77–78) has justly remarked that “there have always been at any
one time disparities between the speech sounds of the younger and older gen-
erations” and that therefore “the speech of the young is traditionally character-
ized by the old as slovenly and debased.”
2. Daniel Jones later also abandoned the term Public School Pronunciation (PSP)
and followed H. C. K. Wyld in using Received Pronunciation, a label actually
first utilised by A. J. Ellis (Gimson 1984: 45; Strässler 2005). Interestingly, the
term “BBC English” originally had a rather negative connotation and was used
“among regional BBC staff resentful of the better prospects of speakers with
public-school accents” (McArthur 1992: 109). The fact that in the 1920s the
two terms “BBC English” and “public school accent” were perceived as syn-
onymous speaks volumes about the public image and composition of the early
BBC.
3. Biographical information, unless stated otherwise, is taken from BBC-internal
documents and the Oxford Dictionary of National Biography (ODNB 2004–6).
4. The monophthongisation process of, in this case, the triphthong in /!s`H?c/ to
/!s@9c/ is commonly referred to as smoothing (see Wells 1982: 238–242, 292–
293).
5. See Mugglestone (2003) on homophones (114–115), aspirated w /hw/ (186–
188), vocalisation of /r/ (86–89), and linking and intrusive /r/ (91–94).
6. The final paragraph of the preface in the first edition was omitted and substi-
tuted with the following sentence in the second: “All the pronunciations con-
tained in this booklet have been recorded on two ten-inch gramophone records
which are published by the Linguaphone Institute” (Minutes, 26 March 1931;
R6 / 201 / 1).
7. The problems are outlined and enlargement is recommended in a seven-page
document (most probably drawn up by Lloyd James) from the autumn of 1929.
C. T. Onions, co-editor of the Oxford English Dictionary and reader in English
philology at Oxford University, and Lascelles Abercrombie, a leading poet and
at that time professor of English literature at Leeds, were invited to join the
Committee (R6 / 196 / 3; ODNB 2004–6). The former eventually resigned un-
der rather acrimonious circumstances (R6 / 196 / 4), the latter died in 1938
(Minutes, 8 December 1938; R6 / 201 / 2).
8. The advisory committee minutes of 30 November 1933, for example, report an
exchange of letters between Lloyd James and Professor George Philip Krapp of
Columbia University in New York. Krapp – the author not only of the two-
volume The English Language in America (1925) but also of popular hand-
books like A Comprehensive Guide to Good English (1927) (Garraty and
Carnes 1999: 901–902) – approved of Lloyd James’s suggestion “that an
American Advisory Committee on Pronunciation might be formed to act in con-
junction with the B.B.C. Committee, so that one Committee might co-operate
with the other in ascertaining general usage in debatable cases.” In the same
meeting it was further suggested that “in view of the introduction of Empire
Broadcasting steps should be taken to obtain the co-operation of authorities on
pronunciation in Australia, New Zealand, South Africa, and Canada” (R6 / 201
/ 1).
9. Note that Reith, who had created the ethos that led to the Committee’s forma-
tion, left the Corporation in 1938 and Lloyd James, not only the Committee’s
honorary secretary but also its driving force, had suffered from “depressive in-
sanity” due to the “stress and anxieties of war” and committed suicide in 1943
(ODNB 2004–6).
10. Though in 1981, Burchfield still advised announcers to “be careful not to gar-
ble” words like library or secretary (at 11).
11. Wells (1982: 296) lists merrily among various types of words “where RP
speakers differ from one another, some using /H/ and some using /?/.” Twenty
years on, Hughes, Trudgill, and Watt (2005: 48) write that “in general, younger
people are more likely to have /?/, upper-class speakers are more likely to have
/H/.”
12. Many thanks to the BBC Written Archives Centre, Caversham, for kindly grant-
ing permission to quote extensively from BBC-internal documents; to its staff,
particularly the ever-helpful and deeply knowledgeable Jeff Walden and Jac-
quie Kavanagh; to Jens Poulsen for competently and efficiently checking vari-
ous sources in Zurich to which I did not have access in Lausanne; to Peter
Trudgill, Catherine Sangster and Anne-Laure Gex for their very helpful com-
ments on an earlier (and rather different) version of this paper; and above all to
Peter Jackson, who not only very kindly and generously passed me his extensive
research and archive work on the same topic, but whose careful proofreading of
a near-final version also saved me from numerous errors and inaccuracies.
References
1. Unpublished sources, BBC Written Archives Centre, Caversham

R6 / 196 / 1–11: Advisory Committees, Spoken English Advisory Committee, Files
1–11, 1926–1943.
R6 / 199 / 1–2: Advisory Committees, Spoken English Advisory Committee, Index
of Words Discussed, vol. 1: A-N; vol. 2: O-Z, A-Z Index of Proper
Names.
R6 / 201 / 1–2: Advisory Committees, Spoken English Advisory Committee, Min-
ute Books: 1926–1938.
2. Published sources
Aitchison, Jean
2001 Language Change: Progress or Decay? 3rd edition. Cambridge:
Briggs, Asa
1995 The History of Broadcasting in the United Kingdom. Volume I: The
Birth of Broadcasting 1896–1927. First published 1961, reissued
1995. Oxford: Oxford University Press.
Broadcast English I–VII

1928–1939 Recommendations to Announcers. Collected and Transcribed for
the B.B.C. Advisory Committee on Spoken English by A. Lloyd
James. London: The British Broadcasting Corporation.
Burchfield, Robert
1981 The Spoken Word: A BBC Guide. London: British Broadcasting
Corporation.
Garraty, John A. and Mark C. Carnes (eds)
1999 American National Biography. Volume 12. New York: Oxford Uni-
versity Press.
Gimson, A. C.
1984 The RP Accent. In Language in the British Isles, Peter Trudgill (ed.),
45–54. Cambridge: Cambridge University Press.
1994 Gimson’s Pronunciation of English. 5th edition, revised by Alan
Cruttenden. London: Arnold.
Hughes, Arthur, Peter Trudgill and Dominic Watt
2005 English Accents and Dialects: An Introduction to Social and Re-
gional Varieties of English in the British Isles. 4th edition. London:
Hodder Arnold.
Jones, Daniel
1924 & 1937 English Pronouncing Dictionary. 2nd & 4th editions. London:
Dent.
Labov, William and Wendell A. Harris
1986 De Facto Segregation of Black and White Vernaculars. In Diversity
and Diachrony, David Sankoff (ed.), 1–24. Philadelphia: John Ben-
jamins.
McArthur, Tom (ed.)
1992 The Oxford Companion to the English Language. Oxford: Oxford
University Press.
Milroy, Lesley
1999 Standard English and Language Ideology in Britain and the United
States. In Standard English: The Widening Debate, Tony Bex and
Richard J. Watts (eds), 173–206. London: Routledge.
Mugglestone, Lynda
2003 “Talking Proper”: The Rise of Accent as Social Symbol. 2nd edition.
Oxford: Oxford University Press.
ODNB
2004–6 The Oxford Dictionary of National Biography online:
http://www.oxforddnb.com. Last accessed 21 August 2006.
OED
Oxford English Dictionary online: http://www.oed.com. Last ac-
cessed 21 August 2006.
Radio Times
16 July 1926; 16 September 1927; 3 May 1929; 24 May 1929; 28 June 1929;
12 February 1932.
Reith, J. C. W.
1924 Broadcast Over Britain. London: Hodder and Stoughton.
Smith, Jeremy
1996 An Historical Study of English: Function, Form and Change. Lon-
don: Routledge.
Strässler, Jürg
2005 Jones, Daniel. In Key Thinkers in Linguistics and the Philosophy of
Language, Siobhan Chapman and Christopher Routledge (eds), s.v.
Edinburgh: Edinburgh University Press.
Trudgill, Peter
2002 Sociolinguistic Variation and Change. Edinburgh: Edinburgh Uni-
versity Press.
Wells, J. C.
1982 Accents of English. 3 vols. Cambridge: Cambridge University Press.
Who Was Who
1979 & 1991 (vols. 6 & 8). London: Black.
Chapter 10
Liverpool to Louisiana in one lyrical line:
Style choice in British rock, pop and folk singing
“Can Blue Men Sing the Whites?”

(The Bonzo Dog Doo Dah Band)
Franz Andres Morrissey
1. Introduction
When Pink Floyd’s original guitarist/songwriter Syd Barrett died in July

2006, the BBC News website featured a quote by British singer David
Bowie, “He was the first guy I’d heard to sing pop or rock with a British
accent – his impact on my thinking was enormous”. Indeed, as we shall
see, the two 1967 singles “See Emily Play” and “Arnold Layne”, which
represented the Pink Floyd’s entry into the musical mainstream, show un-
mistakably just how British Barrett’s vocals sound.
Bowie’s admiration for Barrett’s adoption of “a British accent” and the
fact that this should be noteworthy can be explained in terms of a phe-
nomenon that is typical for British popular music.1 At the time relatively
few singers of note used a British singing style consistently, the best
known – and exceptional – exponents being former actor Anthony Newly,
who focussed mainly on chanson- or music-hall-oriented material, and Jake
Thackray, the “founding father of English chanson” (Clayson 2002). By
contrast, ever since the Jazz era, but particularly from the 1950s onwards,
British singers have modelled their singing style on what they perceived to
be an American pronunciation, a tendency that continues to the present
time. This is true for Jazz singers, Skiffle groups, Rock’n’Roll, R’n’B or
Hard Rock bands as well as doowop pop and soul vocalists to the extent
that this style represents a standard style, a norm that seems to permeate
even the material of musicians who as of the mid-Sixties have begun to
adopt singing styles other than American ones. The reason for the choice of
standard style is clearly that the original input for these types of popular
music is based on Afro-American Blues and its musical developments,
mixed with other elements, such as white American pop music (“Schlager-
196 Franz Andres Morrissey
musik”) (Flender and Rauhe 1989: 72), American folk song influences,
Country and Western, etc. Unmistakably the resulting melting pot had and
still has a strong American flavour.2
This choice of singing style often contrasts drastically with the singers’
speech style, e.g. Keith Richards’ and Mick Jagger’s insistent – and given
their background, not always wholly convincing – Cockney speech style
whereas their singing is predominately based on American models, particu-
larly in the early Blues and R’n’B material, with which the Stones first
made a name for themselves in the purist Crawdaddy blues club in Rich-
mond, London (Shapiro 1999: 831). However, despite the continued preva-
lence of American pronunciation in the singing style of British vocalists,
there have been trends towards other, mostly British accents, in rock and
pop vocals, a trend that began, according to Trudgill (1983: 150–151), as
of 1964, became more widespread in the Seventies and Eighties with Punk
and the New Romantics, and reached a new high with the popularity of Brit
Pop and Indie Rock from the Nineties onward.
This paper discusses various factors which influence style choice in
rock, pop and folk singing and attempts to present reasons for some of the
dialectological inconsistencies which Trudgill (1983) and Simpson (1999)
have pointed out. Their approach primarily focuses on sociolinguistic theo-
ries, which will also form part of this analysis. I would posit, however, that
there are explanations for style choice and the inconsistencies mentioned,
for which we may need to look further afield. Admittedly, some of these
explanations may be somewhat marginal to linguistic investigation in a
narrow sense, focussing on issues such as musical genre, song topics and
cultural considerations, as well as the fact that performance has an impor-
tant role to play. Therefore, the final focus of this paper will include pho-
nological considerations, which to date seem to have been somewhat ne-
glected.
The data used are a number of recordings by seminal artists represent-
ing a variety of genres, in particular rock, pop, mainly of the Sixties, folk
rock, singer-songwriter material and, given the starting point of this paper,
psychedelic and prog rock. Clearly such an analysis, even if it were con-
ducted in a format which would afford more space, needs to remain im-
pressionistic and indicates tendencies rather than universal realities.
Liverpool to Louisiana in one lyrical line 197
2. Theoretical framework and scope
For the discussion, the term style will be used in preference to accent. Ac-
cent according to Swann et al. (2004: 2) “identifies a speaker in terms of
regional origin, social standing and, possibly, ethnicity”, which does not
describe the deliberate nature of the choice that appears to be at the basis
of what singers do. For this, the term style seems to be more appropriate,
firstly, because style “refers to a distinctive way of speaking … People
adopt different styles in different contexts” (Swann et al. 2004: 299, em-
phasis added). Furthermore, Bell (1984: 2001) uses the term in his discus-
sion of audience design, which presents the aptest model for the phenome-
non under discussion.
At this point a caveat is called for. A crucial fact when we discuss style
choice in popular music is that it is part of an act, a performance. Flenders
and Rauhe (1989: 72) point out that recorded instances of performances
play a crucial role for the dissemination of popular music; sheet music is
usually available but, unlike in classical music, is not crucial for reproduc-
tion and hardly ever does justice to an actual performance. Moreover, per-
formances, even by the same artist, frequently differ considerably from the
original recording – and in concerts may even be expected to.3 Thus, to a
certain extent the performance poses similar problems for notation as for
linguistic analysis, a point we shall return to below. What is more, it is a
cultural construct, not an instance of natural language, a fact that we can
afford to ignore with as much impunity as a discourse analyst examining an
excerpt from a play. A literary text, like the – intrinsically theatrical – per-
formance of a singer, can yield interesting insights, but conclusions drawn
will apply only partially to the “real world”.
Another characteristic of performance is its unidirectionality and the re-
sulting lack of interpersonal interaction that much of sociolinguistic analy-
sis is based on, which poses problems for such an analysis. Trudgill
(1983), for instance, rightly dismisses accommodation as an explanatory
framework for style shift in the absence of interlocutors a speaker could
accommodate. The fundamental peculiarity of popular music and of mass
communication in general lies in the spatial and usually the temporal dis-
tance between the audience (addressee), and the performer. As a mass, the
audience make performers into stars, but at the same time may well force
them to “replay concert after concert popular songs that [they] will hear but
the musician wants to leave behind” (Bell 1984: 193, emphasis added). By
contrast, performers “hold the initiative to express themselves, and the
audience has no adequate channel for feedback” (Bell 1984: 193, emphasis
added), which is an integral part of (face-to-face) verbal interaction. Fur-

thermore, if we consider Bell’s model of concentric circles (1984: 159) to
describe the various roles of participants in a speech situation, we can see
that the audience at a performance may have characteristics of all the “per-
sons and roles in [a] speech situation”, and yet lack some as a result of the
temporal and spatial distance mentioned (see Diagram 1 and Table 1).
Thus, the audience exhibits characteristics of addressee, auditor, over-
hearer and eavesdropper, the latter in the sense that they may not be known
to the performers, especially if the performance is recorded rather than a
live event.
Table 1: Attributes and audience roles in a speech situation according to Bell
(1984), with popular music audience added
Known Ratified Addressed
Addressee + + +
Auditor + + -
Overhearer + - -
Eavesdropper - - -
Popular music audience -/+ + +/-
popular music au-

speaker dience
addressee
auditor
overhearer
eavesdropper
Diagram 1. Persons and roles in a speech situation according to Bell (1984: 159)
with popular music audience added
When style shift occurs in face-to-face verbal interaction it tends to be a

reaction to the persons and their roles in the interaction as shown in Dia-
gram 1 and therefore represents, in Bell’s terminology, Responsive Style

Design. However, style shift can also signal a change in topic or the
speaker’s attitude, e.g. a shift towards a more formal style if the speaker
wants to make a point with authority and uses, say the H-variety in an L
conversation (e.g. Standard German in a Swiss German conversation; cf.
Ferguson 1959) to do so. Style shifts of this type are not responsive, but as
they are initiated by the speaker, represent instances Initiative Style De-
sign. As Bell (2001) puts it:
Initiative style shift is essentially a redefinition by speakers of their own
identity in relation to their audience. The baseline from which initiative
shifts operate is the style normally designed for a particular kind of ad-
dressee. (Bell 2001: 147)
Furthermore, such style shifts “are not predictable” and thus represent “the
marked case, which draws its force from the unmarked, responsive use of
style” (Bell 1984: 184). They depend for their effectiveness on their rela-
tive infrequency. By contrast, an institutionalisation would also mean a
loss of markedness.
However, as “pop musicians … are acting according to conventions”
(Frith (1988: 4, original emphasis), as quoted in Simpson (1999: 344)),
their choice of singing style is one of them and thus institutionalised,
which suggests that we are faced with a special type of initiative style de-
sign, the so-called Referee Design. According to Bell, “a speaker diverges
from an addressee and towards a third party whom I call the ‘referee’.”
Referees, usually “not physically present at the interaction” are regarded
by speaker (and listener) as so important as to “influence speech events in
their absence” (Bell 1984: 186). The resulting style model could be called
the reference style. Blues, blues-derived rock and pop, and country-based
singer-songwriter material are obviously rooted in American popular cul-
ture and generally associated with an American reference style. Song mate-
rial of this kind, for which the expected American reference style is not
used, will, like marked forms in language, either sound odd or have a spe-
cific effect on the audience. The same holds true for other genres of popu-
lar music, associated with a specific style, such as British English, perhaps
regionally coloured, for music-hall or regional accents for traditional folk
material.
The tendency for British singers to adopt an American singing style
represents an instance of Bell’s Outgroup Referee Design: “speakers lay
claim to a speech and identity which are not their own but which hold pres-
tige for them on some dimension”, but the absence of the outgroup referee
and thus the “absence of feedback” often leads to an “inconsistent adoption

of the forms of outgroup speech” (Bell 1984: 188). One could argue that
this insight provides the perfect explanation for the often less than accurate
imitation of American pronunciation, for instance in the case of “hyper-
rhoticity” in Cliff Richard’s “Bachelor Boy” where “(I’ll be) a bachelor
boy” is rendered as Z?¢ azsR?k?¢ aNH\ (Trudgill 1983: 149).
However, with the move away from American forms as the main model
Ingroup Referee Design has started to play a role as well. Here group mem-
bership is emphasised in interaction with “members of an outgroup … with
a shift towards the speaker’s own (absent) ingroup” (Bell 1984: 187). In-
stances of this in the context of Brit Pop and British Indie Rock will be
discussed below.
This raises the question as to which features speakers and listeners
would consider salient in an American or a British reference style. Simp-
son (1999: 345) presents five such features, the “USA-5”, which represent
the distinctive differences between Standard British English or Received
Pronunciation (SBE/RP) and General American (GenAm, Wells (1982a:
10)). In the following I shall refer to these features as SF for Style Feature
and the number under which they appear in Table 2.
Table 2: Distinctive Style Features (SF) and their realisations in SBE/RP and
GenAm (adapted from Simpson 1999)
Style Standard British English General American (GenAm)
Feature (SBE) / Received Pronuncia-
tion (RP)
SF1 Vowels in bath and hat are The vowel in bath and hat is realised
realised as Z@9\ or [z] respec- as Zz\-
tively.
SF2 Vowels for not and all are The same unrounded lower back
rounded ZP\ and ZN9\- vowel Z@\ is used for not and all.
SF3 The vowel in eye is the diph- The vowel is pronounced as a half-
thong Z`H\- lengthened Z`∂\ (the so-called Con-
federate Vowel).
SF4 Postvocalic /r/ as in car is Postvocalic /r/ is realised as an alveo-
realised as a lengthening of the lar or possibly retroflex approximant,
vowel Z9\- i.e. Z¢\ or [±\-
SF5 Intervocalic /t/ is pronounced Instead of intervocalic /t/ we have a
as Zsç\ or Zs\ (Z>\ in regional Zc\, (Simpson (1999: 345), possibly
variants). dentalised ZcŒ\ (Trudgill 1983) or
something close to the alveolar
flap/tap Z3\.4
In addition, we will also touch upon features such as “yod-dropping”, the

replacement of SBE/RP vowel ZU\ in love with a more central, more
fronted American vowel that could be represented as Z2∂\ and the comple-
mentary distribution of Zk\ and [4] in SBE/RP versus dark /l/ in GenAm.
These are less salient features and in the case of the elision of /j/ “not en-
tirely uniform” in GenAm (Wells 1982: 247). Nevertheless, “yod-
dropping” will have a bearing on the considerations in Section 4.
3. Instances of style choice
3.1. Defining oneself through a reference style
On the Lindisfarne live album “Caught in the Act”, recorded in Newcastle

City Hall in 1983, we can observe a telling phenomenon: in “Lover not a
Fighter” the lyrics are sung in an unmistakable and relatively consistent
American singing style.
(1) I’m a lover not a fighter
Z`∂l ? 42∂u?¢ m@3 ? e`∂3?¢\
and I’m Z`∂l\ really Z¢h94H\ built for ZeN9\ speed
There is widespread use of flapped intervocalic /t/ (SF5) and relatively
consistent postvocalic /r/ (SF4), with the exception of the missing /r/ in
“for speed”. In addition we have two elements which go beyond the USA-
5, the pronunciation [2∂\ for /U/ in lover and the dark /l/ where we would
expect a light /l/, especially in “Geordie”, which, unlike SBE/RP has the
light /l/ feature in all positions (Wells 1982: 374). However, in the music-
hall flavoured “I Must Stop Going to Parties” we are confronted with a
high incidence of unmistakably (North Eastern) English features.
(2) I must Z@ lTrs\ stop going Zf79Hm\ to parties Z!o`93hy\, …
My hair is changing colour Z!jTk@\ and I’m getting kind ZjDHmc\ of thin
…
Next time ZsDhl\ you see me there, will you kindly Z!jDHmckH\ throw me
out ZDT>\-
The sample, apart from the typically northern absence of the FOOT-
STRUT split in ZlTrs\ and Z!jTk@\, highlights a whole range of Geordie
features (cf. Wells 1982: 374) such as the rather open word-final Z@\ in
colour, the monophthongal realisation Z79\ in going, the ZDH\ in kind, time
and kindly (but Z@\ for I) and the ZDT\ in out together with the glottal stop.
However there is a departure from the almost textbook Geordie style in the
pronunciation of parties, where we would expect at least a glottalised inter-
vocalic /t/, if not a glottal stop. However the weak vowel /H/+ both raised
and relatively long, again corresponds to the Tyneside style.
What we observe here is not so much an inconsistency in singing style
between the two examples (although some of the features are inconsistent
with the style chosen) but a deliberate change between the two songs, in
keeping with the different genres, namely, the Swamp Blues “Lover not a
Fighter” by Louisiana producer J.D. Miller and the “folky” self-penned
Lindisfarne song. Whereas it makes sense to use an American model for
the former, it seems just as appropriate to use a Geordie style for the latter.
The departure from the “unmarked” American reference style condi-
tioned by genre, i.e. Rock’n’Roll, can also be observed when the Bonzo
Dog Doo Dah Band’s Viv Stanshall covers the Elvis hit “Suspicion”
(1974) and in Nottingham-based Roaring Jelly’s Buddy Holly parody
“Trev and the Rock’n’Roll Rockets” (1977) complete with Holly’s legen-
dary yelps. In the former, Stanshall sings the lyrics in his trademark RP
accent, in the latter the pronunciation never wavers from Midland / North-
ern English style. In both cases much of the resulting comic effect stems
from the use of this “marked” singing style. To put it another way, the un-
marked style choice reflects the style of the culturally dominant group from
where the genre originates, i.e. an American singing style for Blues, blues-
derived rock and pop. By contrast, the marked use of a style, regional in
the case of Lindisfarne (Geordie) and Roaring Jelly (Midlands), SBE/RP in
the case of the Bonzo Dog Doo Dah Band, contributes to what is known in
the recording industry as their novelty status.
The cultural dominance manifested in a reference style can be illus-
trated particularly well with the Rolling Stones version (3a) of “Little Red
Rooster” (1964), where Jagger demonstrates how closely his singing style
is modelled on that of the original singer, black Blues shouter Howlin’
Wolf (1961) (3b).
(3a) Ain’t had no peace in the farmyard Z!e@9li@9c\
Since my Zl`∂\ little [4H34<\ red rooster’s Z!qt9r3?y\ been gone
(3b) There’s ZCD?y\ been no peace in the barnyard Z!a@9mi@9c\

Since my Zl`∂\ little [4H34<\ rooster Z!qt9r3?\been gone
Both use the “Confederate vowel” (SF3), both are clearly non-rhotic (SF4),
and in both versions we can observe a flap (SF5) between two vowels or
between a vowel and a syllabic /l/.
Another example that illustrates the strength of cultural dominance,

even when the image of the band or the setting of their material are in con-
flict with this are “The Who”. They started their career on the London pub
circuit with a sound that has been described as “the hard, vulgar and relent-
less Rock’n’Roll style of British working class youth” (Graves and
Schmid-Joos 1990: 854). Although Townsend often wore a jacket made of
a Union Jack and many early publicity photographs show the band posing
against the British flag, the singing style of their enduring first hit “My
Generation” (1965) obviously owed more to their genre of choice, R’n’B,
than to their projected British image:
(4) People try Zsq`∂\ to put us ZoT3?y\ down
talkin’ Z!s@9jHm\ about my Zl`∂\generation
…
Hope I die Z`∂ c`∂\ before I Z`∂\ get old
Clearly in evidence is the American model for SF2 in Z!s@9jHm\+ SF3 for “I
die” etc., and SF5 for “put us”. This pattern remains the same even in “Pin-
ball Wizard” (1969) with its unmistakable English setting; in the opening
lines “Ever since I was a young boy, I played the silver ball/ from Soho
down to Brighton I must have played them all” /`H/ is consistently realised
as Z`∂\, and “all” and “ball” as Z@94\ and Za@94\ respectively.
A similar pattern is evident in the singing of Kent songwriter and folk
guitarist Ralph McTell. The 1974 hit “Streets of London” shows “McTell’s
admiration for practitioners of the country blues” (Jenkins 2003) both in
the guitar picking style and in the chord progressions. However, the musi-
cal orientation towards American country blues and the focus on lonely
denizens of the Capital create an interesting conflict. One might well ex-
pect the topic to have a stronger influence on the singing style than the
musical orientation, but the result is less clear-cut. While there is a consis-
tent absence of postvocalic /r/ (SF4), in my mind based on SBE/RP rather
than a non-rhotic Southern American style, there is a case of yod-dropping
in the first verse in the line “Yesterday’s papers telling yesterday’s news
Zmt9yfifi\”, which would be in line with just such an American model. How-
ever, as (5) illustrates, there are several other instances where GenAm and
SBE/RP seem to sit side by side:
(5) In our winter city Z!rH3H\ the rain cries Zj¢`Hy\ a little pity Z!kH34< !oçH3H\
for one more forgotten Ze?!fP3m<\ hero …
We have, thus, for SF3 in Zj¢`Hy\for SF2 in “forgotten” Ze?!fP3m<\ the

SBE/RP pronunciation; the same applies for the differentiation between
light and dark /l/ in Z!kH34<\.
However, for SF5 the intervocalic /t/ is consistently flapped as would be
typical for GenAm. In the chorus we have a further American feature, al-
beit not a phonological one, when McTell sings “and say that for you the
sun don’t shine” instead of doesn’t.5 However, this may be as much an
issue of rhythm as a bow towards an American reference style.
In later McTell recordings similar phenomena persist. Both “The En-
emy Within” and “Care in the Community” on the 1995 album “Sand in
Your Shoes”, take up two catchphrases of the Thatcher era. “Care in the
Community”, described in the liner notes as a follow-up to “Streets of
London”, also presents maladjusted inhabitants of London’s streets. Unlike
its predecessor, however, it contains references to two actual places, the
Strand and Waterloo station, as well as to the Tube. The song is set to a
laid-back, jazzy tune, which stands in stark contrast to the bitter sarcasm of
the lyrics. It seems that, here too, it is the music that determines the choice
of the singing style more than the topic (the effect of Thatcherite politics)
or the setting (London) as (6) demonstrates:
(6) At the world Zv29k\ blurring past Zozrs\ on the street.
[…]
You better Z!aD3?\ take care ZjD?\in the community Zj?!lit9m?3H\
The community Zj?!lit9m?3H!aD3?\better take care ZjD?\
There is no immunity ZH!lit9m?3H\
Everybody ZDuqH!a@cH\ must bear ZaD?\ his share ZRD?\.
SF1 being realised as /z/+ SF2 as unrounded ZDuqH!a@cH\ and SF5 consis-
tently as a flap rather than a voiceless plosive point towards an American
reference style, which could well be the result of the choice of tune and
orchestration. However both SF3 (/`H/ diphthong rather than the Confeder-
ate vowel) and SF4 (absence of rhoticity) are relatively consistently real-
ised according to an SBE/RP model.
By contrast, we have a British reference style in “The Enemy Within”,
the Thatcherite reference to the miners during the 1984 Miners’ Strike. The
tune is based on a hymn and the arrangement features a colliery band. It
may well be that all elements come together in this song, the topic (the
miners’ strike), the setting (the practice room of a colliery band) and the
reference style. All style features from SF2 to SF5 are consistently
SBE/RP, as illustrated to an extent in (7), but there are two features in the
excerpt which deserve special attention.
(7) … the colliery band Za`mc\ in front

Some said we’d been defeated ZcH!eh9s?c\
but it felt ZeD“4s\ as if we’d won
As we can see SF1 is realised closer to /a/ rather than /z/, and the /D/ in
“felt” seems to be lowered somewhat in comparison to the SBE/RP pro-
nunciation; these two elements would seem to point towards a “folky” rural
style with a slight Northern English flavour which we can find in the sing-
ing style of some folk singers with rocky leanings as a shift towards what
could perhaps be called a “generic non-urban” singing style.
This generic non-urban style is not usually in evidence with more tradi-
tional folk singers, whose singing style does not differ markedly from their
speaking style, even when the topic of the song or its origin (and setting)
might make such a shift appropriate. Admittedly, Scottish folk singer Dick
Gaughan sings “Song for Ireland” (1981) in less pronounced Scots than the
traditional “The Bonnie Earl O’Moray” (1978) but nevertheless makes no
attempt at including Irish style features in the former. The same can be said
of Irish singer Christy Moore or Northerner Dave Burland and their respec-
tive versions of Ewan MacColl’s “Sweet Thames Flow Softly”
(1972/1978), which depicts a relationship from the tender beginnings to the
break-up as a journey along the Thames flowing through London. Neither
affect a singing style like MacColl’s, nor do they opt for (perhaps London-
tinged) SBE/RP to reflect the fact that the song was originally written for a
radio adaptation of Romeo and Juliet set in London.
In contrast to the tendency among traditional folk singers not to adopt a
singing style markedly different from their speaking style, English folk
rockers like Fairport Convention, The Albion Country Band and Steeleye
Span have a tendency to make use of certain features that would signal the
“generic non-urban” singing style that McTell uses in “The Enemy
Within”. Bell (1984) has made the case that for Outgroup Referee Design it
is enough that the style chosen contains the features associated with the
outgroup, but need not be so convincing as to pass muster with it. The to-
ken gestures of the folk rock singing style illustrates this, perhaps even
going beyond the concept of an existing outgroup because the style fea-
tures that can be observed do not seem to belong to one actual style. Thus,
apart from a lowering of /z/towards Z`\ alongside Z@9\and a similar low-
ering of the unrounded front vowels /d/ and /D/, we have /`T/ and /?T/
sounding relatively close to Southern rural styles (cf. Trudgill 1999). To
complete the picture of a lack of specificity in terms of dialect area and
group we sometimes have a slight raising, possibly with some lip-rounding
of /U/ towards ZT\, but without the full realisation of a Northern vowel.
Consonant articulation includes intervocalic /t/ being usually an aspirated
voiceless alveolar stop, a penchant towards l-darkening also in places
where in SBE/RP the /l/ would be light, a tendency towards lenis pronun-
ciation of some word-initial voiceless stops, in particular /k/, as well as
occasional h-dropping. However, even if the subject matter suggests a rela-
tion to specific regions as in Fairport Convention’s concept album “Babba-
combe Lee” set around the Torquay area, the typical style features which
one would associate with South Western dialect varieties, initial fricative
voicings (Z!yDu?m\ instead Z!rDu?m\(+ lenis articulation of medial obstru-
ents (ZcY`∂fHc\ for jacket) and the typical West country “burr” (cf. Wells
1982: 343–344 and Elmes 2005: 29–30), are largely absent. The style that
emerges shows a mixture of local features, predominately but not exclu-
sively rural, from a variety of traditional dialect areas being added to what
is otherwise a relatively clearly identifiable form of SBE/RP.
3.2. SBE/RP as a dominant reference style
Whereas the picture that emerges for a folk rock reference style is slightly
uneven, an interesting case can be made for a somewhat more uniform
reference style in prog or art rock. Pink Floyd guitarist Syd Barrett’s use of
SBE/RP style features has been referred to briefly at the beginning of this
discussion. To illustrate, in the second line of “See Emily Play” “She often
inclined to borrow somebody’s dreams till tomorrow” the pronunciation of
tomorrow is Zs?!lPq?T\, with the vowel in the second syllable possibly
slightly lowered and fronted (SF2); in the opening line of the second verse
“Soon after dark” Zrt9m !@es? c@9j\, there is not only no trace of post-
vocalic /r/ (SF4), but we also get the typical SBE/RP /@/ in the pronuncia-
tion of after rather than the GenAm /z/ (SF1). In “Arnold Layne” the pro-
nunciation is also SBE/RP throughout, for instance with the rounded back
vowel /N9/ in wall and tall (SF2) in “on the wall hung a tall mirror” and the
clearly non-rhotic Z!lHq?\ (SF4). The adoption of these SBE/RP features,
especially the focus of SF1 (/@/ instead of /z/) has remained a feature in
much of Pink Floyd material, clearly observable in many of the songs sung
by Barrett’s successor Dave Gilmour (cf. for instance “Comfortably
Numb”, where Gilmour sings “I can’t Zj@9ms\ hear what you’re saying”).
But also in “Another Brick in the Wall Part II” (both 1979) sung by Roger
Waters we have “no dark Zc@9j\ sarcasm Z!r@9jzyl<\ in the classroom
Z!jk@9rqt9l\” although in other places Waters may use the GenAm pro-
nunciation for instance SF2 in “is there anybody ZdmH!a@9cH\ in there” and
SF5 in “on your feet again ZeH93?!fDm\” in “Comfortably Numb”.
What may account for the adoption of SBE/RP as the reference style for
much of prog rock, at least in Britain, is that these compositions aspired to
be much more sophisticated than the standard three-minute pop song; with
the increasing length and musical complexity of the material and its aspira-
tions towards “serious” music, the lyrics had to move away from the
(American) Rock’n’Roll model towards the language of the literary sophis-
tication of the – theatrical – stage, in other words towards SBE/RP.6
Another reason for the emergence of SBE/RP as a reference style in
prog rock may be the celebration of Englishness, possibly resulting from
the popularity of English bands with audiences worldwide in the second
half of the Sixties. At that time, British musicians met with considerable
success in the US and the resulting tours by British bands was often re-
ferred to as the “British Invasion”. This rise and the attendant increase in
what one could call cultural confidence began with The Beatles, although
at the time, about 1964, they still focussed on American material for their
covers on “Beatles for Sale” (mainly black Rhythm and Blues, Rockabilly
and Country). Robertson (1994: 26) points out that on some recordings this
album “sounds like the peak of British beat music”, a fact that becomes
obvious when we consider the cover of Carl Perkins’ “Everybody’s Trying
to Be My Baby”, on which “Harrison’s scouse drawl made Perkin’s lyrics
almost impossible to decipher” (Robertson 1994: 31). Scouse, however,
had always been part of their media personality, first on the (RP) BBC and
later during interviews in the States (cf. Atkinson 2007: 17). The trend
away from American dominance is evident on the albums that followed in
the next two years, “Help”, “Rubber Soul”, “Revolver”, “Sgt. Pepper”,
which show how the Beatles left their rock roots and embraced new con-
cepts, musically and lyrically. As of “Rubber Soul” they entirely eschewed
cover material, only doing their own songs.
Interestingly enough, Beat music, spearheaded by the Beatles, seems to
have given rise to a reference style of its own, which in turn created a phe-
nomenon not often discussed in the literature, i.e. that not all transatlantic
style imitation went from the New to the Old World. Probably the best-
known example for this are The Monkees. Formed in 1965 by Columbia in
response to the commercial success of the Beatles films A Hard Day’s
Night and Help the group, selected from a field of 400 applicants, com-
prised a carefully tailored line-up of two folk/country musicians, Michael
Nesmith and Peter Tork, and two singing actors, Micky Dolenz and Man-
chester-born Davy Jones, the latter included no doubt for both his boyish
good looks and the added bonus of being a Northerner. The “manufac-
tured” band, “an American Beatles” (Nixon 1999: 858), meant to be the
transatlantic answer to the Fab Four, were often unflatteringly referred to
as “the Prefab Four”.
It is hardly surprising that when Jones sings lead vocals, the pronuncia-
tion is unmistakably British. In “Here Comes Tomorrow” we have SF1 as
Z@9\ in can’t but also some non-standard features such as a glottal stop in
Z?T v@> ? rvh9s f29k\, the “Velar Nasal Plus” in ZkPMf\, a feature typical
of “most of the western half of the midlands and the middle North, includ-
ing … Manchester and Liverpool” (Wells 1982: 365). Possibly the most
interesting instance is Jones’ adoption of specific Northern English style
features in the 1967 hit “Daydream Believer” (8):
(8) Oh, I Z`∂\ could hide Zg@Hc\ ‘neath the wings
…
But ZaTs‘\ it ZHs‘\ rings and I Z@H\ rise Zq@Hr\
The /`H/ diphthong seems to conform mostly to the midland pronunciation
(Wells 1982: 358).7 More surprising is perhaps that Mancunian Jones uses
Scouse t-affrication in ZaTs‘ Hs‘\, which, similar to his more recent fellow
Mancunian Liam Gallaher in “Wonderwall” (see Simpson 1999: 363), may
well indicate an imitation of the Beatles (even though they rarely used
Scouse features in their singing8).
However, SBE/RP features play an important part in the Monkees’ sing-
ing style even when American Micky Dolenz acts as the lead vocalist. “I’m
a Believer”, a number 1 hit in 1966, shows a surprising adherence to a Brit-
ish style in the consistent absence of rhoticity (SF4) (“for” ZeN9\ `mc “be-
liever” ZaH!kh9u?\(, and /`H/ being mostly realised as a diphthong (SF3) in
Zl`H l`Hmc\. Back vowels, on the other hand, seem to follow mostly
American models as in Zm@s\ and Z!g@9msHc\ although there is a relatively
British realisation of ZN94\ in “haunted all my dreams”. Similarly, intervo-
calic /t/, only in evidence at a word boundary in this song (e.g. “not a
Zm@3 ?\ trace”), corresponds to the American voiced flap rather than the
British unvoiced plosive. “Last Train to Clarksville” (1966), on the other
hand, presents a wider spectrum of USA-5 features but here too we have a
mixture of styles.
(9) Take the last Zkzrs\ train to Clarksville Z!jk@9jruHk\.
I’ll Z`Hk\ be waiting Z!vdHsçHM\ at the station.
We’ll have time Zs`Hl\ for coffee Z!jPeH\ - flavoured kisses
And a bit of conversation Z$j@9mu?!rdHRm<\.
SF1 unmistakably follows the American model, SF2 once again veers be-
tween GenAm and SBE/RP, whereas SF3 and SF4 are consistently British.
That Dolenz’s style seems so strongly influenced by SBE/RP is somewhat
unexpected when we consider the song’s subject matter – a young man
about to be drafted and shipped off to Vietnam (hence “I don’t know if I’m
ever coming home”) arranging a last date with his girlfriend – as well as
the geographical reference, i.e. Clarksville, Tennessee near Fort Campbell,
Kentucky, the young man’s presumed destination. What this would point
towards, on the one hand, is the strength of British cultural dominance in
pop and beat music at the time, which led to the adoption of British fea-
tures. On the other hand, it demonstrates that reference style in popular
music, when it goes from Britain to the New World, is subject to the same
lack of consistency in the adoption of salient features that would apply to
British singers aiming for an American reference style.
Another instance where the reference style is SBE/RP with an Ameri-
can singer trying to sound British can be observed in Jefferson Airplane’s
“White Rabbit” (1967). The title alludes to Alice in Wonderland and the
musical style is reminiscent of Pink Floyd’s prog rock / psychedelic piece
“Astronomie Domine”. Both factors, but certainly the former, may be the
reason for the style choice in this instance. The first verse (10) shows a
number of interesting elements
(10) …and one pill ZoHk\makes you small ZrlN9k\+
And the ones that mother Z!lUC?\ gives you
Don’t do anything at all Z?sçN9k\-
Go ask Z`rj\ Alice
When she’s ten feet tall ZsçN9k\.
Most style features follow the SBE/RP model: low back vowels are
rounded in tall and at all (SF2), mother is non-rhotic (SF4), SF5 is realised
as an aspirated /t/ (at all) and /a/ in ask (SF1), while not corresponding
entirely to SBE/RP seems closer to a Northern English than a GenAm pro-
nunciation. It is intriguing, however, that the /l/ in pill, all and tall is pro-
nounced as a light [l] where we would expect Z4\ in SBE/RP; this appears
to be a hypercorrection of a (perceived) British feature, which results in
something resembling an educated Irish accent.
While American singers adopting SBE/RP features is a relatively rare
occurrence, the weakening of the American model as the only or the
strongest reference style as of the mid-Sixties (Trudgill 1983) is evident in
a number of cases, for instance in the Small Faces “Lazy Sunday After-
noon” (1968), which has mainly Cockney vocals. What is interesting, how-
ever, is that the last chorus after an instrumental break has lost all Cockney
features and, in keeping with the R&B musical style, reverts to the Ameri-
can style model in all features in evidence, i.e. SF1 [zes?!mt9m\, SF2 Zf@s\,
SF3 Zl`∂ `∂y\.
A similar tendency towards a British reference style can be observed in
Brit Pop and Indie Rock of the Nineties. Both genres owe an allegiance to
the British musical past, the former to the Sixties, the latter more to the
punk era of the Seventies and early Eighties. In several cases the tendency
is to employ regional features, e.g. the Scouse t-affrication in Mancunian
Oasis’s song “Wonderwall” referred to above. Other examples include the
Colchester band blur (11) or Catatonia from Wales (12).
(11) City dweller Z!cvDk@\, successful fella Z!eDk@\
… I’m caught in a rat race [jN9> Hm ? q`> qdHr\ terminally
(“Country House” 1995)
(12) You should be making it easy on yourself Z!ldHjHM H3 !h9yH @m iN9rDke\

…
It’s all over Z!ntu?\ the front page, you give me road rage, Z!qntc 3dHcY\
--- possibly ZQdHcY\
you’re driving Z!c3`HuHM\ me crazy Z!j3dHyH\ (“Road Rage” 1998)
In (11) we have evidence of Southern English features, for instance, non-
rhoticity with a tendency towards an open rather than a central word-final
vowel for dweller and fella. SF2 follows the SBE/RP rounded back vowel
and /t/ mostly the “Estuary” style model as a glottal stop. Similarly there
are typical Welsh elements in (12), notably the Znt\ in over and road and
the absence of rhoticity (SF4), but probably the most strikingly Welsh fea-
ture is the /r/ in rage, driving and crazy. Carmosino (1999: 7) has argued
that /r/ in rage is a uvular /Q., a rural Welsh feature known as tafod tew
(thick tongue) (Wells 1984: 390), but this is a little difficult to assess be-
cause of the instrumentation. It is also possible, and in the case of driving
and crazy more obvious, that what we are faced with is the flapped realisa-
tion – or possibly a very short alveolar trill – that is also widespread in
Welsh English (Wells 1984: 390).
Two points are noteworthy here: firstly, the commercial success of Brit
Pop and Indie Rock on both sides of the Atlantic may well have been a
driving force for the tendency to retain British reference styles. Clarke
(1998: 135) has suggested that Brit Pop and Indie Rock of the Nineties are
characterised by a “backward-looking anglocentricity”, and thus hark back
to periods in which there was a similar cultural confidsence as during the
Sixties and Seventies.9 Secondly, the prominence of these features hints at

an interesting phenomenon, which Carmosino (1999) links to Giles and
Couplands’ concept of ethnolinguistic identity (1991): the marked use of
such features can “accentuate ingroup communicative markers” (Giles and
Coupland 1991: 97). This ties in with Bell’s notion of Ingroup Style De-
sign and would explain the salience of regional style features connected to
the bands’ origins or possibly their real or suggested social background as
they emphasise group membership, i.e. (regional) Britishness, with “mem-
bers of an outgroup”, the audience, “with a shift towards the speaker’s own
(absent) ingroup” (Bell 1984: 187).
4. Beyond sociolinguistics
What is interesting in the case of Catatonia (12) – as well as in other ex-

amples – is the presence of GenAm features, i.e. SF2 realised as an un-
rounded lower back vowel and SF5 flapped, which represent some of the
many inconsistencies in style choice presented in this chapter and referred
to in the literature. The questions they raise are summed up in the conclu-
sion of Trudgill’s 1983 paper:
… even in the narrow field of pop-song pronunciation it is not possible, in
terms of [Le Page’s] theory [of linguistic behaviour] (or any other), to ex-
plain why particular … consonantal, vocalic or other variants are retained,
rejected or selected, and not others. We may therefore await theoretical re-
finements, since it may be the case that until we are better able to account
for why, say, the Clash … sing Zj@9msfDcŒ?!gDc\ (rather than
ZjzmsfDcŒ?!gDc\ or Zj@9msfD>?!gDc\, we as sociolinguists may perhaps
not be able to make much progress either. (Trudgill 1983: 159–160)
The present discussion neither claims to present these theoretical refine-
ments, nor to advance the theory that would help the sociolinguistic pro-
gress Trudgill refers to. However, a partial answer to the question why The
Clash, like so many others, are inconsistent in their use of SF5, among
other style features, and why they opt for a GenAm rather than a British
realisation, could well stem from the fact that the data we are examining
are performances of songs. As such they are subject to rules and restric-
tions not necessarily and certainly not only governed by sociolinguistic
considerations.
As any vocalist appreciates, there are certain speech sounds that lend
themselves better to singing than others because, to put it simply, they
“carry” the tune. Hawkins, somewhat more technically, points out that
“more ‘sonorous’ sounds have greater carrying power …, which corre-
sponds in articulatory terms to the freedom of passage of air through the
vocal tract” (1984: 98). Whereas (voiceless) stops restrict the free passage
of air, vowels, at the other end of the spectrum, have the greatest carrying
power; particularly “[o]pen vowels are (as any singer appreciates) more
sonorous than close vowels” (Hawkins 1984:98). In other words, the more
sonorous speech sounds vocalists can use in their singing, the easier it is to
hold their own against the electric competition of the band.
Sonority of speech sounds can be illustrated with the following dia-
gram:
Most sonorous low (open) vowels

mid vowels
high (close) vowels
glides
flaps
laterals
nasals
voiced fricatives
voiceless fricatives
voiced plosives
voiceless plosives
Least sonorous complex plosives /affricates
Diagram 2: Sonority scale (Burquest and Payne 1993: 101)

This scale ignores the issue of lip-rounding, which would also affect sonor-
ity because the relative opening through which the sound escapes is re-
duced in comparison to unrounded (and not spread) vowels produced in the
same articulatory region.
Reconsidering the style features discussed above, we see an interesting
pattern emerge: several of the favoured – and, one could postulate, un-
marked – style features, i.e. the GenAm realisations, have a higher sonority
than their SBE/RP equivalent. SF5 is a case in point. Intervocalic /t/ real-
ised as a voiceless stop, irrespective of whether it is alveolar or glottal, has
a very low sonority; by contrast, if we accept the sonority scale above,
flaps are considerably more sonorous, which is also the reason why it
makes sense to transcribe this speech sound in singing as Z3\ rather than
Zc\ or ZcŒ\. In practical terms the SBE/RP realisation of /t/ or a regional Z>\-
variant means that the sound produced by the two vowels is interrupted by
a voiceless stop, which is difficult to sing unless the song allows for a stac-
cato phrasing. This is the case, for instance, in the chorus of the Small
Faces’s “Itchycoo Park”, an acoustic, bluesy song with relatively strong
SBE/RP style features. “It’s oh so beautiful” is realised as Z!ait9sHeTk\, i.e.
with an aspirated /t/ with each syllable articulated separately. However, in
(12) the phrasing for “making it easy” is much more legato. The same is
true for “to make it better Z!aDsç?\better Z!aD3?\better Z!aD3?\ in the
Beatles’ “Hey Jude” (1968) and in “We’re so pretty ZoqH3dH\+oh so pretty
ZoqHsçH\” in “Pretty Vacant” (1977) by the Sex Pistols (cf. Simpson 1999:
349). In both cases we have the SF5 GenAm and SBE/RP variants side by
side, but the musical phrasing in “Hey Jude” would make it almost impos-
sible to pronounce the intervocalic /t/ as a voiceless stop in the ascending
melody after the first better. The Sex Pistols may simply sound inconsis-
tent but it is at least possible to interpret this as vocalist Rotten succumbing
to the temptation of opting for the more “singable” flap, even in this ten-
dentially more staccato phrasing. To put it another way: the flapped reali-
sation of intervocalic /t/ is relatively sonorous and thus more singable,
which is why we can take it as the unmarked form, particularly in legato
phrasings; by contrast, the unvoiced stop /t/ may represent a marked reali-
sation, which signals a conscious choice of a British reference style as is
the case in “Itchycoo Park”.
Singability may also have an impact on rhoticity (SF4). A postvocalic
/r/ represents a closure in comparison to the position of tongue and mouth
during the pronunciation of the vowel. For the singer this creates two prob-
lems: firstly, the closure reduces the opening for the air flow, thus reducing
sonority, and, secondly, it requires a decision as to when the tongue should
begin to move from the vowel constellation towards the alveolar approxi-
mation of the Z¢\. For a singer, non-rhoticity may therefore be preferable,
particular in sustained notes. In shorter notes, on the other hand, it is easier
to produce the approximant, because the two problems mentioned above
represent less of an issue. Nevertheless, they may be a reason why non-
rhotic realisation is relatively widespread, especially in monosyllabic
words, at the end of a line in the lyrics or generally in a sustained note, an
observation which also applies to singers not actively aiming for an
SBE/RP or a Southern/Black Southern reference style (3a and 3b).
What holds true for the difficulties in producing post-vocalic /r/ also
applies to SF3, the .`H.-diphthong and to an extent to yod-dropping, al-
though in the opposite direction: the diphthong requires a movement from
a relatively sonorous open vowel to a less sonorous closed vowel. Both, the
reduction of sonority and the need to decide when this movement is best
executed, create a problem for singers, which can easily be avoided if a

monophthong is substituted. This is clearly the case in a long note, where
the decision when to start the move is not an easy one; furthermore, notes
sustained on .h9. create a screechy, rather unpleasant sound and are there-
fore avoided. On a very short note the problem of the screech obviously
does not apply but may make it difficult to execute the diphthongal glide.
The decision to produce a SBE/RP /`H/ is therefore probably limited to a
middle-length note on a syllable with a coda. In yod-dropping the move is
from the relatively less sonorous glide /j/ to a more sonorous vowel; this
means that the potential difficulties mentioned above apply, albeit in the
reverse direction. Life for a singer may therefore be easier if the (Ameri-
can) realisation with yod-dropping is an option.
As far as SF2 is concerned, we can perhaps say that this is, together
with SF5, the feature which is used most inconsistently as far as adherence
to a reference style is concerned. Our examples show that singers who use
a SBE/RP or a British regional reference style may use the unrounded back
vowel Z@\in preference toZP\and ZN9\. Ringo Starr’s “I need somebody
Z!rUla@cH\ to love” in “With a Little Help from My Friends”, where he
also uses the “velar nasal plus” feature in “if I sang Zr`Mf\out of tune”
illustrates this preference. Avoiding the reduced “freedom of passage of air
through the vocal tract” (Hawkins 1984: 98) as a result of lip-rounding
could present an explanation for the prevalence of the unrounded and more
open back vowels that we can observe in so many of the examples dis-
cussed in this paper. Therefore, if Grace Slick not only opts for an aspri-
tated /t/ but also the rounded back vowel in Z`sçN9k\ (10), we can postulate
that this constitutes a marked form, a clear signal that the reference style
used here is SBE/RP in keeping with the Carrollian theme of the song.
This leaves us with the one style feature not affected by sonority, the
choice between Zz\ and Z@9\ (SF1). Both are open vowels and their length
can be varied and thus adapted to the needs of melody. Therefore, when the
Beatles sing “I can’t [kz}nt] get you out of my mind” in the Coasters’ cover
“Young Blood”, they plainly adopt the (black) American pop vocalists’
reference style.10 In the same way, Sophie Ellis-Bextor pronunciation of
Z!c@9mrekN9\ in “Murder on the Dance Floor” (2002) must represent a con-
scious decision to opt for an SBE/RP reference style, possibly to convey an
image of Sloane Ranger poshness demonstrated in the video.
5. Conclusion
In conclusion we can say that there are a variety of elements that affect
style choice. There is clearly the issue of reference style which may be
influenced by the origin of the song, the demands of the genre, to a lesser
degree the topic and to a certain extent the cultural confidence in the image
the performers are trying to project. Many of these choices reflect what
could be considered an unmarked style, where the genre and the reference
style are in keeping with each other, i.e. an American pop song would be
sung with American style features, a Sixties beat number may be studded
with English or even Northern style features, a folk rock ballad would use
elements from a variety of rural English dialects, etc. By contrast, a devia-
tion from this unmarked use of a reference style may lead to a variety of
effects, often comical, e.g. Spinal Tap’s prog rock parody “Stonehenge”
(1984), where the spoken intro and much of the lyrics are intoned in a Lon-
don accent (including h-dropping in Zrs`Tm!Hmf\), or the Bonzo Dog Doo
Dah Band’s SBE/RP pronunciation of the blues lyrics in “Can Blue Men
Sing the Whites” (1974).
However, a discussion of the phenomenon of style choice in popular
music must also take into account the aspect of performance and its impact
on pronunciation. It is a truism that singing is not speaking and that singing
style and speaking style are therefore subject to different parameters. As
the literature makes abundantly clear, singers do not have the sociolinguis-
tic knowledge that would allow them to monitor the accuracy of their
adopted singing style, but it is clearly also important to take into account
that their performance needs to conform to different criteria, one of the
foremost being lyrics affording the potential for carrying sound. For this
reason it makes sense to broaden the scope of the discussion of reference
styles to include the notion of sonority and to what degree it can help to
answer the question that Trudgill raised at the end of his seminal 1983
paper.
Notes
1. In this paper I shall use “popular music” as an umbrella term for rock, pop and
modern folk music, both traditionally influenced and contemporary.
2. Liverpudlian dockers and sailors who first brought recordings of American
Rock music to Britain in the Fifties were known as “Cunard Yanks”, Cunard
being one of the principal shipping lines by which these records and other
American goods were brought across the Atlantic.
3. Compare, for instance, the recording of Eric Clapton’s “Layla” (1972) with
the unplugged version of 1992.
4. I shall use the /3/ for reasons that will emerge later in the discussion but as
Trudgill (1983: 141) has pointed out the sound is “a voiced alveolar flap of
some kind.”
5. This could be a Cockney feature but the pronunciation makes this possibility
rather unlikely.
6. Theatricality was also a strong feature in the almost operatic performances
with elaborate costumes and stage effects (e.g. Peter Gabriel with Genesis and
the legendary special effects at Pink Floyd concerts).
7. Nevertheless there are some instances of American realisations, e.g. SF3 for I
in the first line.
8. Lennon’s pronunciation of git Zfds\ in “I’m so tired” (“White Album”), the
Liverpudlian style in “Polythene Pam” or the Scouse /r/ in “spinal cracker” on
“Come together” (“Abbey Road”) are relatively rare exceptions.
9. For a more detailed discussion of Brit Pop, cultural precedents and the percep-
tion of it representing a reaction against American dominance in popular mu-
sic compare Mitchell (1996).
10. The recording represents a wild mixture of styles in the line “just look at her”
sung in succession by all four band members, running the gamut from GenAm
to Scouse. This may be a sign of markedness as the band are known to have
tired of doing BBC studio session at the time and may have been sending up
the song with this strategy.
References
Atkinson, Pete
2007 Scouse: The accent that defined an era. The Times Higher, June 29,
2007, 17.
Bell, Allan
1984 Language style as audience design. Language in Society 13 (2): 145–
204.
2001 Back in style: Reworking audience design. In Style and Sociolinguis-
tic Variation, Penelope Eckert and John R. Rickford (eds.), 139–169.
Bowie, David
2007 BBC News website. Available at
http://news.bbc.co.uk/2/hi/entertainment/5169344.stm. Accessed on:
11.06.06/20.3.07
Buckley, Jonathan, Orla Duane, Mark Ellingham and Al Spicer (eds.)

1999 Rock: The Rough Guide. 2nd expanded and revised edition. London:
The Rough Guides.
Burquest, Donald A. and David L. Payne
1993 Phonological Analysis: A Functional Approach. Dallas: Summer
Institute of Linguistics.
Carmosino, Giovanni
1999 “You wanna sing like the common people”: British regional accents
in 1990s British pop music. Unpublished paper, Berne: University of
Berne.
Clarke, David
1998 Penguin Encyclopedia of Popular Music. 2nd revised edition. Har-
mondsworth: Penguin.
Clayson, Alan
2002 Jake Thackray: Original lyricist and founding father of the English
chanson. Guardian, 28 December.
Elmes, Simon
2005 Talking for Britain. A Journey Through the Nation’s dialects. Har-
mondsworth: Penguin.
Ferguson, Charles A.
1959 Diglossia. Word (15): 325–340. Reprinted in J. B. Pride and J
Holmes (eds.). 1972. Sociolinguistics. Harmondsworth: Penguin.
Flender, Reinhard, and Hermann Rauhe
1989 Popmusik. Aspekte ihrer Geschichte, Funktionen, Wirkung und Äs-
thetik. Darmstadt: Wissenschaftliche Buchgesellschaft.
Frith, Simon
1988 Performing Rites: On the Value of Popular Music. Oxford: Oxford
University Press.
Giles Howard and Nikolas Coupland
1991 Language: Context and Consequences. Milton Keynes: Open Uni-
versity Press.
Graves, Barry and Schmid-Joos Siefrid
1990 Das neue Rock Lexikon: Gianna Nannini to ZZ-Top. Volume 2.
Reinbek bei Hamburg: Rowohlt Taschenbuch Verlag.
Hawkins, Peter
1984 Introducing Phonology. London: Edwards.
Jenkins, Paul O.
2003 Recurring Themes in the Songs of Ralph McTell.
http://www.mctell.co.uk/paul_jenkins_essay.htm. Accessed 22.6.07.
Mitchell, Tony
1996 Popular Music and Local Identity. London and New York: Leicester
University Press.
Nixon, Neil
1999 The Monkees. In Buckley et al., 856–857.
Robertson, John
1994 The Complete Guide to the Music of The Beatles. London, New
York, Sydney: Omnibus Press.
Shapiro, Peter
1999 The Rolling Stones. In Buckley et al., 831–834.
Simpson, Paul
1999 Language, culture and identity: With (another) look at accents in pop
and rock singing. Multilingua 18 (4): 343–367.
Swann, Joan, Ana Deumert, Theresa Lillis and Rajend Mesthrie
2004 A Dictionary of Sociolinguistics. Edinburgh: Edinburgh University
Press.
Trudgill, Peter
1983 Acts of conflicting identity: The sociolinguistics of British pop-song
pronunciation. In On Dialect; Social and Geographical Perspectives,
Peter Trudgill (ed.), 141–160. Oxford: Blackwell.
1999 The Dialects of English. Second Edition. Oxford: Blackwell.
Wells, J. C.
1984 Accents of English. Volumes 1 and 2. Cambridge: Cambridge Uni-
versity Press.
Part II
English usage in non-native contexts
Chapter 11
‘Standard’ English, discourse grammars and
English language teaching
Tony Bex
1. Introduction
In this chapter, I intend to consider some of the emerging views of dis-

course grammars with particular reference to the CANCODE project. This
is a unique project established by the universities of Cambridge and Not-
tingham in which recordings of spoken English comprising some five mil-
lion words were collected between 1995 and 2000. It is unique because the
corpora have been coded, among other things, according to the relation-
ships between the speakers (Cambridge International Corpus). In particu-
lar, I shall be considering the proposed applications of this research in lan-
guage teaching (e.g., Carter 1998, 1999; Carter and McCarthy 1995;
Hughes and McCarthy 1998; McCarthy and Carter 1995; McCarthy 1998).
I shall be concentrating on English language teaching (ELT) because I
believe English has a unique status in the world with a variety of functions
that is not matched by other languages. In passing, I shall refer to the de-
bates concerning ‘standard’ English. My concluding argument will be that
the native-speaker English teacher has become an expensive irrelevance for
those countries which teach English as a second language, and that such
teaching should be restricted to countries which have English as a first
language.
2. Grammar and discourse grammars
For these purposes, it is necessary to investigate the nature of grammar,

and its place within language teaching projects. Traditionally, language
teachers have regarded grammars as sets of instructions for producing well-
formed sentences within the target language. Typically, such instructions
are treated as the ‘rules’ of a language. There are two major problems with
this characterisation: the first being a failure to ascertain the nature of such
222 Tony Bex
rules, and the second being a failure to recognise that such grammars only
represent a very small part of a dialect of the language so-called.
Clearly, such ‘rules’ cannot be immanent within the language since this
would involve proposing that languages can exist independently of their
speakers. It could be argued pace Chomsky and his followers that they
somehow exist within speakers’ heads. But this involves treating grammar
(and I have chosen the singular form deliberately) as universally abstracted
from individual languages, and of little interest to the language teacher.1 It
would thus seem that such ‘rules’ are the construction of particular speak-
ers and are therefore normative rather than natural. This normativity de-
rives from those speakers who typically use the dialect that has been codi-
fied which, in the case of English, is typically referred to as ‘Standard
English’.
It could be argued that these are ‘Saussurean’ grammars which give par-
tial synchronic accounts of the language. But such grammars give a dis-
torted picture of how meanings are actually made. The whole process of
producing a meaningful utterance is far more provisional than such abstrac-
tions allow and as discourse grammarians are discovering. As Cumming
and Ono (1997) state in their extremely useful introduction to this area:
the actual production of syntax is locally managed – that is, transpiring in
real time, second-by-second, and always contingent on negotiation with the
other participants in the speech event. Moreover, although the grammatical
patterns resulting from this process would often be considered ‘syntactically
ill-formed’ in traditional accounts, interactants are extremely tolerant of
such constructions. (Cumming and Ono 1997: 132)
Of course, this is not to argue that grammar does not exist. There may well
be a universal semantic grammar which identifies processes, participants
and circumstances, but in any particular language the items (and their in-
flections) which can fill these slots are not predetermined but are subject to
the exigencies of the communicative event. Thus, the Saussurean distinc-
tion between langue and parole dissolves since parole is langue as it un-
dergoes a process of grammaticalisation (Hopper and Traugott 1993;
Toolan 1994).
If this is true, and it gains further support from researchers (e.g., Bechtel
and Abrahamsen 2002) who are investigating neural networking within the
brain, it poses a dilemma for language teachers for the particular grammars
they are teaching are necessarily always provisional. Widdowson (1997)
has captured this insight neatly when he asserts à propos systemic func-
tional grammars:
‘Standard’ English, discourse grammars and ELT 223
The question arises though as to how far a grammar can actually account for
language use, and how far there is indeed a correspondence between catego-
ries of description and instances of communicative behaviour. (Widdow-
son 1997: 149)
If this is a problem with systemic functional grammars, it is no less a prob-
lem with other types of grammar. Certainly, discourse grammars are less
normative than traditional descriptive grammars have tended to be (cf.
Taylor 1997: 136) since they are no more than a model of what has oc-
curred. However, there is always a risk that they may become predictive,
and this is what seems to be happening with the findings from the CAN-
CODE project.
It is clear that Carter and McCarthy are attempting to develop a dis-
course grammar of the spoken language for they adopt a similar position to
that of Cumming and Ono:
… speakers make different grammatical choices according to the context in
which the language is used … it is also significant that speakers regularly
make choices which reflect the interactive and interpersonal nature of the
communication. (McCarthy and Carter 1995: 211, emphasis in original)
However, they go further in arguing that:
[a] more discourse-oriented approach, we suggest, acknowledges the indis-
soluble link between structure and function in context and aims to incorpo-
rate systems of appropriateness and use at the heart of the explanation.
(Hughes and McCarthy 1998: 281)
Although such an aim is highly desirable, it is ultimately unrealisable sim-
ply because it involves idealisations at three separate levels and of three
different kinds of entities.
The first idealisation is that which all linguists typically practise: that of
identifying a set of features which count as the units of language (i.e.
words, sentences, etc.). Without some such idealisation, the study of lan-
guage would be well-nigh impossible. However, the next idealisation, that
of relating given structures with given functions, is, as already indicated,
highly questionable. At best, it is a convenient fiction derived from obser-
vations as to how linguistic utterances appear to ‘mean’, although it comes
with no guarantee that such utterances will always ‘mean’ in the same sort
of way. The third idealisation is the most illegitimate, since it involves
assuming that contexts are replicable. A moment’s reflection will show that
this is strictly impossible since all contexts are necessarily unique either by
virtue of the times at which they are occasioned, or by virtue of the partici-
224 Tony Bex
pants involved, or by both of these. There are some theorists (e.g. Halliday
and Hasan 1985) who would argue that text and context are indissoluble in
that texts construct their own contexts. Although this is an attractive ap-
proach, it is ultimately untenable since, while texts are necessarily part of
contexts, there are other random elements which may be salient to partici-
pants involved in linguistic exchanges (cf. Sperber and Wilson 1995).
These problems are even more apparent in their claims that the mean-
ings they are interested in are not simply semantic but also affective (e.g.
McCarthy and Carter 1997). While I have argued that grammars are ‘emer-
gent’ at the moment of utterance, it follows that the same will be true of
semantic meaning. Thus, although it may be the case that the investigation
of linguistic units at word or sentence level can reveal a great deal about
the conceptual and logical organisation of a particular utterance at a par-
ticular time, it gives us no guarantee that any particular speaker or hearer
will activate precisely that concept or set of concepts upon hearing the
utterance at another time. And it becomes even more difficult to argue that
the affective meanings of any particular utterance bear a direct relation to
the forms of that utterance. Indeed, if such were the case it would be ex-
tremely difficult to understand how two or more people could fail to agree
whether a given utterance were meant sarcastically. If meanings (whether
affective or semantic) were realised by forms, such misunderstanding
would be impossible. The counter-argument that such misunderstandings
derive from the different pragmatic circumstances in which the utterance is
processed will not work here because either these pragmatic circumstances
include the affective states of the interlocutors, in which case the forms do
not map as proposed, or the utterances lead directly to affective states, in
which case the affective interpretations of the utterance should be identi-
cal.
This being the case, it adds weight to my earlier claim that the grammar of
any particular exchange and the meanings that derive from the linguistic
tokens that are exchanged are always sui generis. As Roy Harris (1998) puts
it
the signs that occur in first order communication are those that the partici-
pants construe as occurring, and what is signified is what the participants
construe as having been signified. There is no higher court of appeal. (Har-
ris 1998: 21, emphasis in original)
One way of capturing this insight has been explored by Potter (1996) who
draws on ethnomethodological research to show how the constructed
meanings of particular exchanges are always the result of ‘work’.2 Al-
though Potter does not explore this area in detail, it seems reasonable to
assume that the emergent grammars of each utterance are the result of this
work and are shaped rhetorically not merely to represent a state of affairs
in the world but also to persuade the interlocutor that such a state exists.
It further follows that the emergent grammars are the ‘possession’ of the
speakers engaged in the exchange and that native speakers have no abso-
lute lien of possession on the grammar of any particular language. This
may seem a fantastic claim since it is clearly the case that native speakers
are usually able to use ‘their’ language more effectively than non-native
speakers and they also possess a richer metalanguage for discussing it.
However, it is not so surprising when we remember that such speakers are
employing the conventionalised grammatical forms on a day-by-day basis.
Clearly, the role of memory comes into play so that what has worked be-
fore will be used again when the situation appears to warrant it. In this
sense, the grammar of a particular language represents the collective mem-
ory of those who use that language most frequently.
To date, grammars of English have been of written English and the par-
ticular value of the CANCODE project lies in demonstrating that spoken
English operates within different constraints. Carter (1999) gives a useful
summary of some of these findings, identifying such features as heads and
tails, ellipsis, different word orders and the uncertain status of clauses, and
observes:
that there are forms of spoken English which are perfectly standard and
which are indeed grammatically correct. These forms do not appear in stan-
dard grammars, however, and it is easy, therefore, for them to be judged as
non-standard and ungrammatical. (Carter 1999: 158)
There is no doubt in my mind that the CANCODE findings are of immense
interest. Not only have they disclosed previously hidden features of the
language and described their distributional features; they have also indi-
cated some of the ways in which speech forms are similar to, or different
from, written forms. Their educational potential is beyond question in that
they can help the mother-tongue teacher of English develop appropriate
speech or writing habits in their pupils.
However, they also need to be treated with caution. McCarthy (1998)
observes that:
. . . some things emerge time and time again, whichever words or structures
we look at, and these are that face-to-face interaction brings the people who
use language slap-bang into the centre of investigation; it is simply impossi-
ble to idealise the data away from who said it, to whom, at what point, with
226 Tony Bex
what apparent goals and purposes, in the context of what relationship, and
under what circumstances. (McCarthy 1998: 173)
In the tradition of the discourse grammarians that I cited earlier, there is a
tacit recognition that meanings are constructed in real time, and their gram-
maticalisation is a function of all the variables McCarthy lists. Although
there may be high statistical correlations between given forms and imputed
meanings, there is no guarantee that forms map onto meanings in a non-
problematic way. This is a comfortable observation for mother tongue
teachers of English because they can situate the forms within a cultural
context that is immediately salient. It is far less comfortable for teachers of
English as a second or foreign language since they will have to construct a
cultural context in which such forms achieve meaning, and it may be one
that is alien to their pupils. Carter and McCarthy’s proposals, therefore, to
introduce the teaching of spoken English into EFL based on the findings of
CANCODE need to be treated with some scepticism.
3. CANCODE and ELT
Needless to say, their programme has aroused considerable interest within

the ELT profession. Prodromou (1996) rejects the three assumptions which
he sees lurking beneath McCarthy and Carter’s earlier (1995) paper argu-
ing firstly that what may be interesting to the native speaker of English
may not be to the non-native speaker, at least in the ELT classroom; sec-
ondly, that non-native speakers will not necessarily achieve raised lan-
guage awareness by studying informal native speaker varieties; and thirdly,
he doubts that such varieties can be taught by non-native speaker/teachers.
He also asks the pertinent question (1998: 266) “ ... how much ‘native
speaker reality’ can the non-native speaker and the EFL classroom take?”
echoing an earlier comment from Guy Cook (1998: 62): “For language
teachers the issue remains as to what the principles for selection, idealiza-
tion and simplification should be.”
Prodromou’s pragmatic objections must be taken seriously, for while it
may be possible to incorporate some elements of the research findings of
the CANCODE project into the foreign language classroom in an interest-
ing way, and in a form likely to interest the learner, it would certainly re-
quire a teacher with near native competence. His final question, although
pertinent, could be better reformulated as “How much ‘native speaker real-
ity’ does the non-native speaker and the EFL classroom need?” In this
form, it overlaps with Cook’s concerns.
Prodromou is writing from the Greek perspective, and it may be that
Greek learners of English, under the pressure of the European Union, will
be taught a variety of English that is based on British forms. However,
even when this is the case it does not follow that Greek learners necessarily
wish to sound like, or to imitate, native speaker norms. For them, English
is not a language imbued with the culture of Britain or the US, but that of a
largely deracinated language which has a world function. This being the
case, it is important for the ELT profession to understand exactly how Eng-
lish is situated within the world. Further, because the profession is very
significant economically, teachers need to guard against the charge of ‘lin-
guistic imperialism’ (Pennycook 1994, 1998; Phillipson 1992) whether
conceived of as the increasing ‘Macdonaldisation’ of world cultures, or as
leading to linguicide.
There seems little argument that English is the most widely used lan-
guage for international communication, and it is at least likely that (if only
in the short term) it will be even more widely used with the development
and spread of electronic communication. One consequence of this spread
has been the development of a wide variety of different dialects both so-
cially and functionally and attempts to describe these different Englishes
have been made by Kachru (1985) who distinguishes between three differ-
ent ‘circles’ in which English is spoken. The inner circle contains those
countries in which English is a mother tongue, and in which it is multifunc-
tional and can be used in all domains. The outer circle consists of countries
where it is largely learned at school as a second language but where it also
has a societal role. The expanding circle is effectively the rest of the world.
Here, English is learned as a foreign language either for educational pur-
poses or as a means of international communication. The boundaries be-
tween the outer and expanding circles are fluid in that countries may shift
between them.
Kachru’s characterisation is useful as far as it goes, but it tends to ig-
nore the basic question of what we mean when we talk about English.
Elsewhere (Kachru 1986: 159) he refers to English as pluricentric, arguing
that different countries appeal to different models as their standard. There
is, then, a tension between the centripetal forces which allow us to regard
all these Englishes as dialects of ‘a language’ and the centrifugal forces
which are driving these Englishes apart as they develop more localised
forms to express particular cultural identities. Nevertheless, Kachru ac-
knowledges that the development of English as an international language is
228 Tony Bex
likely to ensure some degree of mutual intelligibility between these differ-

ent dialects. The problem, at least for those who wish to use English as an
international language, is how to maintain and nurture this mutual intelli-
gibility.
One solution would be to adopt a purely laissez faire attitude and as-
sume that the pressure of communicative needs will create its own solu-
tion. To some extent, I am in sympathy with this position, but it is of little
help to governments who wish to nurture the teaching of English in their
schools and universities. They need some model to appeal to and which
they can offer to their pupils as ‘correct’. For countries in the inner circle
and, to a lesser extent, the outer circle this seems largely non-problematic.
There is a range of internal models which are perceived as more or less
‘correct’. For countries in the expanding circle, however, there are no such
models so they have to appeal to some external model and usually one
drawn from the inner circle. But, as Kachru (1986) points out, native
speaker models are not necessarily those which will obviously lead to mu-
tual intelligibility in an international context:
... in the case of English, we must be clear about whom we have in mind
when we talk of participants in a linguistic interaction. What role does a na-
tive speaker’s judgement play in determining the intelligibility of non-native
speech acts that have intranational functions in, for example, Asia or Africa.
(Kachru 1986: 94)
4. ‘Standard’ English versus ‘standardised’ English
Needless to say, Kachru’s view has not been universally accepted. There is
a body of opinion that asserts that there is one (or two, if one distinguishes
between the UK and the USA) variety of English which is multifunctional
and can serve as a model both for mother tongue teaching as well as for
EFL teaching. This model derives historically from the East Midlands dia-
lect as used in the 15th century, and its syntax is predominantly based on
the written language. Typically, it is referred to as ‘standard’ English. If,
however, it can be shown that ‘standard’ English does not exist, at least as
a linguistically describable variety, then Kachru’s functionalist arguments
will be further supported by evidence of a different kind.
In Bex (1993, 1996), I attempted to show the unreality of ‘standard’
English. However, in Bex and Watts (1999: 7–8) we also argued that
throughout history there have been powerful forces attempting to ‘strait-
jacket’ the language into acceptable forms both grammatically and phonol-
ogically. Such forces are frequently represented by people who are deemed
to possess linguistic ‘authority’ and in Bex (1999), I discussed some of the
authorities of the 20th century and noted that they were not obviously
trained linguists and this tradition persists, as exemplified by John Hum-
phrys (2004). Occasionally, however, professional linguists will declare
themselves on the prescriptivists’ side. In recent years, a notable example
has been John Honey (1997) who argues that ‘standard English’, apart
from being based on written forms of the language is that used by ‘edu-
cated’, or prestigious, speakers. He includes among such people those
... [who have graduated] from (often famous) universities, of literary reputa-
tion, or [who have] the ability in all other respects to use the language in
highly acceptable ways – or who are in some other way high-status figures
(like royalty) ... (Honey 1997: 161–162)
This is puzzling for a number of reasons. In particular, although it is quite
clear that Honey’s characterisation of ‘standard English’ is as a class dia-
lect (something that he denies elsewhere), it is unlikely that all the speakers
he lists adopt similar locutions even in similar communicative situations,
and when he argues that these people use “the language in highly accept-
able ways”, he forecloses the debate by not indicating who regards such
uses as ‘acceptable’. But, although Honey’s position is untenable, and has
been effectively refuted by Crowley (1999), he does represent a body of
opinion that believes there is a linguistically describable ‘standard English’
which should be taught to all pupils, and that should also form the basis of
EFL teaching.
My preceding arguments have attempted to show that the grammars of
natural languages are far less stable than were previously thought, and that
the model of English typically recommended to teachers is more of a social
‘myth’ (Milroy and Milroy 1991) than a linguistic reality. However, when
we enter the pedagogical debate, we are faced with a conundrum since
language education, and especially foreign language education, is by its
nature prescriptive. We are therefore obliged to select a set of grammatical
rules and persuade our pupils that they are ‘correct’. But this begs the ques-
tion of where these rules come from. Some theorists (e.g. Carter 1999;
Cheshire 1999), faced with this dilemma, have chosen to assume the exis-
tence of ‘standard’ English while recognising that it has not yet been ade-
quately described and for these people the task is to construct a more inclu-
sive grammar which takes into account the spoken features which have
tended to be ignored in existing grammars.
230 Tony Bex
I believe this is both misguided and potentially dangerous. Leaving

aside the question as to whether there is a ‘standard English’ which can
ever be satisfactorily codified, and recognising that some form of codifica-
tion is essential for any kind of syllabus, we need to ascertain why people
wish to learn English, and then develop a syllabus which is appropriate to
their communicative needs. Although I have no certain evidence, I would
assume that most non-native Englishes are used for business, the dissemi-
nation of scientific information, diplomacy and tourism and that most in-
teractions are between non-native speakers. It therefore leaves me wonder-
ing whether it is appropriate that these speakers should learn the kinds of
spoken grammatical constructions that McCarthy and Carter wish to teach
them and which have developed from native speaker interactions.
While we need to distinguish between learners in different situations,
for the moment I wish to focus on those learners who are learning EFL in a
country where English is the primary means of communication. Although
necessarily heterogeneous, the students I am thinking of are primarily stu-
dents who spend some time in a country where English is the first language
and typically attend a (private) language school. My own experience3 sug-
gests that such students come to Britain with the intention of developing
their English skills for clearly motivated, though not identical, purposes.
The majority of those that I have taught have come from Kachru’s ‘outer’
circle. They, therefore, represent a rather different constituency than the
Danish learners discussed in Preisler’s interesting chapter in Bex and
Watts (1999) in that they are exposed more to ‘English from above’ than
‘English from below’ (1999: 247). Those that I have spoken to have indi-
cated that they regard the possession of a certificate in English as a tool to
further their careers. They have, therefore, commodified the language and
expect the product they are buying to be of the highest quality. In this
sense, they have a clear concept of the standard of English they wish to
achieve, even though they may not know exactly what this entails.
The notions of correctness that they bring with them derive largely from
their experiences of learning English in their home countries and where
their instruction has been highly prescriptive, they tend to assume that
grammatical form, communicative function and consequent meaning have
an invariant relationship. As Bright and McGregor (1970) observed:
In our first language we pick up strong emotional associations within the
home. They are built into the learning situation . . . [But this is not so in the
second language classroom]. The result is a lack of emotional involvement
in the language and hence great difficulty in seeing any other meaning than
plain sense. (Bright and McGregor 1970: 30)
Or, as discussed earlier, they tend to be oblivious to the affective uses of
English.
The teaching staff, who are all native speakers, tend to deal with such
issues pragmatically being subjected to three, often conflicting, constraints.
First, they have their own beliefs about what constitutes ‘correct’ English;
second, they recognise that students may be at a particular stage of devel-
opment and that presenting them with alternative constructions could inter-
fere with the learning process by overloading their memories, and therefore
confuse them; and third, they are following a syllabus (often established by
the given textbooks) which require them to teach a set of items within a
given time period.
The first of these constraints is difficult to quantify and will depend to a
large extent on education and training. However, those that I have spoken
to are comfortable with the local dialect4 (which the students are necessar-
ily exposed to) while recognising that certain forms are unacceptable in the
language classroom. The second and third are best seen as two sides of the
same coin. While some teachers may prefer to downplay constructions
which are problematic either because they are outside the context of the
current syllabus, or because they present peculiar ad hoc difficulties,5 oth-
ers will offer a brief explanation with the promise of a fuller one at the
appropriate time. Much the same will apply to vocabulary items which may
be heard on the street. So, for example, the use of wicked or awesome to
refer to something perceived as excellent can be both acknowledged and
treated as ill-advised at the same time. The assumption here is that students
should not be denied knowledge, but should be made aware that certain
usages tend to be restricted to a social group or particular type of commu-
nicative activity.
Textbooks tend to be less problematic. Although there may be some
teachers who slavishly follow the sequence of items in the book, I have not
met them. The majority seem to supplement the books with their own ma-
terials and ignore items which they feel are badly presented, although they
do have to work within the limits of the overall syllabus designed for stu-
dents of a given level. Perhaps the biggest constraint lies in the tapes, vid-
eos and CDs which accompany certain textbooks. These tend to present
educated south-eastern English as the target pronunciation, although there
has been a recent welcome increase in other regional accents. Perhaps none
of the above will surprise many readers, but the point I wish to make here
232 Tony Bex
is not that the variety of English being modelled is ‘standard’ English, but
that it is ‘standardised’ English (cf. Bex 1993).
5. Standardised English in the wider world
However, it is also interesting to observe how the students interact outside

the classroom. Although they necessarily draw on what they have learned,
the pressures of immediate communication lead to the construction of in-
teresting ‘emergent’ grammars, and this has led me to believe that the
models of English taught in both the expanding and outer circles should be
based on the kind of ‘standardised’ English referred to above but should be
thoroughly expropriated by the country in which it is being taught.
There is an apparent paradox in this suggestion, since by taking control
of English language teaching to such an extent the maintenance of ‘mutual
intelligibility’ would appear to be under threat. Again, I believe this risk to
be more apparent than real. The variety that I have been describing has
been codified largely from written forms, and the extent to which it is held
in common by the majority of writers of English for non-specialised writ-
ing minimises the risk of non-intelligibility. It is clear that as writers ex-
plore their way into more advanced genres of writing, they will need to
develop more sophisticated modes of expression, but these are shaped by
their purposes and are not the preserve of native speakers but of the people
engaged in the communicative acts.
There is a potential problem with phonology, and it could be argued that
native speakers are particularly useful in modelling this feature of the lan-
guage. But, again, this is often overstated. When appeals are made to the
native speaker in this area, one can usefully ask the question ‘Which native
speaker?’ And it is inadequate to respond with British, Australian, Ameri-
can, etc. since within all these countries there are wide phonological diver-
gences. The only possible answer will be to appeal to a phonological model
that is internationally comprehensible, but in doing this one is no longer
necessarily referring to the phonology of any particular native-speaker
since such a phonology can be manipulated by all speakers of English. It
may retain traces of a Spanish, Chinese, Japanese, French accent, but these
traces do not render it incomprehensible. Rather they serve to indicate the
provenance of the variety in the same ways that a Scottish or New York
accent indicates the geographical provenance of the native speaker. In fact,
from my preceding argument it becomes desirable to retain these variations
of accent since they serve as a marker of cultural distinction. In most lin-
guistic exchanges there is a desire both to transmit one’s message as

clearly as possible and to understand to the best of one’s ability what is
being communicated in return. This desire for mutual comprehension over-
comes such local difficulties as accent and pronunciation and, where it
seems not to be occurring, the interlocutors can always request a repetition
or re-formulation. Thus, I am suggesting that Cuban, Chinese, or Japanese
businessmen can negotiate both their business deals and their linguistic
interactions quite successfully by using their own varieties of spoken Eng-
lish. In fact, I am going further and arguing that it is desirable that they
should do so. If this means that they are speaking in ways that seem inap-
propriate to native English speakers, this is insignificant since what is im-
portant is that all users of English, so long as it remains an international
language, should take control of the language and use it to express their
own identities and cultures.
If context is a determining factor in grammatical choice, then the con-
text in which communication between non-native speakers takes place
should also determine the grammars adopted. If they follow native-speaker
norms, then they are implicitly signalling that they wish to convey the val-
ues of such native speakers. This may seem a large claim, but it is relevant
here precisely because all linguistic exchanges take place in a cultural con-
text, and native-speaker exchanges are imbued with the cultures of their
origin. Thus, I am suggesting that any syllabus which follows Carter and
McCarthy’s recommendations will be inadvertently inviting the learners to
become mini-British or mini-American.
A further consequence of McCarthy and Carter’s proposal is that it
privileges native speaker teachers for they are the most likely people to be
comfortable with this grammar. However, this is dangerous for two rea-
sons. To the extent that this grammar is not yet codified, it means that the
textbooks and the teachers will be out of step. There will therefore be a
dissonance between what learners hear from their teachers and what they
read in their books. Of course, this is relatively trivial since successful
classroom practice would be able to resolve the potential problems created
by this dissonance.
More important is the implicit linguistic imperialism inherent in the
proposal. Phillipson (1992) has pointed out the very real dangers of lingui-
cide as English becomes more and more widespread, while Pennycook
(1994) has illustrated how the British Council has used its teaching opera-
tions as a hegemonic practice which bypasses the needs of the countries in
which it operates. It is arguable that Phillipson overstates his case, and it is
certainly not clear to me how we can resist the spread of English so long as
234 Tony Bex
the rest of the world regard it as a valuable resource. However, there are
ways in which the increasing McDonaldisation of the world can be re-
sisted, and one of these ways is for those countries in which ELT occurs to
take complete control of the syllabus and to teach a variety which is appro-
priate to their needs.
Earlier, I argued that countries within Kachru’s expanding circle were
necessarily exonormative in their search for an appropriate variety of Eng-
lish. Initially, it seems reasonable for such countries to look to places
within the inner circle to supply a model. By extension, it also seems ap-
propriate to hire ‘experts’ and even teachers from the inner circle so that
the variety of English is appropriately modelled. However, I have also ar-
gued that, for various reasons both linguistic and social, this may not be the
best way of proceeding. On the contrary, I am suggesting that it is better
for such countries to avoid the importation of inappropriate cultural models
and avoid the economic costs of hiring expatriate ‘experts’.6 However, it
remains to indicate how this can best be done.
6. Conclusion
Following on from what I have argued above, the kind of ‘standardised’

English I refer to has been codified in various grammars and forms the
basis for much English language teaching. This variety is typically used for
non-threatening service encounters and adequately serves for such activi-
ties as tourism. Clearly, it is inadequate for business transactions or more
specialised activities. Nevertheless, it serves as a useful basis from which
learners can develop both more appropriate varieties as they come into
contact with other English users and, if they so desire, a degree of personal
idiosyncrasy which will allow them to express their own personalities.
More importantly, it represents a variety which is likely to have been
learned by the majority of other non-native speakers.
It may seem that this is a utopian suggestion. The alternative, adopting
models derived either from US or UK usages, is dangerous since it leads to
the insidious spread of an alien culture which eventually runs the risk of
dominating the cultures with which it comes in contact. While this is not an
inevitable consequence of appealing to native speakers both for models of
English and teaching expertise, there is sufficient evidence (Phillipson
1992) to suggest that it has occurred in some African and South East Asian
countries, is occurring in Greece (Prodromou 1992), Turkey (Alptekin
1993) and Morocco (Bentahila and Davies 1989) and needs to be resisted.
My argument, then, is that the findings of the CANCODE project are

indeed fascinating, but that they should not be pressed into the service of
EFL teaching. As a discourse grammar, they represent the choices of native
speakers communicating with each other. Their use in mother tongue
teaching of the language is therefore highly desirable. However, their im-
portation into ELT is dangerous both because they involve incorporating
the cultural values of native speaker use and because they privilege the
native speaker teacher of the language. In most expanding circle countries,
English is being learned as a way of representing the countries desire to
engage in international activities. Native speaker varieties of English will
hinder this desire, while the use of native speaker teachers is economically
and culturally undesirable. Rather, such countries should rely on their own
teachers of English who would be able to mediate the countries’ own cul-
tures through the medium of English.
Notes
1. Vivian Cook (1988, 1989), however, argues that it can help in the understand-
ing of second language acquisition and, by extension, help in second language
teaching.
2. The concept of work involves (in speech) joint orientation to the evolving
meanings. Although this has parallels with Grice’s CP, it is different in that it
does not take co-operation as a fundamental given. Rather, co-operation can
be suspended or even resisted when alternative constructions of reality are in-
volved in the interchange.
3. Since 2001, I have been teaching on a part-time basis at Churchill House
School of English Language. This is an independent, average-sized and well-
respected school with an international intake at all levels. I am grateful to the
staff and students for allowing me to draw on my experiences there. My ob-
servations are entirely personal and do not necessarily reflect either the
school’s policy or the views of individual staff members.
4. For example, the Thanet dialect tends to generalise ‘was’ in the past continu-
ous for all numbers and persons.
5. For example, I recall a staffroom discussion of the phrase: “I was sat there,
minding my own business”. Given that it is clearly deviant from the more
common (EFL) expression: “I was sitting there, minding my own business”,
there is a potential problem in explaining it to any student who comes across it
and tries to generalise the form. One possible explanation could be that it is a
dialect form. This, though, seems unlikely since I have heard the expression
used quite widely. It can more easily be explained as a passive form analogous
236 Tony Bex
to “I was hit by a passing bicycle”, whereby speakers are conveying the sense
(affectively and effectively) that they were not participating actively in the ‘sit-
ting’. However, because EFL students are struggling with the language such a
construction is likely to be confusing simply because it fails to fit with the
’rules’ they have learned.
6. These costs can be excessive when one takes into account housing, education,
medical care, air fares, etc.
References
Alptekin, Cem
1993 Target-language culture in EFL materials. ELTJ 47 (2): 136–143.
Bechtel, William and Adele Abrahamsen
2002 Connectionism and the Mind: Parallel Processing, Dynamics, and
Evolution in Networks. 2d ed. Oxford: Blackwell.
Bentahila, Abdel-ali and Eirlys Davis
1989 Culture and language use: A problem for foreign language teaching.
IRAL XXVII (2): 99–112.
Bex, Anthony R.
1993 Standards of English in Europe. Multilingua 12 (3): 249–264
Bex, Tony
1996 Variety in Written English. London: Routledge.
1999 Representations of English in twentieth-century Britain: Fowler,
Gowers and Partridge. In Standard English: The Widening Debate,
Tony Bex and Richard J. Watts (eds.), 89–109. London: Routledge.
Bex, Tony and Richard J. Watts (eds.)
Bright, John A. and Gordon P. McGregor
1970 Teaching English as a Second Language. London: Longman.
Cambridge International Corpus
http://www.cambridge.org/elt/corpus/corpora_cancode.htm
(Accessed on: 6 June 2007).
Carter, Ronald
1998 Orders of reality: CANCODE, communication and culture. ELTJ 52
(1): 43–56.
1999 Standard grammars, spoken grammars: some educational implica-
tions. In Standard English: The Widening Debate, Tony Bex and
Richard J. Watts (eds.), 149–166. London: Routledge.
Carter, Ronald and Michael McCarthy
1995 Grammar and the spoken language. Applied Linguistics 16 (2): 141–
158.
Cheshire, Jenny
1999 Spoken standard English. In Standard English: The Widening De-
bate, Tony Bex and Richard J. Watts (eds.), 129–148. London:
Routledge.
Cook, Guy
1998 The uses of reality: A reply to Ronald Carter. ELTJ 52 (1): 57–63.
Cook, Vivian J.
1988 Chomsky's Universal Grammar. Oxford: Blackwell.
1989 Universal Grammar Theory and the classroom. SYSTEM 17 (2): 169–
181.
Crowley, Tony
1999 Curiouser and curiouser: Falling standards in the standard English
debate. In Standard English: The Widening Debate, Tony Bex and
Richard J. Watts (eds.), 271–282. London: Routledge.
Cumming, Susanna and Tsuyoshi Ono
1997 Discourse and grammar. In Discourse as Structure and Process,
Teun van Dijk (ed.), 112–137. London: Sage.
Halliday, Michael A.K and Ruqaiya Hasan
1985 Language, Context and Text: Aspects of Language in a Social-
Semiotic Perspective. Victoria: Deakin University Press.
Harris, Roy
1998 Introduction to Integrational Linguistics. Oxford: Pergamon Press.
Honey, John
1997 Language is Power. London: Faber
Hopper, Paul J. and Elizabeth Closs Traugott
1993 Grammaticalization. Cambridge: Cambridge University Press.
Hughes, Rebecca and Michael McCarthy
1998 From sentence to discourse: Discourse grammar and English lan-
guage teaching. TESOL Quarterly 32 (2): 262–287.
Humphrys, John
2004 Lost for Words: The Mangling and Manipulating of the English
Language. London: Hodder and Stoughton.
Kachru Braj B.
Learning the Language and Literatures, Randolph Quirk and Henry
Widdowson (eds.), 11–30. Cambridge: Cambridge University Press
in association with the British Council.
Kachru, Braj B.
1986 The Alchemy of English: The Spread, Functions and Models of Non-
native Englishes. Oxford: Pergamon
238 Tony Bex
McCarthy, Michael
1998 Spoken Language and Applied Linguistics. Cambridge: Cambridge
University Press
McCarthy, Michael and Ronald Carter
1995 Spoken grammar: What is it and how can we teach it? ELTJ 49 (3):
207–218.
1997 Grammar, tails, and affect: Constructing expressive choices in dis-
course. Text 17 (3): 405–429.
1991 Authority in Language. 2nd ed. Routledge, London.
Pennycook, Alastair
1994 The Cultural Politics of English as an International Language. Lon-
don: Longman.
1998 English and the Discourses of Colonialism. London: Routledge.
Phillipson, Robert
1992 Linguistic Imperialism. Oxford: Oxford University Press.
Potter, Jonathan
1996 Representing Reality: Discourse, Rhetoric and Social Construction.
London: Sage.
Preisler, Bent
1999 English in a European EFL Country. In Standard English: The Wid-
ening Debate, Tony Bex and Richard J. Watts (eds.), 239–267. Lon-
don: Routledge.
Prodromou, Luke
1992 What culture? Which culture? ELTJ 46 (1): 39–50.
1996 Correspondence. ELTJ 50 (1): 87.
1998 Correspondence. ELTJ 52 (3): 266.
Sperber, Dan and Deidre Wilson
1995 Relevance: Communication and Cognition. Second edition. Oxford:
Blackwell.
Taylor ,Talbot J.
1997 Theorizing Language: Analysis, Normativity, Rhetoric, History.
Oxford: Pergamon.
Toolan, Michael
1994 On recyclings and irony. In Literature and the New Interdisciplinar-
ity, Roger D. Sell and Peter. Verdonk (eds.), 79–92. Amsterdam:
Rodopi.
Widdowson, Henry G.
1997 The use of grammar, the grammar of use. Functions of Language 4
(2): 145–168.
Chapter 12
Towards a new English as a Foreign Language
curriculum for Continental Europe
Urs Dürmüller
1. Introduction
English today is no longer the same subject it was fifty years ago. The in-
ternationalization of English has dramatically changed especially what in
English as a Foreign Language (EFL) studies could be perceived as just
one language, one culture and one literature. English does not belong ex-
clusively to its native speakers any more (cf. Gnutzmann and Intemann
2005). There is an awareness of the difference between English as a first
language (EL1), and English as a second (EL2) or foreign language (EFL),
and of English as a Language of Wider Communication (lingua franca,
ELF) that is not necessarily identical with EL1 prestige varieties. In addi-
tion, English is finally seen as a language symbolizing globalization and
westernization. EFL Teachers begin to mind the ideological load of the
dominant EL1 varieties and to think about redefining separate standards for
ELF and EFL, independent from the standards set in EL1 education, and
with special regard to the goals set for language competence, access to
culture, and knowledge about literature.
This means that the English studies curricula, in countries where Eng-
lish is not a first language, have to be redesigned. The new curriculum for
future teachers of English at secondary school level at the PHBern Teacher
Training College in Bern, Switzerland, does not only illustrate openness
towards English as a global auxiliary language and the various English
language cultures, but also shows respect for the continuing needs of stu-
dents to become familiar with prestige varieties, even in the age of Global
English (cf. Graddol 2006). In what follows I am going to discuss the ways
the status of English as an international language, its many varieties and
the cultural and ideological loads carried by these challenge the traditional
understanding of English in EFL curriculum development in outer circle
nations, particularly in Continental Europe, and I am going to show how
such discussions can produce a new orientation for EFL teacher training.
240 Urs Dürmüller
2. Re-Defining the subject of English as a Foreign Language (EFL)
Discussing the subject of English as a Foreign Language makes it neces-

sary to look at English from the outside, rather than from the inside, as has
been customary in the past. Not the perspective taken by L1 speakers of
English – e.g., the native speakers of English in Britain or in the USA –
should be taken, but that of those speakers who are adding English to their
personal language repertoires. The question is not simply “What is Eng-
lish?” any more, but “What is English to its users in Switzerland or Bul-
garia, or any other country where English is a foreign language?”
This new perspective has been encouraged since Kachru (1985,
1986a/b) presented his idealized model for the spread of English, in which
he distinguished between three types of areas affected by the expansion of
English. These are now called, respectively, the inner circle (EL1 nations),
the outer circle (EL2 nations) and the expanding circle (EFL nations). In
this paper I am not concerned with the subject of English in a nation where
it is a national and/or official language, but with English as a subject of
study and language education in one of the many nations where it is simply
an additional and foreign language (EFL). When doing so, I will rely on
my first-hand knowledge of English in Continental Europe, especially of
English in Switzerland (Dürmüller 1992, 2003), and on my experience of
teaching English to students at the PHBern Teacher Training College, Bern
University of Applied Sciences (Pädagogische Hochschule Bern).
For the reasons mentioned above, outside the core-English (the EL1)
nations the subject of English is in need of redefinition. Many more stu-
dents than ever before are willing and eager to take up the subject today,
especially to learn a language that has become an international means of
communication. While in the past it was taken for granted that these learn-
ers of EFL would model their English after EL1 speakers, now it is not so
easy anymore to say what exactly they want or need to learn.
Questions like the following are being asked:
– What is English today, especially to non-native speakers?

– Who does English belong to?
– Which of the many varieties of English should be the target of second
language acquisition?
– Should EFL learners be encouraged to imitate an L1 accent?
– Should identification with a certain L1 variety be a goal of EFL learn-
ing?
Towards a new EFL curriculum for Continental Europe 241
– Thus, what is going to be the standard for English instruction? Could

that be a variety that has no native speakers?
– Which of the numerous authors writing in English should students read
and study?
– Which parts of English-speaking culture, civilization, lifestyle should
they know something about?
The learning and teaching of English as a foreign language has generally

been based on one of the two dominant EL1 varieties, Standard British or
Standard American. There may have been discussions on whether Ameri-
can or British English was to serve as the teaching model, but other ques-
tions, e.g., relating to alternatives to the supremacy of UK and US English
in EL2 acquisition, have hardly ever been asked.
Such questions, however, cannot be ignored any longer. The voices of
English today are many, and the values associated with the various varie-
ties are quite diverse. Educators have to become aware of what it means
today to choose a certain variety as their standard for EFL teaching and to
actually speak it. They must understand that the globalization of English
has put an end to the uncritical use of EL1 varieties in EFL contexts.
Outside English-speaking countries, it is becoming more and more ob-
vious that it is not only British or American citizens that use English, but
people from all over the world, particularly also people from and in nations
where English is not a national language. English today is used much more
by people for whom it is a second or foreign language added to their lan-
guage repertoire than by people who speak it as a first language.
Nobody really knows how many speakers of English there are. Crystal
(2002) suggests that 400 million people speak English as a first language
and that another 400 million use English as a second language, while the
number for speakers of English as a foreign language might even reach a
billion. If today there are many more speakers of English as a second and
foreign language – with a ratio of (at least) 2 to 1 in favor of non-native
speakers –, that is the result of the spread of English, which is due to vari-
ous factors, political, military, economic (Crystal 2002), and cultural (Dür-
müller 1986, 1996).
The expansion of English from the inner to the outer and the expanding
circle has been accompanied by a move of the center of English from the
UK to the USA. In the 19th century and the early half of the 20th century,
Britain was one of the world’s leading industrial and trading nations, the
biggest colonial power and easily the most important military power of the
world. But after World War Two this leadership role was taken over by the
242 Urs Dürmüller
USA. This has helped English to ensure its position as a world language in
spite of the fall of the British Empire. For the EFL curriculum today this
means that a global definition of English has to include US English not
only next to British English, but before British English. To people affected
by the spread of English in the expanding circle and partly also in the outer
circle, American English has become more important than British English.
Indeed the factors encouraging the spread of English in the second half
of the 20th century were mainly US-related, namely: the military and eco-
nomical power of the USA, the affiliation in large parts of the world with
much of US culture, a culture with enormous popular appeal, the concen-
tration of technological and general scientific advances in US corporations
and universities, and the development of the new communication technolo-
gies like wireless, telephony, telegraphy and internet (cf. Crystal 2002;
Graddol 1997). English became a very powerful language, attracting new
speakers everywhere – and has continued to do so. (For the parameters of
the power of English see Kachru [1986a: 130].)
In spite of the obvious differences between the status of English in the
Inner Circle on the one hand and in the Outer and Expanding Circles on the
other hand, the subject of “English” has traditionally been associated with
native-speaker English and culture everywhere, and schools have based
their courses on the prestige varieties of the core-English nations, or,
rather, on what the English (British) textbook industry – itself based in
Inner Circle nations – has offered them as the standard for English lan-
guage teaching. Even where in the Expanding Circle national educational
authorities had their own English language materials developed (as, e.g., in
Switzerland), the L1 models were closely followed. However, the fact that
there are so many non-native speakers of English now that use English to
interact with other non-native speakers of English casts some doubts on
programs for EFL education in Expanding Circle nations that are based on
Inner Circle curricula. What do learners of EFL really need to know about
English? If they are studying English as a lingua franca, do they need to
know anything about the culture, say, of England, or the lifestyle of
Americans? If they focus on English as a language for international com-
munication, is it necessary for them to concentrate on the two dominant
varieties, i.e., American and British English? Or can the target of EFL be
some other variety of English, even one that does not exist among L1
speakers?
Indeed: Most students of English in the Expanding Circle do not take up
the subject because they want to learn about the British or the Americans,
but because they want to become fluent in the English language. They ex-
pect that they will be using English as a language for wider communica-
tion, both intranationally and internationally. If they are using English for
communication with other non-native speakers, their medium will be some
form of International English, and that is not necessarily one of the varie-
ties used as a national language in core-English speaking countries. To
learner-speakers of EFL English is not “English” in the restricted sense of
relating to Anglo-Americans, their culture and language, but just a tool for
communication between people whose linguistic and cultural backgrounds
can be quite different from that of EL1 speakers. Reference to Anglo-
American culture is often irrelevant and in some instances even undesired.
It is true that in the past EFL education used to take as its target the na-
tional standards of Britain or the USA as they appeared in the teaching
materials produced in these EL1 countries. In Continental Europe there
was a clear preference for the type of British English promoted by the lan-
guage education industry of the UK. There was no doubt that the L1 stan-
dard varieties would also serve the EFL functions of wider communication
and of dealing with specific purposes (ESP). If in the Expanding Circle
everywhere the same “textbook variety” were used in English language
education, then, indeed, this would be the obvious language of wider com-
munication. However, where English is not L1, new oral-vernacular Eng-
lishes have come into being and now compete with “Textbook English”. I
am not just referring to the so-called new varieties of English, mostly
forms of English in the Outer Circle, but to the “impure” forms of English
that can be encountered in the Expanding Circle, also, of course, in Conti-
nental Europe. These are mixtures of American and British and possibly
other L1 Englishes, with additions of words and phrases imported from
various culture fields, such as trendy sports, fast food, discos, advertising,
pop music, comics, films, etc., all of which colored by the local languages
of the EFL speakers. In the profile of these new varieties, particularities
can be detected in pronunciation, spelling, lexicon, grammar, semantics
(word, phrase and text meanings), and in pragmatics which make them
differ, not only from each other, but also from the established standard
varieties.
For Continental Europe Barbara Seidlhofer (2004; cf. also Seidlhofer,
Breiteneder and Pitzl 2006) has discussed such characteristics which are
mainly simplifications of English based on contact-linguistic, i.e., interfer-
ence phenomena in second-language acquisition.
Together with Jennifer Jenkins, she is proposing a locally (contact-
linguistically) simplified English as a possible lingua franca in Europe
(Jenkins and Seidlhofer 2001; cf. also Jenkins 2007). The main maxim for
244 Urs Dürmüller
that kind of EFL would be: Teach only what is crucial for communication!
This might mean allowing /th/ to be replaced by either /s/ or /z/ or /d/ or /t/,
getting rid of the redundant {-s} as verb ending in the present, considering
the relative pronouns who, which, that as interchangeable, doing without
the correlation of the present perfect with for/since, or even no longer in-
sisting on the use of articles (Jenkins and Seidlhofer 2001). It would also
mean okaying words and phrases that are unique to European EFL varie-
ties, and doing without British or American idioms.
Such a form of English would certainly no longer be grounded in Brit-
ish or American uses of English, but in contexts where English is used as a
foreign language, and it would mean recognizing the fact that English is no
longer the preserve of native speakers only. This kind of Continental Euro-
pean English would certainly be removed from the “classical” forms of
EFL and the standards set for language education in British and American
English. It might be seen as a truly international form of English, making
English into a lingua franca, as Seidlhofer and Jenkins claim, to be used
not among native speakers, nor between non-native speakers and native
speakers, but among non-native speakers first of all. It appears that a sim-
plified, streamlined variety, stripped down to the essentials for communica-
tion, might work as a lingua franca just as well as any EL1 variety used
among native speakers.
However, practitioners of EFL teaching and researchers in English sec-
ond language acquisition will know that English-target interlanguages vary
greatly according to their source languages. Interference-based errors are
not the same in the English produced by EFL learners in Bulgaria, in
Finland, Hungary, or in Italy. For a Bulgarian and a Finnish or a Hungarian
and an Italian EFL learner to understand each other might not be so easy if
each is making use of their own form of English. It is not surprising that
Jenkins and Seidlhofer were criticized when they popularized their ideas
and talked about “Bringing Europe’s lingua franca into the classroom” in
the Guardian Weekly (2001).
Mixing EFL learner varieties cannot produce the standard language to
be used in Continental European or in global EFL teaching. The interfer-
ences caused by the L1 of Polish, Portuguese, Greek or Latvian learners of
English are so different from each other that it will always be possible to
distinguish not only Polish, Greek or Latvian accents but also L1-specific
lexical and grammatical particularities in the EFL varieties of these Euro-
peans. What can be perceived as German English, Romanian English or
Slovenian English may work in their own L1 contexts, but communication
is bound to break down when the L1 boundaries are crossed. Multilingual
Switzerland can serve as an example: The English learnt as a foreign lan-

guage in German-speaking Switzerland is markedly different from that
produced by learners in French-speaking Switzerland and that heard in
Italian-speaking Switzerland. If the Swiss should ever choose to use Eng-
lish as a lingua franca intranationally, they would all have to use the same
variety, one that, in addition, would also be understood internationally (cf.
Dürmüller 2003).
If the wider communication function of EFL is considered, the aca-
demic construct of a common European English lingua franca cannot be an
option for language education in Continental Europe. Another reason
against settling for the simplified English as a lingua franca model, is the
fact that internationally EFL is very often used by non-native speakers in
the shape of English for Specific Purposes (ESP). ESP is chiefly learnt and
used not to indulge in social talk with native speakers but to get access to
various global communities of experts, e.g., scientists, engineers, medical
doctors, lawyers, bankers, air traffic personnel, business people, media
experts, advertisers, etc. ESP varieties of English that are used internation-
ally make it possible for people from all kinds of linguistic, cultural and
political backgrounds to communicate with each other in the special lan-
guage of that community, about topics that are the concern of them all.
What is needed here is not a simplified lingua franca, but a highly devel-
oped tool for communicating even specialized information with great detail
and high accuracy.
Does that mean Continental European language educators have to fall
back on the old standards for EFL and again follow the practice of the past
and use the standards set for EL1 education in Continental European EFL
teaching as well? If so, would we opt for British English or for American
English?
To answer that question, one should take into consideration that in-
creasingly the ESP jargons are grounded in American English. It is not
only in popular culture, but also in most academic and other professional
circles that US English has overtaken UK English and established itself as
the language of choice in many domains where ESP is used. While (Stan-
dard) British English is still present in international fora, American English
is now clearly dominant. Keeping that in mind, Continental Europe might
be well advised to prefer American English over British English and set
that as a standard for EFL education.
However, there is a problem with that as well. Language does not only
serve the purpose of neutral, subject-centered communication but also the
purpose of identification in interpersonal exchange across cultures where it
246 Urs Dürmüller
might become necessary to express one's identity as a member of a particu-

lar cultural community. This is obviously the case for Inner Circle speakers
of English. It is also the case where English has been localized, i.e., where
new varieties have arisen, as typically in Outer Circle nations, e.g., as Ni-
gerian English or Sri Lankan English. The special local forms of these
varieties, which are the result of long contact between English and non-
English local languages, can give a lot of information about their speakers
that can be interpreted culturally or socially.
On the other hand, Expanding Circle speakers of EFL, e.g., from Conti-
nental Europe, say from Switzerland, or Bulgaria, do not have such local-
ized or nativized varieties of English. If they use the Seidlhofer/Jenkins
ELF, which is a compromise of “Textbook English” with interference-
caused errors produced by EFL speakers having different L1s, they would
be perceived not as speaking a variety of European English, but a form of
English as a learner language. These “localized” forms of English are not
what people want to speak – even if many, inevitably, do speak an L1-
marked form of English. Italian English, French English, German English
are not accepted as targets of EL2 acquisition, as they are easily recognized
as the result of imperfect L2 learning. Nor are they accepted for the expres-
sion of European identities. Until quite recently the target of Continental
European EFL education tended to be a student emulation of an L1 pres-
tige variety, above all the standard of British L1 education, which would
mean imitation of the accent (RP) and use of the variety-specific vocabu-
lary and phraseology, including idiomatic expressions. Linguistically and
culturally the goal was identification with a certain L1 speaker group.
However, neither non-native speakers talking to each other nor non-
native speakers speaking to native speakers need to assume a native-
speaker identity. For, when using English as a lingua franca they remain
what they are: German, French, Russian, Chinese. Why should they emu-
late the speech habits, including accent and lexical idiomaticity, of an
American or a British person? Doing so might even send out undesired or
wrong signals. For, when looking for a standard for EFL in Continental
Europe and the Expanding Circle, simply opting for the dominant varieties
– American and/or British English – has become problematic. Apart from
the regional diversification of English, it is also the ideologies associated
with particular varieties that influence people’s choice of English. Atti-
tudes towards a language are important: i.e., what we think English will do
for us and what others think of us when we use that language. The recent
US and British invasion of Iraq has made this quite obvious again. Speak-
ing like the members of the US administration may be worth imitating to
some, but others might want to distance themselves from them (cf. Graddol
2006: 112). International English acquired as a foreign language in the
Expanding Circle ought to be a neutral tool rather than an ideological pro-
tagonist or an ambassador of a particular culture.
A new definition for EFL in the Expanding Circle seems to demand a
degree of critical distance towards the dominant presence of American and
British based forms of English worldwide. Traditionally, quasi native-
speaker proficiency has been a target for EFL courses in the Expanding
Circle areas. Students would not only be encouraged to imitate prestige
accents, use the idiomatic expressions of elite L1 communities, but also to
try and integrate themselves with these cultures and societies. Learning
English thus has meant accepting the dominant Anglo-American mindset
and supporting what Kachru (1986b) has described as “the power” of Eng-
lish. Now the ongoing internationalization of English is giving EFL stu-
dents an opportunity to learn English without having to become advocates
of Anglo-Americanism at the same time.
Even if they are not going to take any of the new nativized varieties (say
Kenyan English) or their own learner languages (the Seidlhofer/Jenkins
English as a lingua franca) as their target for EL2 acquisition, but stick to
the educational standards of British and/or American English, they will not
have to concentrate on culture elements that are specifically British or
American, but can try to familiarize themselves with as many cultures that
find expression in and through English today as possible or desirable.
Using English as an international language means communicating with
speakers from various and differing backgrounds. The majority of ELF
interactions worldwide take place between speakers for none of whom
English is the mother tongue and for none of whom English is a cultural
symbol. For example, if a Swiss person conducts business in China, Eng-
lish is likely to be used. English language cultural elements will most
probably not enter the conversation at all (cf. House 1999: 84). What EFL
speakers need today is not information about traditional Britishness and
Americanness, but the teaching of transcultural politeness strategies that
are part of the pragmatics of International English.
In order for students to acquire some knowledge about the identities and
the social and cultural lives of speakers of International English, the litera-
tures (and other culture products) from the Outer and Expanding Circle
must be given more and more weight. While international communication
in English may still be served best if L2 acquisition continues to focus on
standardized forms of UK and/or US English, the accompanying teaching
of English language culture does not have to focus on Britain and the USA
248 Urs Dürmüller
exclusively, not even principally. Reading English in EFL classes still

means reading Inner Circle authors, but no longer exclusively so. In the
chorus of World English, their voices may be the loudest, but they may no
longer indicate the rhythm and melody of the English language or express
the central interests and concerns of those who speak it.
Actually, what American and British English symbolize today, may be
quite contradictory. We all know that English has spread as a result of ex-
ploitation and colonization. But Anglo-American English has also become
a language of support and liberation. English represents the success of a
certain economic model, but also the egoism of capitalism, it stands for
human rights and charity, but also for the perversion of democracy and
Christianity (cf. Bailey 1991: 165).
Thus, it is possible for many learners of EFL to both love and loathe
that language, something which, on the level of specific cultural interest, is
quite obvious. While young learners of English, for example, may radically
distance themselves from the discourse of the present US and UK political
leadership, they may want to identify as much as possible with the rap of
certain music culture stars.
3. Effects on the EFL curriculum
How does all this affect the subject of English as a foreign language? How
can all these considerations be integrated into an EFL curriculum? In the
remainder of this essay I am outlining how the new study plan for English
at the PHBern Teacher Training College, in Bern, Switzerland, reflects
what I have said about the changes the internationalization of English
might bring about in the definition of “English” in Expanding Circle nation
EFL education, particularly in Continental Europe.
In EFL Teacher Training the educational focus must center on high lan-
guage competence in a form of English that can serve both, the lingua
franca function and the need for detail and accuracy in ESP. The English
defined as target should be instrumentally neutral. This might mean creat-
ing a compromise variety of British and American English as the basis for
EFL education, something suggested by Modiano already in 1998. His
Mid-Atlantic English does not make it necessary for EFL educators to
really set up a new hybrid variety of English, but encourages them to be
open towards American and British English at the same time, thus not forc-
ing EFL learners to choose one or the other. Mid-Atlantic English mirrors
the situation of Continental Europeans who are exposed to both, British
and American English, at the same time all the time. It makes it possible
for those learning EFL to keep a certain distance from the values associ-
ated with both British and American English, the ideologies carried by
them, the economic models propagated through them, and the lifestyle
advertised by them, while still using an English that will be recognized as
prestigious even if it does not belong to any group of native speakers.
The English Modules now taught at the PHBern Teacher Training Col-
lege first give students a survey of what the English language is today (Ta-
ble 1). English Today deals with the spread and globalization of English,
the claims for English made by the speakers of English as a first language
and those made by speakers of English as a second and foreign language, it
presents the “old” varieties of English in the Inner Circle and the many
new varieties of English in the Outer and Expanding Circles: it discusses
the cultural and ideological load of the various Englishes, particularly of
the “prestige” varieties of Standard British (RP) and Standard American
(GA), and it considers the lingua franca function of international English,
for which it suggests an English that is based on the educational standard
varieties of both American and British English.
Table 1. Modules of the PHBern English Study plan (from the 2006/7 curriculum
for secondary school teachers)
Language Competence C1/C2: C1
C2
Language Skills: Oral Skills
Writing Skills
English Today
Introduction to English Linguistics
Introduction to Literature in English
English Second Language Acquisition
School Literature
Culture, Language and Literature: Culture Topics
Selected Literature
Language, Culture and Society
With all that information at hand, prospective teachers of English at secon-

dary school level in Switzerland, when making a choice for a certain vari-
ety as their standard in the classroom, can be expected to know why they
250 Urs Dürmüller
are choosing one rather than another. If they use British or American text-
books, as most of them might have to, they are encouraged to approach
these materials critically, and to keep in mind that they are teaching EFL
and not EL1.
In EL2 Acquisition, apart from discussing theories of L2 acquisition and
how these could be applied to the acquisition of English as a Second or
Foreign Language, the choice of a standard – for pronunciation, vocabulary
and grammar above all – is discussed in further detail, as is the inevitability
and acceptability of interlanguage forms. Communication among non-
native speakers is contrasted to communication among native speakers.
Culture Topics may examine characteristics of communities and life-
styles in the English-speaking parts of Asia, Africa, Europe, Australia and
America. English-language culture is not restricted to learning about tradi-
tional British or American features like cricket, high tea and semi-detached
houses or high school graduation, baseball and big breakfasts. While Brit-
ish and American cultures are, of course, not excluded, other cultures are
included on purpose. These reflect the internationalization of English: as-
pects of modern life in Africa and Asia that can no longer do without Eng-
lish, and the dependence of various transnational communities – like those
of business people, researchers, scientists, indeed all kinds of professionals
– on English and the ensuing forms of ELF and ESP.
Language, Culture and Society takes a closer look at the factors that
have promoted the spread of English in continental Europe; it tries to de-
fine the place English occupies within the language repertoire of multilin-
gual and multicultural Europe or a multilingual nation like Switzerland; it
investigates the changes English is undergoing when used by speakers of
other languages. And it asks questions like: How are varieties of Oral Ver-
nacular English created and who makes use of them, when and where?
How do these affect the international standard varieties of English? Local
advertisements and graffiti can be examined as examples of particular text
types that integrate English language elements, while Black slang may be
studied as a source of borrowings into local (European) languages and
learner EFL.
Regarding literature, the idea of going through the classical canon of
English literature is entirely given up. There is an introduction to what
literature is and how it can be approached and studied (Introduction to
Literature in English), as there is a general introduction to language and
linguistics (Introduction to English Linguistics). But passing on the Eng-
lish literary heritage is understood as a task for schools in countries where
English is a national language. Selected Literature focuses on literature in
English, written by authors from all corners of the world, selected for its
relevance with regard to content and form. It is a course about texts that are
hoped to find the interest of students and to strengthen their attachment to
reading literature in general. Of course, such a course will not exclude
British or American authors. But it will not prefer them over Indian, Egyp-
tian, Nigerian, Caribbean, or, indeed, Swiss (e.g., Alan de Botton) authors
writing in English. In School Literature the prospective EFL teachers learn
about what texts are suited to the needs and expectations of adolescents,
both from the point of view of contents and linguistic complexity. The
process of text simplification is of greater importance than the question
whether a story is grounded in Anglo-Saxon culture.
While all these modules reflect the awareness that English today is dif-
ferent from what it was fifty years ago, the language skills modules, Oral
Skills and Writing Skills, and Language Competence (C1/C2 Common
European Framework of Reference, CEFR) swing back to traditional main-
stream English. Students need to be able to give an effective oral presenta-
tion and to write an academic paper or a CV, and they need to be able to do
so in a form of English that is developed, has international recognition and
will not be mistaken for a restricted learner language exhibiting L1 inter-
ference phenomena jeopardizing successful communication. That is why
Language Competence is a course preparing them for language certificates
that have wide recognition and promise material benefits. C2, the highest
level of the CEFR, must be attained before the students are certified as
English teachers. Many of them will make use of the training received as a
preparation for the corresponding Cambridge (Proficiency) or TOEFL
exam.
4. Conclusion
The internationalization of English is challenging not only the hegemony

of Standard American and British English, but also the way English as a
foreign or second language is taught in the Expanding Circle. For EFL
teacher training in Continental Europe native-like proficiency based on
what is considered the standard in Inner Circle nations – especially Britain
and the USA – and integration within the dominant L1 speaker communi-
ties cannot be the main goal any longer. Instead, English must be perceived
as a truly international language, increasingly expressive of many and dif-
fering cultures, literatures and lifestyles, not necessarily congruent with the
Anglo-American worldview, even if the lingua franca function of English
252 Urs Dürmüller
is guaranteed best if the target for EFL acquisition is set as an approxima-

tion of British and/or American English.
This means that with regard to culture and literature there has to be a
truly global orientation, and that the linguistics of EFL in grammar, spell-
ings, vocabulary, and also in pronunciation should favor a compromise
model of what has been called Standard American-British English or Mid-
Atlantic English (Modiano 1998). Like the problematic English as a Lin-
gua Franca model (Jenkins and Seidlhofer 2001), this can guarantee criti-
cal distance to English language imperialism, and the accompanying impo-
sition of social, economic and cultural values, while at the same time
giving access to an English developed enough for special purposes and
recognized everywhere to ensure the lingua franca function.
References
Bailey, Richard W.
1991 Images of English. A Cultural History of the Language. Ann Arbor:
University of Michigan Press.
Crystal, David
2002 The English Language (2nd edition). London: Penguin.
Dürmüller, Urs
1986 The status of English in multilingual Switzerland. Bulletin CILA 44:
7–38.
1996 Englisch auf dem Vormarsch. Schweizer Schule 6 (96): 3–9.
1992 The changing status of English in Switzerland. In Status Change of
Languages, Ulrich Ammon and Marlies Hellinger (eds.), 355–370.
Berlin/New York: de Gruyter.
2003 English in Switzerland: From foreign language to Lingua Franca. In
Perspectives on English as a World Language, David Allerton (ed.),
115–123. Basel: Schwabe.
Graddol, David
1997 The Future of English. London: The British Council.
2006 English Next. London: The British Council.
Gnutzmann Claus and Frauke Intemann (eds.)
2005 The Globalisation of English and the English Language Classroom.
Tübingen: Narr.
House, Juliane
1999 Misunderstanding in intercultural communication: Interactions in
English as lingua franca and the myth of mutual intelligibility. In
Teaching and Learning English as a Global Language, Claus Gnutz-

mann (ed.), 73–89. Tübingen, Stauffenburg.
Jenkins, Jennifer
2005 Implementing an international approach to English pronunciation:
The role of teacher attitudes and identity. TESOL Quarterly 39: 535–
543.
2007 English as a Lingua Franca: Attitudes and Identity. Oxford: Oxford
University Press.
Jenkins, Jennifer and Barbara Seidlhofer
2001 Bringing Europe’s lingua franca into the classroom. The Guardian
Weekly, 18th April.
Kachru, Braj
Learning the Language and Literatures, Randolph Quirk and H. G.
Widdowson (eds.), 11–30. Cambridge: Cambridge University Press.
1986a The Alchemy of English. The Spread, Functions and Models of Non-
Native Englishes. Oxford: Pergamon Press Ltd.
1986b The power and politics of English. World Englishes 5 (2/3): 121–
140.
[n.d.] The New Englishes.
Modiano, Marko
1998 The emergence of Mid-Atlantic English in the European Union. In
The Major Varieties of English: Papers from MAVEN 97, Hans
Lindquist, Staffan Klintborg, Magnus Levin, and Maria Estling
(eds.), 241–248. Vaxjo: Acta Wexionensia.
2000 Euro-English: Educational standards in a cross-cultural context. The
European English Messenger IX (1): 33–37.
Seidlhofer, Barbara
2000 Mind the gap: English as a mother tongue vs English as a lingua
franca. Vienna English Working Papers 9 (1): 51–68.
2004 Research perspectives on teaching English as a lingua franca. Annual
Review of Applied Linguistics 24: 209–239.
Seidlhofer, Barbara, Angelika Breiteneder and Marie-Louise Pitzl
2006 English as a lingua franca in Europe: Challenges for applied linguis-
tics. Annual Review of Applied Linguistics 26: 3-43.
Chapter 13
Language learning and medium of instruction:
Understanding contemporary discourses and
practices in Swiss schools and classrooms
Daniel Stotz
1. Introduction: An impetus for immersion
In the late 1980s, a number of linguists, practitioners and politicians began

pushing for more intensive forms of foreign language education in Swiss
state schools. This movement arose in parallel to an insufficiently coordi-
nated reform to introduce a first foreign language (in fact, the second na-
tional language, French or German) to the primary school curriculum.1 One
of the events that enlivened the debate on desirable forms of foreign lan-
guage teaching (FLT) was an international symposium organised by a
group of linguists and pedagogues around Richard J. Watts at the Univer-
sity of Berne in 1988. The aim of the gathering was not only to discuss
novel forms of intensive language learning in and through schools, but also
to develop ways of enhancing mutual understanding across language bor-
ders in Switzerland, in other words to counter the risk that Switzerland
“falls apart into monolingual regions between which communication was
barely possible any longer”2 (Andres and Watts 1990: 9).
The symposium participants were invited to search for models of
schooling that could provide better results with respect to the acquisition of
a second national language and, at the same time, to civic integration. De-
spite some doubts about the link between school language learning and
mutual understanding in a multilingual country as well as about the poten-
tial of educational institutions to facilitate language learning, there was a
good measure of optimism: the implementation of French or German in the
primary school curriculum was seen as an intermediate goal, but genuine
improvements in overall competence would best be accomplished with
immersive forms of language learning. This would mean that subjects such
as history, geography or mathematics could and should be taught and con-
ducted in the language to be acquired. Hence the name of the project that
256 Daniel Stotz
ensued from the symposium: “Unterrichtssprache Französisch/Deutsch”,

which can be translated as “French or German as a medium of instruction”.
This chapter traces the pathways of attempted and completed reforms in
the second/foreign language domain over the last decade and provides
some empirical classroom data in order to discuss issues along the fault
lines between discourses on societal multilingualism and classroom reali-
ties in a context characterised by multiethnic plurilinguality. The article
questions the notion of standard, in particular with respect to the diglossic
situation in Swiss schools, and points to the additional sociolinguistic
complexities that the standard/dialect dichotomy introduces. It also prob-
lematises simplistic models of form-oriented foreign language learning and
the ideology of the monolingual use of the target language in FLT. In brief,
the argument is that while immersion-style provision of language classes
may look like a persuasive solution to the school system’s task of promot-
ing students’ multilingual competence, it comes with the inherent contra-
diction that it attempts to introduce, as a medium of instruction, standard
forms of language in an environment characterised by various forms of
non-standard use such as dialect, learner language and hybrid mixing of
different varieties. In processes of gatekeeping, e.g. giving or withholding
access to valued resources such as vocational education and good jobs, the
current language ideology privileges some varieties over others (Standard
German over dialect) and some languages over others (English over
French). Rather than engineering classroom interaction by decreeing a
particular medium of instruction, schools could raise awareness of how
languages are used meaningfully and provide more opportunities for nego-
tiating topics which are both educationally relevant and linked to the stu-
dents’ social worlds and identities.
The research presented here stems from a larger-scale project on multi-
lingualism, identities and school language learning3 (Stotz 2005), which
attempts to discover, with the help of ethnographically based sociolinguis-
tics, the contribution that language learning can make to the construction of
identities among children and adolescents and, hence, to suggest ways in
which schools could promote a kind of multilingualism that supports young
people in “mastering their lives” (Andres and Watts 1990: 9).
The following section looks at the educational and political goals of in-
tensive language learning through immersion and analyses the contexts for
such programmes in German-speaking Switzerland. The situation is made
complex by the phenomenon of diglossia, where dialects and Standard
German are used in different domains and under intriguing conditions of
power and access. The third section provides a bottom-up perspective of
Language learning and medium of instruction 257
language education in one school community in the Canton of Zurich, Stät-

ten,4 which has undergone significant changes. The event described is an
information forum for adolescents and parents in this multiethnic suburb of
Zurich. An analysis of representations and uses of standard and dialect
varieties of German helps to focus on the question of linguistic resources
and their value at the transition between school and occupation. In the
fourth section, evidence from language classrooms is used to make a case
for a close emic perspective on the parallel development of skills in four
languages or varieties (dialect, Standard German, French and English). It
asks how ways of communicating and referring to subject content shape
the opportunities that learners have to develop the desired skills. The con-
cluding section retraces the recent history of language education reform
and, on the background of the evidence discussed, asserts the need to offer
adolescents more binding ways of engaging with content issues and move
away from a simplistic focus on form-oriented language teaching.
2. Language as a learning objective and as a medium of instruction
In the late 1980s, majority immersion programmes had the allure of a

magic potion for FLT. They were exploited by second language acquisition
theorists as a “natural approach” which could do away with many tenets of
“instructed” language learning (Krashen and Terrell 1983; Krashen 1985;
Krashen 1998). Canadian models of French immersion were cited in the
Berne symposium report as case studies in the “transmission of meaningful
content and subject matter in a foreign language” (Stotz 1990: 37) with
only a minor focus on the target language as a system. The programmes
usually led to native-like receptive competences while the ability to pro-
duce and interact in French remained more limited or even fossilised
(Harley et al. 1990). In her symposium contribution, Monica Heller shed
some light on the sociolinguistic context of the French immersion wave:
The programmes provided “access to authentic communication situations
in the foreign language” (1990: 108) thanks to native or bilingual teachers
of the target language, but, she concludes:
Malgré l’immersion, l’école n’est finalement pas en mesure d’apporter une
contribution décisive à l’abolition des barrières linguistiques. (…) Tout
l’environnement social des élèves reste anglophone, ce qui signifie que,
pour des contacts avec des francophones, leur compétence sociolinguistique
n’est tout de même pas assez développée. (Heller 1990: 108)5
258 Daniel Stotz
To judge from the success of the Canadian programmes, it appears that

they fulfil a desirable socio-economic function in that they allow middle-
class anglophone students to broaden their employability in a declaredly
bilingual state. There is a sociolinguistic paradox in this: much as in Swit-
zerland, the principle of territoriality locks the national languages in their
places, while at the same time urban communities are becoming more and
more multiethnic and multilingual as a result of old and new forms of mi-
gration. While nationally oriented bilingualism is lived by few, societal
multilingualism is often treated as a problem to be overcome by unilateral
integration on the part of immigrants.
If one wanted to transfer the Canadian immersion model to European
contexts, a sociolinguistic perspective that includes considerations of
power and difference would be called for. Thus, a paradox of the Canadian
experience was that Anglophone majority students who had achieved
higher competence in French started competing for posts reserved for bi-
lingual speakers. Among the French-speaking minority this may have led to
a sense of “la francophonie menacée” (Heller 1990: 109). In Switzerland,
the franco- and italophone minorities are under greater economic pressure
to learn German than the German speakers are to acquire French or Italian
(Grin 1999). However, the minorities are confronted with societal diglossia
in the German-speaking part of Switzerland, which may frustrate their ef-
forts to learn Standard German (Watts 2001). While French Canadians
who learn English gain a world language and a desirable form of bilingual-
ism (at least in some functions, cf. Heller 2004), the francophone Swiss
with school-learnt German may not even be able to cope with the ‘Swiss
model of multilingualism’ according to which everyone speaks their
mother tongue and is capable of understanding their conversational part-
ners. The reluctance of many dialect-speaking Swiss to use Standard Ger-
man in public and private encounters with speakers of other languages
exacerbates the situation (Watts 1999).
Any discussion of educational language politics in Switzerland must go
beyond the “topos” (Hägi and Scharloth 2005) that “Standard German is a
foreign language for the pupils” and seriously analyse the roles of dialects
and standard. It is on the basis of the underlying duality of dialect and stan-
dard that Swiss German pupils learn foreign languages such as French and
English. Watts reflected on this from a vantage point of contrastive analy-
sis already in the early 1980s (Watts 1981) and later from a critical socio-
linguistic point of view (Watts 1997, 1999, 2001).
The German-Swiss linguistic situation has been characterised in two
ways. It is sometimes said to be a type of medial diglossia (Kolde 1981;
Werlen 1998), where written communication is mainly conducted in Stan-

dard German while speakers use their local dialect in informal and even
formal situations. Other researchers describe it as a case of outright socie-
tal bilingualism (Ris 1990: 42). Werlen distinguishes between receptive
and productive uses of Standard versus Swiss German and notes that
speakers are often engaged in “asymmetric bilingualism” (Werlen 1998:
26), where Standard German is more often heard than spoken, e.g. as a
result of widespread consumption of TV channels broadcasting from Ger-
many and Austria. However, written forms of dialect are increasingly used
in electronic media of communication such as text messaging and Internet
chat.6 Even though there is now some consensus that the hypothesis of
medial diglossia cannot be upheld due to the widespread use of both varie-
ties in different functions and social situations, we must stress the link
between social difference and the ability to speak Standard German flu-
ently. Werlen points out that “spoken Standard German represents an inte-
gral part of occupational life, albeit for only about a third of the working
population” (1998: 28). Census data corroborates the suspicion that the
standard is used more frequently by academics and professionals (Lüdi and
Werlen 2005: 47). While mastery of Standard German may not be the most
important key to successful careers, it is certainly a concomitant feature
and an imperative stressed both by university professors with respect to
their students and by employers in local communities speaking to secon-
dary school leavers (see Section 3 below). Gate-keepers for desirable posi-
tions often use the written standard as a device for controlling access (e.g.
letters of application and CVs).
Significantly for the present discussion, Standard German cuts across
the spoken/written medial divide at school, the most important arena for
spoken use of Standard German in most Swiss German biographies. In
official terms, exposure to and interaction in the standard variety at pri-
mary school is seen as the key ingredient for the acquisition of Standard
German. For instance, the Zurich curriculum stresses the role of school as
an opportunity for pupils to “build up a broad repertoire of spoken registers
in High German [sic] and to refine it in various configurations of speaking”
(Bildungsrat des Kantons Zürich 2005: 105). The educational goal is thus
one of language acquisition through the use of a range of registers in the
context of schooling with the help of specific arrangements for implicit and
explicit learning, but in the absence of native speakers of the standard vari-
ety. Traditionally, there has always been a partial focus on the grammar
and lexis of the written standard through instruction. Contemporary ap-
proaches such as Hägi and Scharloth’s stress the need to teach “oral com-
260 Daniel Stotz
munication abilities in the standard more purposefully and consciously”

(2005: 16). This mixed process of implicit acquisition and instructed learn-
ing of Standard German is expected to unfold in conjunction with subject
knowledge learning and the acquisition of social skills. In other words,
what is theoretically postulated is nothing less than genuine immersion,
albeit not with a foreign language as a medium of instruction, but the stan-
dard variety, which is not used regularly in the everyday routines of the
local community.
In reality, the easy acquisition of Standard German through schooling is
hampered by a variety of factors:
– Many teachers tend to use a characteristically Swiss subvariety of Stan-

dard German, which is often described as showing a lack of typical fea-
tures of orality and fluency (cf. Sieber and Sitta 1986; Bachmann and
Ospelt 2004).
– Interaction in educational contexts is characterised by asymmetric par-
ticipation structures, giving the pupils less share in front-stage oral lan-
guage production and interaction.
– Young pupils often start school with high idiomatic competence in the
standard, acquired through media exposure and active use in peer role-
play (Häcki Buhofer et al. 1994; Watts 1997), but tend to adapt their
phonology and delivery to a less native-like version used by teachers.
– There is less emphasis on the acquisition of presenting, debating and
other rhetorical skills compared to, for example, the Anglo-Saxon edu-
cational tradition.
In recent years, educational authorities and public discourses have increas-

ingly emphasised the use of Standard German at primary and secondary
schools as a result of perceived mediocrity of PISA test results and an at-
tempt to provide better educational conditions for children with a back-
ground of migration. Brochures and projects have been created to lend
support to a less uptight approach to Standard German. Paradoxically,
however, the school system sets itself the goal of rectifying a situation for
which it has itself been responsible. As dialect use is spreading to new
niches and as new subcultural varieties such as “Balkans German” (possi-
bly a crossing phenomenon, Rampton 1995) are emerging, the discourse of
educators and politicians on Standard German is often fraught with notions
of language decay (“Sprachzerfall”) and the emergence of despised hybrid
language forms. This suggests that Standard German as a medium of in-
struction keeps supporting the reproduction of socially selective ways of
schooling. The following two sections underpin this suspicion with ethno-
graphically gained evidence.
3. Communicating language values: A public forum on the transition

between school and work
Stätten is a town of 13,000 inhabitants on the outskirts of Zurich, charac-

terised by a tradition of industry and a multiethnic present; it counts among
the less desirable residential areas, but is socially too differentiated to be
called a low-income ghetto. Social geographers have described it as part of
an “edge city” which offers spaces for the development of new service
industries and retailing (Leuthold 2005).
A key event to observe and describe linkages between school language
learning and use of languages in occupational life is an information and
discussion forum organised by the parents’ council of the local secondary
school. The evening, to which 8th graders and their parents have been in-
vited, is about the challenges and opportunities of the search for appren-
ticeship contracts, the main avenue into employment from lower secondary
school. In the following, some salient goings-on are reported and discussed
with respect to the ways in which language values are communicated.
The president of the parents’ council introduces the evening in Zurich
dialect, reading out from notes. A young teacher briefly welcomes the par-
ticipants in Zurich dialect. A professional moderator leads the discussion,
speaking Standard German, using notes occasionally. After having given
an overview of the evening, she engages a former student in a semi-
spontaneous interview, asking questions in Standard German, while the
apprentice, a young female florist, replies in Zurich dialect. None of these
choices of varieties and switches between them are explicitly justified or
commented on. It is obvious to residents of a multiethnic town such as
Stätten that a fair share of parents will not be able to understand the dia-
lect. The school itself normally emphasises the need to use the standard,
not only for the benefit of families with a background of migration but also
to support Swiss dialect speakers in the acquisition of the standard. The
invited speakers on the podium, who represent various occupational do-
mains from hairdressing through the hospitality to the construction indus-
try, seem unsure which variety to use. The first speaker replies in Standard
German, but the representative for a training centre for public transport
announces that she will speak Swiss German and then quickly asks if there
is anyone who does not understand her. However, the moderator intervenes
262 Daniel Stotz
and asks her to use Standard German without giving a reason, possibly in
order to save the face of those listeners who do not want to call attention to
their own lack of competence in the local dialect. In the rest of the panel
presentation and discussion, the speakers use a variety of Swiss High Ger-
man which is clearly enunciated, delivered slowly but fluently with little
hesitation. On the evidence of this public event, the claim that Standard
German is used like a foreign language by adult Swiss speakers (e.g.
Berthele 2004) cannot be maintained. But it also becomes clear that speak-
ing it requires an effort and contextualises the speech event as official and
portentous.
When asked which kinds of knowledge and abilities are crucial for a
successful application for an apprenticeship, some panellists stress the
importance of German as a school subject, or, as one speaker says, “learn-
ing German”. The moderator repeats and emphasises this conspicuously
(“Deutsch ist wichtig, ich möchte es nur noch sagen, dass man’s bis ganz
hinten hört”).7 In the question period, half the students use Standard Ger-
man, the others dialect. One student questioner quotes his German teacher
as saying that a punctuation error, e.g. a missing comma, in the letter of
application would lead to the letter being thrown out and wants to know
from the employers if that is correct. The responding restaurant owner
rejects this but emphasises that some of the applications are written in
“really bad German” and quotes the worst as running: “Sie Chef, ich gerne
Koch wolle. Du mir habe Stelle?”,8 to the mirth of the audience. The quote
plays on an immigrant hybrid caricature of German rather than a dialect-
based version.
Written Standard German is thus presented as an important gate-passing
skill, with an emphasis on formal aspects of the language. All of these stu-
dents have been learning foreign languages for many years (French for 5
years, English for at least 3 years), but skills in and uses of these languages
are never touched upon during the whole evening. They do not seem to
play a crucial role in the transition from secondary school into the world of
work. This does not mean that languages other than German do not surface
in the daily lives at school as carriers of meaning potential, disenfran-
chisement or symbolic power. On the contrary, the multilinguality of the
community itself is a constant undertow that pulls at the resources and
beliefs of the students and teachers.
The following section provides insights into the situated practices of the
language classroom in the same community, involving those same 8th-
graders whose occupational futures were delineated in the forum.
4. Approaches to language teaching: Three languages in one class
We turn to social and linguistic practices in the language classrooms to ask

what opportunities the students of this mixed suburban school community
have to develop the language skills which will allow them to find their
desired trajectories into occupational training and at the same time to de-
velop ways of expressing themselves. The issues which the analysis deals
with are the following:
– How are languages and their varieties contextualised as linguistic sys-

tems and ways of communicating in the classroom?
– How are topics introduced and negotiated in the language classroom?
Do they serve as ‘empty’ carrier content or are they integrated into a
broader agenda for learning, including other subjects and social skills?
– What role does the teacher and his/her interaction management play in
the conduct of the lesson and what may be the consequences for the stu-
dents and their learning opportunities?
Here I will focus on three lessons which were audio- and video-recorded
with a view to approaching these questions empirically. The class of 19
students is in the B track of secondary school, in other words, third in a
hierarchy of four academic tracks. Six of the students name German or
Swiss German as their first language, the other mother tongues are Alba-
nian, Arabic, Bosnian, Croatian, Italian, Portuguese, Serbian, Spanish and
Turkish. All except for two students were born in Switzerland, but only
eight have Swiss citizenship. Each of the three languages, German, French
and English, is taught by a different teacher. One reason for choosing these
lessons as a basis for the analysis is the striking impression that the stu-
dents’ behaviour and interaction in the lessons differ greatly as a result of
the teacher and his/her style and interaction strategy. The German lesson is
taught by the class teacher, an experienced man shortly before retirement,
while the English class is given by a middle-aged female and the French
class by a young female teacher. The observers9 noted very different levels
of noise, discipline and concentration in the three settings.
The German lesson is somewhat of an exception as the teacher, whom I
shall call Mr Götschi, has decided to offer a special lesson that he has
taught before. The focus of the lesson is a transformation exercise for
which the students have to transfer sentences from written Zurich dialect to
Standard German. The topic is exclusively linguistic, as Mr Götschi an-
nounces: the lesson will deal with the word ‘wo’ (where), which, according
264 Daniel Stotz
to him, is used “in the Swiss German language ... not only locally, but also
temporally or as a relative pronoun”.10
Mr Götschi asks the learners to transform 22 dialect sentences on a
worksheet into “korrektes Hochdeutsch” (correct High German). An illus-
trative exchange is given below.11
(1) GÖT ah probiert doch einfach (.) Artan
just give it a try Artan
ART (reads) höch uf de B- äh uf d Bäum wo d Vögel
high on the trees where the birds
sitzed cha kchei Chatz chlättere (..) höch auf den
are sitting no cat can climb high on the
Bäumen wo die Vogel wo die Vögel sitzen kann keine
trees where the birds are sitting no
Katze klettern (looks up)
cat can climb
GÖT ja (.) kannst du’s noch einmal ganz deutlich sagen
yes can you say it again very clearly
ART ähm höch
high (Swiss German phoneme)
GÖT hoch
high (Standard German phoneme)
ART äh hoch (.) hoch auf den Bäumen wo die Vögel sitzen
high on the trees where the birds are sitting
(.) kann keine Katze klettern
no cat can climb
GÖT äh hört ihr den Fehler (.) hoch auf den Bäumen kann
do you hear the mistake high on the trees
keine Katze kle- klettern (.) das ist nicht ganz
no cat can climb that is not quite
korrektes Deutsch (..) hoch auf den Bäumen kann
correct German high on the trees
keine Katze klettern
no cat can climb
MAR hoch auf DIE Bäume
high onto THE trees
GÖT guet (.) sagst du den Satz ganz
good can you say the whole sentence
(…)
MAR hoch auf die Bäume wo die Vögel sitzen kann keine
high onto the trees where the birds are sitting no
Katze kle[ttern
cat can climb
GÖT [ jawohl (.) guet

yes good
Artan, a 15-year old male student of Albanian extraction, transfers the

Swiss German umlaut in ‘höch’ to Standard German. The teacher corrects
this mistake and then recasts the sentence for the learners to find the sec-
ond mistake, a problem of case marking. Marco, a monolingual student,
provides a correction and is asked to repeat the whole sentence. The rou-
tine feedback move by the teacher is Swiss German ‘guet’ for ‘good’. The
group of learners are frequently disciplined to provide complete sentences,
they are mildly criticised if they deviate from the standard, and only once
is there an oblique hint by the teacher as to the questionable nature of the
example sentences when he comments on a form as “nöd ganz guets
Züritüütsch” (“not quite good Zurich German”).
Briefly summarised, the lesson is conducted like a foreign language
class, with a focus on form but little in the way of metalinguistic explana-
tion. Given the quaint meanings encoded by the sentences, we may wonder
why the class, which can be unruly and noisy in other lessons, plays along
well and why the general atmosphere is sober but friendly. On the one
hand, the teacher commands respect and does not tolerate deviance, on the
other, the lesson focuses on correct written Standard German, a resource
that keeps being pointed out to the students as invaluable on the career
path. The use of ‘wo’ as a non-local relative pronoun in Standard German
is more than a grammar mistake – it can be read as a signal for incomplete
mastery of the standard and may expose the speaker or writer more se-
verely than a comma mistake. The difficulties that bilingual students have
reading out written dialect and translating it into the standard may further
intimidate them into compliance. A year before their graduation from sec-
ondary school, they are given a demonstration of how awkwardly they use
both dialect and standard and how complicated these valuable resources
are.
The English and the French lesson follow a similarly form-oriented
agenda. As in the German lesson, the English teacher, Ms Keller, controls
the management of the interaction in the whole-class phase and, in doing
so, attempts to ensure that the target language is used as the medium of
communication and that complete sentences are normally produced. The
teacher initiates the sequence by asking the students what they know about
Picasso; the learners had to read a biographical text on the painter in their
course book. However, the topic is addressed not for its own sake, but with
a grammatical focus, as Ms Keller made clear in the previous lesson: “Pi-
266 Daniel Stotz
casso is dead, that’s the reason why we must use past form.” The famous
painter is treated as mere carrier content, and the reproduction of a painting
in the book is only briefly considered. However, as opposed to the German
lesson, there is a factual background to the sample sentences, and the fol-
lowing exchange is noteworthy as one student, Tabea, corrects a statement
of the teacher’s (Ms Keller, KEL).
(2) ART he was born in eighty eighty one
(…)
KEL I have to look up – eighteen [ ninety one yeah
ART [ ninety one
KEL okay (.) more (..) yes
[(two girls, Tabea and Nathalia, are talking to each other, looking
at the teacher critically)
SXX [he and his family nei he and his father was painters
KEL ye:s [(.) hey-ey girls (eye-gaze direction to Tabea and Nathalia)
DAM [ (raises hand) ääh
KEL was a painter he said
(..)
KEL yes (to Tabea)
TAB she is the eighty EIghty one
KEL eighteen ninety one – or eighteen eighteen eighty one
[ yes and he mo::ved to Vigo in ninety one yes (.)
SXX [ xxx
(KEL) you know better than I do
Tabea and Nathalia are admonished by the teacher for talking back-stage,
and when Tabea gets her turn to speak, she corrects the painter’s year of
birth (1881 rather than 1891), although her use of English makes her hard
to understand. The teacher acknowledges her correction and comments
appreciatively. In a second phase, not shown here, the students have to ask
questions about Picasso, which are supposed to trigger the reply: “Picasso
did”. The content focus of the course book then moves away from Picasso
to other famous dead people. Fatlum, another Albanian-origin student,
provides a question based on a prompt in the book, a picture of Alexander
Bell, believing that Bell invented the electric light bulb:
(3) FAT (points to his book) who invented the electric lights
(..)
KEL ääh (sound of rejection)
(…)
FAT who did
KEL no no no (..) who invented that’s okay

(writes on blackboard)
FAT it’s correct (..) the electric light (looks at KEL)
KEL is it correct?
MAR (XXX)
KEL Alexander Bell invented the electric light?
(Marco and Sevda raise hands)
KEL oh no I’m sorry no he didn’t invent the electric li::ght
SEV the telephone
KEL yes (.) the whole sentence please
This sequence highlights a typical conundrum of FLT, the oscillation be-

tween a metalinguistic focus on form and an epistemic focus on content
matter. The frame (Goffman 1974) for much of the lesson has been a focus
on form, but Ms Keller’s reaction to Tabea’s initiative has shifted the
frame slightly. Her self-correction and evaluative comment “you know
better than I do” can be seen to contextualise factual correctness as indis-
pensable. Hence the teacher is now constrained to insist on Alexander Bell
as the inventor of the telephone and not the electric light. Fatlum has trou-
ble understanding the import of her rejection and starts to correct the syn-
tactic form of his contribution (“who did –“) rather than the content.
What happens here could be called a re-keying along the lines of Goff-
man’s frame analysis, where ‘key’ refers to a set of conventions by which a
given activity (which is already meaningful in terms of a primary frame-
work) is transformed into something patterned on this activity but seen by
the participants to be something else (Goffman 1974: 43–44). In a form-
oriented foreign language lesson, exemplars are usually treated in isolation
from their meaning potential. In a communicatively oriented lesson, there
are often two layers or laminations at play, the semi-authentic communica-
tion, where samples of language are used to emulate real-life exchanges,
and a concentration on language forms as well as their pragmatic and so-
ciolinguistic uses. In a genuine immersion situation, however, the primary
frame would be that of subject knowledge acquisition through whatever
configurations the players in the classrooms constitute as relevant and ap-
propriate (Nikula 2005). Opportunities to use the target language – and
hence improving competence in it – will arise as a function of learners’
involvement, the social organisation of tasks and the exposure to the lan-
guage that the lesson and the materials offer.
A precondition for an immersive process is the widespread if not exclu-
sive use of the target language in the classroom. Interestingly, foreign lan-
guage pedagogy has long insisted on this principle even outside properly
268 Daniel Stotz
immersive classes. Communicative language teaching methodology aims to

increase the opportunities for use and teachers adhering to it will often be
adamant about it, as is Ms Keller when she insists on the use of English
even in pairwork activity. Despite her insistence on the English-only prin-
ciple, other languages are obviously used: Swiss German in group work
interaction and Standard German to frame sentence patterns (e.g. “who
marschierte first on the moon?”). Occasionally, in this multilingual class,
another language does slip in on the front stage as when a student with
Lebanese background pronounces ‘travelled’ as “travailed” and Fabio uses
the Italian acronym “tivu” for TV. In both cases, the teacher good-
humouredly acknowledges the slip but again insists “on ne parle pas fran-
çais maintenant” and “tivu would be Italian, Fabio”.
In sum then, the teacher’s strategy of conducting the formally-oriented
lesson is a rallying cry for concentrating on linguistically and factually
correct use of the target variety with less emphasis on phonology than on
syntax. The pragmatic hybridity of the situation is evident in a number of
keyings and re-keyings, and thus there is more of a negotiation going on
than in the German lesson, including even the distribution of speaking
rights, with two cases where learners initiate corrections. The borders be-
tween back-stage and front-stage activity are more permeable than in the
German lesson (cf. also Heller 2001). The teacher still manages to maintain
orderliness through her insistence on the English-only practice and the
narrow bandwidth of topical issues, dead artists and inventors whose main
role is to allow practising the past simple.
There is not enough space here to describe and exemplify the French
lesson. It takes place with the same class, but a different teacher. Discipli-
nary matters and open resistance determine long stretches of the interac-
tion. The focus is once again on formal aspects of the language system, and
there is a listening text, entitled “l’ambitieux” about a fictional ambitious
pupil who does everything perfectly, a set piece of a display text which
serves to drill a grammatical feature. It is perhaps ironical that its content
refers to the exact opposite of what the learners are doing. They work their
way obediently through a set of exercises, but at the same time they resist
any opportunity to use French with respect to a content topic or even to the
management of the situation at hand. Suffice it to point out that these
glimpses of real life in the classroom are a long way away from the goal of
immersion education and a kind of language learning which would pro-
mote mutual understanding across Swiss language borders.
This is somewhat surprising and potentially disappointing since the
town of Stätten has taken part in a 5-year project run by the Zurich De-
partment of Education, which aimed to test new forms of language teach-

ing methodology in an approach termed Content and Language Integrated
Learning. A number of the students in this class took part in this project in
their primary school years. The project was part of a re-orientation of edu-
cational language policy in Switzerland, which is tied, on the one hand, to
the discourses on multilingualism and, on the other, to foreign language
teaching methodology. The next section traces the developments that have
led to a general shift in the priorities of foreign language teaching, both in
the dimensions of language sequence and methodological concepts. We
will see that the globalisation turn has become associated with immersion-
type approaches, involving synergies but also contradictions.
5. Trajectories of change in educational language policy in

Switzerland and their local implications
What has happened in the domain of foreign language learning in Swiss

state schools since the Bern symposium mentioned at the beginning of this
chapter? The story has been told and critiqued elsewhere by this author
(Stotz 2002, 2006a/b; Stotz and Meuter 2003), and the discourses on multi-
lingual Switzerland versus the “globalisation of English” were the focus of
a critical analysis by Richard Watts (2001). Although the process of
change has not come to an end yet, I will briefly summarise the develop-
ments, casting them against the background of a statement Watts made in
1997:
As long as Switzerland exists in its present form, English will never be de-
clared the first foreign language in preference to an LN2 [a second national
language]. On the other hand, if the demand for English increases signifi-
cantly – and I stress that there is no real evidence as yet to make this as-
sumption – then it will be a strong indication that a great deal is wrong with
language teaching and learning in the Swiss education system(s). (Watts
1997: 298)
It is a matter of debate what exactly may be wrong with FLT in Switzer-
land. In fact, the trajectory of changes over the last decade did lead to an
advance of English, ending in a situation where 14 out of 17 German-
speaking cantons are now implementing or will introduce English as the
first foreign language.12 During the 1990s, the pressure from parents, poli-
ticians and business representatives to devote more time and resources to
teaching English first resulted in English being installed as an obligatory
270 Daniel Stotz
subject at the secondary school level (age 13 and up) in several cantons.
The process of globalisation and the techno-economic boom before the
year 2000 contributed to a dynamic which, in a populist reform discourse,
somehow got entangled with psycholinguistic theories of early second lan-
guage acquisition (Stotz 2002). While in the previously dominant discourse
on multilingualism, English was construed as a threat to linguistic harmony
(Watts 2001), the most influential language policy document, the Compre-
hensive Languages report (EDK Expertengruppe 1998) is, in Watts’ inter-
pretation, an “attempt to play down the importance of English throughout
all areas and in all social domains in Switzerland, whilst at the same time
recognising its global importance” (Watts 2001: 313). The notion that Eng-
lish would be used as an intra-national lingua franca or link language was
not prominently discussed; some would say it remained a taboo (Kleinber-
ger Günther 2003), a spectre (Watts and Andres 1993), or lacked “empiri-
cal evidence on which to base such conclusions” (Watts 2001: 312).
By the turn of the century, the advocates of early English were dominat-
ing the discursive space with notions such as high motivation, naturalist
approaches and English as the future language of international communica-
tion and, hence, advancement. Most importantly, they started implementing
English as a first additional language in the primary school and thus occu-
pied the field for experimenting with novel methodological approaches.
While in the first canton to start out with a generalised primary English
programme in 2001, Appenzell Innerrhoden, the method adopted was a
story-based communicative approach, in Zurich as well as cantons in cen-
tral Switzerland, English classes are now conducted on the basis of a model
derived from partial immersion, termed Content and Language Integrated
Learning (CLIL, cf. Marsh and Langé 2000; Nikula 1997; Wolff 1997). It
would not be exaggerating to state that English as the language of global-
isation has hijacked the model of immersion-type approaches to the teach-
ing and learning of foreign languages in Switzerland. While experiments
with bilingual French and German classes have been carried out with suc-
cess in communes along the language border (Brohy and Bregy 1998; Le
Pape Racine 2005), they exist in precarious conditions and are rarely trans-
ferred to wider regions. In contrast, English is more often associated with
successful language learning even in the face of critical evaluation reports
such as Büeler et al. (2001) and Stebler and Stotz (2004). The compromise
decision reached by the federal coordinating body EDK/CDIP13 in 2001 to
allow regional conglomerates of cantons to determine the first additional
language (English or the second national language) has exacerbated the
tension since the language strife (“Sprachenstreit”) now appears to breach
unity even between closely related cantons such as Basel Land and Aar-
gau.14
In sum, educational language policy in Switzerland has been dominated
by a rift in the discursive space on multilingualism and has led to a head-
over-heels implementation of primary school English in many cantons and
a patchwork of differing solutions depending on regional and local con-
cerns. In other words, the checkerboard map of Swiss educational language
policy shows traces of both globalising and localising tendencies that could
be characterised with the hybrid term ‘glocalisation’. The assessment that a
great deal must be wrong in Swiss FLT if English were to become the first
additional language to be taught can be turned on its head then: politicians,
parents and practitioners have ambitiously invested in this reform, which
piggybacks on English as a language of prestige and economic value.
French, for the most part of German-speaking Switzerland remains on the
curriculum of primary schools, and optimists hope for better results from
earlier exposure to a popular language. The expectations are high indeed,
and the question arises if the promised rewards in terms of power and ac-
cess can be gained with reformed ways of learning two languages side by
side.
On the basis of the sociolinguistic ethnography carried out in the project

and exemplified in this chapter with insights into a public forum and three
language lessons, the argument unfolds in the following way. The notion of
using the target language as a medium of instruction is valuable (a) if it is
properly translated into the diglossic situation of Swiss German schools,
(b) if it recognises the limits of immersion-type language programmes, and
(c) if it takes into account the reasons learners may or may not have for
investing into language learning. In the event, competence in various regis-
ters of Standard German seems to be a desirable resource, valorised by
teachers, parents and future employers, and a central part of legitimate
repertoires. All learners, in particular those who speak another language at
home, are expected to benefit from the practice of using Standard German
as the main medium of instruction. Indeed, the school board of Stätten has
decided to implement just such a policy: since 2006, all kindergarten
teachers have been using Standard German (rather than Swiss German) as
a matter of course when interacting in the classroom as well as in the
school grounds. They have taken courses promoting a more natural and
272 Daniel Stotz
spontaneous use of the Swiss variety of Standard German. Although some

groups of parents and local politicians oppose the decision vehemently, the
overall echo is positive. With such a move towards using the Standard
German consistently in and around the classroom, it would be odd if les-
sons such as the German class discussed above were still to be held in fu-
ture, where the standard is taught like a foreign language according to the
grammar-translation method.
However, the inherent contradiction remains that in order to teach the
curriculum, a variety is used which is not part and parcel of every-day com-
munication. Thus, when the moderator of the Stätten forum says German is
important what is meant is compliance with a written standard at a crucial
gate-keeping moment where the criteria for evaluation are far from trans-
parent. In the best case, the focus of using Standard German as a medium
of instruction should not exclusively be on language in use; a systematic
raising of awareness and of language reflection should be built into the
curriculum. In this way, the functional and social distribution of dialects
and standard can itself become a resource for heightened sensitivity to
language issues. As the exchanges at the apprenticeship forum show,
value-laden choices of varieties are often associated with insecurity and
face threats. Students who have learnt to negotiate these choices with an
awareness of their consequences will be better able to understand proc-
esses in which the distribution of power and access is at stake. For in-
stance, they would not be led to believe by their teacher that a punctuation
error in a letter of application is the end of their career prospects. They
might even speak out against offensive caricatures of job applicants with
purported lack of competence in German. And they would master the use
of relative pronouns in written German earlier while still being able to
speak idiomatic dialect and standard varieties, mixing in whatever hybrid
forms they need in order to express their variegated identities.
However, what are we to make of the reform process which has
changed the landscape of foreign language teaching over the last decade?
Clearly, English, promoted as a resource in a globalising marketplace, has
had a field day, but the compromise solution that the standing conference
of directors of education (EDK/CDIP) is advocating may yet survive. The
goal would be for all students at the end of obligatory schooling to be able
to use a second national language and English at similar levels of compe-
tence.15 Observations in a classroom where the reform has not really taken
a foothold yet are obviously not sufficient to judge whether such objectives
are realistic. However, our glimpse at current practices through the lens of
ethnographic sociolinguistics suggests a number of issues to be taken up:
– While a form-oriented teaching methodology may be successful in dis-

ciplining learners into social-interactional patterns preferred by the
teacher and the school as an institution, it misses out on opportunities
for use of the language in the authentic context of learning events. It
creates more problems of contradictory keyings than it solves in terms
of language acquisition.
– The ideology of using the target language only militates against actual
practices and may not always be helpful. A more systematic acknowl-
edgement and management of languages and varieties having different
functions in the classroom is called for. Immersion with its blanket
identification of the foreign / second language as a medium of learning
glosses over these challenges.
– Episodes involving learner initiative and multiple language use suggest
that students could bring more to classroom learning than filling in the
gap in the classic three-part Initiation-Response-Feedback exchanges.
Resistance to the teaching agenda may also result from too low de-
mands, especially with groups of learners categorised as academically
less able.
– The evidence from these lessons suggests that ‘content’ is not con-
structed as a relevant issue in foreign language (and even German) les-
sons; the propositions that are expressed do generally not count as
knowledge to be acquired or as expressions of self-identity, but as carri-
ers of forms and lexis.
In all of the four points, foreign language teaching practices as observed

here lead to an impasse: not only do the students lack opportunities for
meaningful use of the new language, there is also a shortage of occasions
where they can invest themselves with respect to who they are and who
they want to be as speakers of various languages and varieties. In other
words, the resources they are about to acquire are only tenuously linked to
their identities and the broader educational agenda which should help the
students gain access to valued resources such as apprenticeship contracts.
This is mainly a consequence of the ways in which course book writers
conceive their texts, exercises and listening pieces, and the ways in which
teachers use the materials in class. While the painter Picasso may not be
the most interesting topic for 14-year olds to deal with in the first place, the
conspiracy with which course book and teacher refrain from engaging with
the artistic, educational and social significance of Picasso’s life and work
is conspicuous. He remains just that: a dead white male.
274 Daniel Stotz
The analysis of a public speech event and three language lessons com-
prising the same group of students suggests that if learners are to develop
desirable resources to which value is assigned, they need to be given op-
portunities to engage with topics and speech activities that are relevant
within their social worlds and tied to a broader educational agenda. This
need is best captured by the notion of investment, developed by Norton
(e.g. Norton 2000), which is in turn based on the economic metaphors
Bourdieu uses in his work. For Norton (2000: 10), the term ‘investment’
refers to the “socially and historically constructed relationship of learners
to the target language, and their often ambivalent desire to learn and prac-
tice it”. Young learners are more likely to invest in a second language if
they stand to “acquire a wider range of symbolic and material resources,
which will in turn increase the value of cultural capital” (2000: 10). In this
vein, an investment in the target language is also “an investment in a
learner’s own identity” (Norton 2000: 11).
In the years between childhood and adulthood, the trajectory of identity
formation is constantly accompanied by language learning efforts which
are situated in social practices of the classroom and its environments. A
higher degree of awareness among teachers, parents and students of the
ways in which languages are used in educational and occupational domains
could contribute to more informed choices in terms of teaching methodol-
ogy and materials design. Something would be wrong with language teach-
ing and learning in Switzerland indeed, not if English were the first lan-
guage to be taught, but if language classes in general mainly used an
orientation on language forms and structures to discipline students into
complying with practices that pursue selection and gate-keeping in connec-
tion with Standard German as a central task of secondary schools. Rather,
foreign language teaching and choices in the medium of classroom interac-
tion should provide for learners to be able to engage in discourses as mean-
ingful ways of using linguistic and other resources, starting from the word
go. This objective constitutes a departure from the “more play-oriented
pedagogy” advocated by Watts (1997: 300) towards an approach which
integrates language use with the construction of ‘content’, or rather, dis-
course experiences within an overall epistemic agenda.
Transcription conventions
KEL speaker acronym

(laughs) non-verbal behaviour
(.) (..) (…) short pause (½ second), medium pause, long pause (>2 sec)
li::ght lenghthened syllable
[ onset of overlapping speech (first speaker)
[ (second speaker)
SXX unidentifiable student speaking
(XXX) incomprehensible passage
Notes
1. German was introduced as the second national language, taught like a foreign
language, in the French-speaking part of Switzerland; in the German- and Ital-
ian-speaking parts French was phased in in similar fashion with the exception
of the Canton of Uri, which chose Italian. In the trilingual Canton of Grisons a
specific solution was sought involving bilingual schooling in Romansh and
German as well as Italian.
2. All citations in German and French have been translated by this author, DS.
3. The 30-month project “Multilingualism, identities and language learning in
Swiss schools and communities” is funded by the Swiss National Science
Foundation as part of its programme 56 “Language variety and language com-
petence in Switzerland” (cf. Stotz 2005).
4. All names referring to places and people with whom research was conducted
have been changed. Prior consent has been sought for the data to be used in
publications under these conditions.
5. Translation: Despite immersion, schools are not capable in the end of contrib-
uting decisively to the abolishing of linguistic barriers. (...) The social envi-
ronment of students remains anglophone, which means that for contacts with
francophones, their sociolinguistic competence is still not developed enough.
6. For an insightful discussion of the diglossia versus bilingualism question cf.
Hägi and Scharloth (2005).
7. Translation: “German is important, I’d just like to say it again, so that you can
hear it as far as the back (of the hall).”
8. A transliteration of this pidginised form of German would run as follows:
“You boss, me want become cook. You have job for me?”
9. The lessons are part of a larger corpus of 48 lessons gathered in two schools as
part of the project “Multilingualism, identities and language learning”, see
endnote 3. There were always two observers present in the lessons, including
the author and a research assistant.
10. Using the relative pronoun ‘wo’ with a non-local reference in Standard Ger-
man can be seen as an incorrect transfer from Swiss German. A constructed
example would be: Swiss German: das isch de Maa, wo mir die Gschicht ver-
zellt hät. Standard German: *Das ist der Mann, wo mir die Geschichte erzählt
276 Daniel Stotz
hat. Correct: Das ist der Mann, der mir die Geschichte erzählt hat. (Transla-
tion: This is the man who told me the story).
11. The transcription conventions are given at the back of this chapter.
12. The bi- or trilingual cantons of Bern, Fribourg, Valais and Grisons are not
counted in this tally. They opted for the intracantonal second language first.
13. The standing conference of cantonal ministers of education who coordinate
educational policy in the absence of a national ministry of education. The con-
stitution assigns education to the 26 cantons.
14. In addition, groups of members of the federal parliament now aim to re-assert
the power of the centre by writing into a languages law a provision that the
first additional language to be taught at schools must be French or German.
15. In Zurich, Zug, Thurgau and Schaffhausen there have been votes about initia-
tives restricting foreign language teaching to one language in the primary
school. The voters rejected these initiatives in all these cantons, albeit with
narrow margins.
References
Andres, Franz, and Richard Watts

1990 Das Projekt "Unterrichtssprache Französisch/Deutsch" – eine Ein-
führung. In Zweisprachig durch die Schule – Le bilinguisme à tra-
vers l'école, Richard J. Watts and Franz Andres (eds.), 9–14. Bern:
Haupt.
Bachmann, Thomas and Barbara Ospelt
2004 Hochdeutsch als Unterrichtssprache: Die Sprechpraxis von Studier-
enden und Lehrpersonen. Project report. Pädagogische Hochschule
Zürich.
Berthele, Raphael
2004 Vor lauter Linguisten die Sprache nicht mehr sehen. Diglossie und
Ideologie in der deutschsprachigen Schweiz. In Dialekt, Regiolekt
und Standardsprache im sozialen und zeitlichen Raum, Helen Chris-
ten (ed.), 111–136. Wien: Praesens.
Bildungsrat des Kantons Zürich
2005 Lehrplan Sprachen. Überarbeitete Fassung vom 7.2.2005.
Brohy, Claudine and Anne-Lore Bregy
1998 Mehrsprachige und plurikulturelle Schulmodelle in der Schweiz
oder: What's in a name? Bulletin suisse de linguistique appliquée 67:
85–99.
Büeler, Xaver, Rita Stebler, Georg Stöckli and Daniel Stotz

2001 "Schulprojekt 21". Lernen für das 21. Jahrhundert? Externe wissen-
schaftliche Evaluation. Schlussbericht zuhanden der Bildungsdirek-
tion des Kantons Zürich. http://www.vsa.zh.ch
EDK Expertengruppe
1998 Welche Sprachen sollen die Schülerinnen und Schüler während der
obligatorischen Schulzeit lernen? Bericht der von der Kommission
für allgemeine Bildung eingesetzten Expertengruppe ‘Gesamt-
sprachenkonzept’. Bern: EDK/CDIP.
http://www.romsem.unibas.ch/sprachenkonzept/Konzept_kurz.html
Goffman, Erving
1974 Frame Analysis. Harmondsworth: Penguin.
Grin, François
1999 Compétences et recompenses. La valeur des langues en Suisse. Fri-
bourg: Editions universitaires.
Häcki Buhofer, Annelies, Harald Burger, Hansjakob Schneider and Thomas Studer
1994 Früher Hochspracherwerb in der Deutschen Schweiz: Der weitge-
hend ungesteuerte Erwerb durch sechs- bis achtjährige Deutsch-
schweizer Kinder. In Spracherwerb im Spannungsfeld von Dialekt
und Hochsprache, Harald Burger and Annelies Häcki Buhofer (eds.),
147–198. Bern: Lang.
Hägi, Sara and Joachim Scharloth
2005 Ist Standarddeutsch für Deutschschweizer eine Fremdsprache? Un-
tersuchungen zu einem Topos des sprachreflexiven Diskurses. Lin-
guistik Online 24: 3.
www.linguistik-online.de/24_05/haegiScharloth.html
Harley, Brigit, Patrick Allen, Jim Cummins and Merrill Swain (eds.)
1990 The Development of Second Language Proficiency. Cambridge:
Heller, Monica
1990 L'immersion et les rapports entre anglophone et francophones au
Canada. In Zweisprachig durch die Schule – Le bilinguisme à travers
l'école, Richard Watts and Franz Andres (eds.), 107–109. Bern:
Haupt.
2001 Undoing the macro/micro dichotomy: Ideology and categorisation in
a linguistic minority school. In Sociolinguistics and Social Theory,
Nikolas Coupland, Srikant Sarangi and Christopher Candlin (eds.),
212–234. Harlow: Pearson.
2004 Paradoxes of language in the new economy. Babylonia 2: 29–31.
Kleinberger Günther, Ulla
2003 Kommunikation in Betrieben. Wirtschaftslinguistische Aspekte
innerbetrieblicher Kommunikation. Bern: Peter Lang.
278 Daniel Stotz
Kolde, Gottfried
1981 Sprachkontakte in gemischtsprachigen Städten. Vergleichende Un-
tersuchungen über Voraussetzungen und Formen sprachlicher In-
teraktion verschiedensprachiger Jugendlicher in den Schweizer
Städten Biel/Bienne und Fribourg/Freiburg i. Ue. Beihefte 37: Zeit-
schrift für Dialektologie und Linguistik. Wiesbaden.
Krashen, Stephen
1985 The Input Hypothesis: Issues and Implications. Harlow: Longman.
1998 Foreign Language Education the Easy Way. Burlingame: Language
Education Associates.
Krashen, Stephen and Tracy Terrell
1983 The Natural Approach: Language Acquisition in the Classroom.
Hayward, CA: Alemany Press.
Le Pape Racine, Christine
2005 Erfahrungen mit frühem immersivem Unterricht: Didaktik im Grenz-
bereich von immersivem zu traditionellem Fremdsprachenunterricht.
In Gesamtsprachencurriculum – Integrierte Sprachendidaktik –
Common Curriculum, Britta Hufeisen and Madeline Lutjeharms
(eds.), 75–87. Tübingen: Narr.
Leuthold, Heinrich
2005 [Stätten] – ein Quartier der Limmattalstadt. Stadtentwicklung in der
Agglomeration. Themenheft 21. Brugg: Metron.
Lüdi, Georges and Iwar Werlen
2005 Sprachenlandschaft in der Schweiz. Neuchâtel: Bundesamt für Sta-
tistik.
Marsh, David and Gisella Langé
2000 Using languages to learn and learning to use languages. In Using
Languages to Learn and Learning to Use Languages, David Marsh
and Giselle Langé (eds.), 1–14. Jyväskylá, Finland: UniCOM, Uni-
versity of Jyväskylä.
Nikula, Tarja
1997 Terminological considerations in teaching content through a foreign
language. In Aspects of Implementing Plurilingual Education, David
Marsh, Bruce Marsland and Tarja Nikula (eds.), 5–8. Jyväskylä: The
University of Jyväskylä.
2005 English as an object and tool of study in classrooms: Interactional
effects and pragmatic implications. Linguistics and Education 16 (1):
27–58.
Norton, Bonny
2000 Identity and Language Learning. Harlow: Pearson.
Rampton, Ben
1995 Crossing. Language and Ethnicity among Adolescents. London &
New York: Longman.
Ris, Roland
1990 Dialektologie zwischen Linguistik und Sozialpsychologie. In La
Suisse face à ses langues – Die Schweiz im Spiegel ihrer Sprachen,
Jean-Pierre Vouga (ed.), 40–49. Aarau: Sauerländer.
Sieber, Peter and Horst Sitta
1986 Mundart und Standardsprache als Problem der Schule. Aarau &
Frankfurt: Sauerländer.
Stebler, Rita and Daniel Stotz
2004 Themenorientierter Sachunterricht in Englisch. Forschungsbericht
zu Handen der Bildungsdirektion ZH. http://www.vsa.zh.ch
Stotz, Daniel
1990 Übersicht über die Immersionsprogramme in Kanada. In Zweispra-
chig durch die Schule – Le bilinguisme à travers l'école, Richard
Watts and Franz Andres (eds.), 37–45. Bern: Haupt.
2002 Englisch in Zürich – no such thing as a free lunch. Babylonia 1: 32–
37.
2005 Multilingualism, identities and language learning in Swiss classrooms
and communities. Project proposal submitted to Swiss National Sci-
ence Foundation. http://www.nfp56.ch
2006a Breaching the peace: Struggles around multilingualism in Switzer-
land. Language Policy 5 (3): 247–265.
2006b Die Dynamik einer Reform: soziolinguistische Implikationen des
Zürcher Englisch-Experiments. Bulletin VALS/ASLA 83, 147–158.
Stotz, Daniel and Tessa Meuter
2003 Embedded English: Integrating content and language learning in a
Swiss primary school project. Bulletin VALS/ASLA 77, 83–101.
Watts, Richard
1981 Swiss German in English language teaching: A plea for the dialect.
Bulletin CILA 34: 71–80.
1997 Language policies and education in Switzerland. In Cultural Democ-
racy and Ethnic Pluralism, Richard J. Watts and Jerzy Smolicz
(eds.), 271–302. Frankfurt: Lang.
1999 The ideology of dialect in Switzerland. In Language Ideological
Debates, Jan Blommaert (ed.), 67–103. Berlin: Mouton de Gruyter.
Theory, Nikolas Coupland, Srikant Sarangi and Christopher Candlin
(eds.), 297–320. Harlow: Pearson.
Watts, Richard and Franz Andres
1993 English as a lingua franca in Switzerland: Myth or reality? Bulletin
CILA 58: 109–127.
280 Daniel Stotz
Werlen, Iwar
1998 Mediale Diglossie oder asymmetrische Zweisprachigkeit? Mundart
und Hochsprache in der deutschen Schweiz. Babylonia 1: 22–35.
Wolff, Dieter
1997 Content-based bilingual education or using foreign languages as
working languages in the classroom. In Aspects of Implementing
Plurilingual Education, David Marsh, Bruce Marsland and Tarja Ni-
kula (eds.), 51–64. Jyväskylä: The University of Jyväskylä.
Chapter 14
Can academic writing style be taught?
Jürg Strässler
1. Introduction
Whatever subject students study at university level, they will sooner or

later be faced with the task of having to write academic papers, either in
their own mother tongue or in a foreign language. In the second half of the
20th century, English has increasingly become the lingua franca for aca-
demic writing and other languages have been neglected more and more,
even in journals that are open to non-English submissions, for instance
Multilingua. When it started in 1982 under the auspices of the Commission
of the European Communities, Sager (1982: 7–8) stated in the editorial that
the policy of Multilingua was “based on the recognition that linguistic
diversity is a positive value which enriches the world and contributes to its
full cultural and intellectual development” and that “[t]he journal will pub-
lish contributions in all European Community languages, with major em-
phasis on English, French, German and Italian. Articles will be preceded
by abstracts in these four languages.” In the first volume, the vast majority
of articles were written in French. When Watts took over the editorship in
1987 and re-launched the journal with a slight shift in orientation towards
cross-cultural communication, he kept the policy that contributions could
be submitted in English, French, German, Italian or Spanish (cf. Watts
1987). Even though the abstracts in the four major European languages
were no longer provided, which might account for the decrease of the
number of non-English articles, the last one being published in 1998, it is
simply a fact that nowadays researchers prefer English as a lingua franca
among academics to writing in their mother tongue.
In his chapter on “The role of English in the research world”, Swales
(2004) shows that there is a similar situation in science, economics, psy-
chology and applied linguistics by giving statistical evidence from a variety
of sources. His most important findings, in my opinion, are that “[i]n to-
day’s Anglophone research world, the status and contribution of the non-
282 Jürg Strässler
native speaker of English has become somewhat more central than it used
to be” and that
there seems little evidence that the larger non-Anglophone research lan-
guages (French, German, Japanese) are resisting the advance of English any
better than smaller ones (Danish, Finnish, Portuguese). Indeed, the reverse
may be the case. (Swales 2004: 52)
Furthermore he concludes that there is a constant proportional decline of
the number of native speakers and that they will be outnumbered in the not
too distant future by non-native speakers. In addition, Swales deplores the
fact that
there has been a massive conversion over the last two decades from other-
language journals to English medium ones, and, as far as I can see, almost
all of the many new journals that have been springing up have an English-
only submission policy. (Swales 2000: 67)
Given the importance of English in the research world, it is therefore rather
surprising that there is only a very limited number of university courses
teaching academic writing at Swiss universities, and that there seems to be
little interest among the students. In 2002, in other words, relatively re-
cently, the University of Zurich and the Federal Institute of Technology
(ETH Zurich) established their common language centre, which
exists to foster the use of foreign languages at an academic level and acts as
an inter-university reference point for matters concerning the acquisition of
foreign languages and languages for specific and academic purposes.
(http://www.sprachenzentrum.uzh.ch)
This centre is now in charge of all language courses apart from those which
the language departments offer to their own students. At the University of
Basle there is a similar language centre (established in 2003) with the same
function, whereas the Centre for Language Competence of the University
of Berne (established in 2006) only offers courses in German as a foreign
language and English for science and for law.
In 2007 the Zurich language centre offered one introductory course in
academic writing in English (level B2)1 and one advanced course each for
arts and social science and for sciences and engineering (C1).2 Although
there is a target group of more than 40,000 students for all three courses,
the advanced course for arts and social sciences had to be cancelled due to
the low number of people interested. One of the reasons might be that the
description states that “the course will focus on both the techniques and the
linguistic conventions of academic writing” and that “the course does not
Can academic writing style be taught? 283
focus primarily on improving participants’ general English language skills”

(Torr 2007). A similar picture emerges at the Bernese Centre for Language
Competence, where in the autumn term of 2007 the course in English Sci-
entific Writing had to be cancelled because here, too, enrolment was low.
The language departments in Switzerland are the only ones that offer
their own courses as compulsory requirements for their students of lan-
guage and literature, who already have a high command of the respective
language and a certain awareness of the importance of different text types,
and it might be argued that they are the ones who need such tuition the
least.
Having a closer look at the topics of this type of courses in English we
realise that they concentrate on issues such as “the accurate use of cita-
tions, attribution of ideas, and the judicious use of figures, tables, equa-
tions and references [as the] critical components of a successful paper”3, on
grammatical and lexical issues as well as on the overall structure of an
academic paper. With respect to the notion of style, however, the course
descriptions refer to “learning-by-doing”, “practise this ability”, or “trying
to learn from the structure of articles and books that you are reading.” In
other words, style is considered to be an element that can neither be learnt
nor taught, but that has to be acquired individually.
Most of the literature on academic writing skills shows a very similar
approach to the topic as the courses mentioned above. Turabian’s (2007: 3)
Manual for Writers, which claims to have been the standard for genera-
tions of students and their teachers and which has sold more than 8 million
copies, states as one of the running themes that “readers will judge [your
research] by how well you report it, so you must know what they will look
for in a clearly written report that earns their respect”. Out of the 466
pages, a mere 10 pages are dedicated to language as such, the rest being on
planning, source citation and style, meaning, spelling, punctuation, names,
numbers, abbreviations, quotations as well as tables and figures. Further-
more, the chapter on revising sentences (Turabian 2007: 109–119) just
concentrates on common-place suggestions such as “[f]ocus on the first
seven or eight words of a sentence,” “make subjects short and concrete,”
“if you never make a verb passive, you’ll write sentences that contradict
the old-new principle,” “use first person pronouns appropriately,” among
others.
The Chicago Manual of Style (2003) as well as the MLA Handbook for
Writers of Research Papers (Gibaldi 2003), the two leading handbooks for
academic writing, show a very similar picture, apart from the fact that the
Chicago Manual contains a chapter covering American English grammar

and usage.
Textbooks on academic writing such as Bailey (2003), Baugh (1993)
Hamp-Lyons and Heasley (1987), Jordan (1999), Pirie (1985), Swales and
Feak (2004), Tribble (1996), to name a few, show a very similar design. As
they are mainly geared towards non-native speakers, they deal more with
actual language problems. However, although specific linguistic features
are linked to different stages of the writing process or to different genres,
that link often remains obscure and information on genre-specific style is
more or less inexistent.
In this paper I will refer to style as
[t]he characteristic use of language in a text. When referring to the speaker,
style is more or less the controlled choice of linguistic means, whereas in re-
ferring to texts, style is the specific form of language. (Bussmann 1996: 459)
I will not use style with reference to citation practice, typographic conven-
tions, formatting, etc., which can be looked up in the manuals mentioned
above. University departments and publishers have their strict norms with
respect to the overall presentation of the texts, often laid down in their
internal style sheets, but what their requirements for the actual use of lan-
guage are is not stated.
The question I will address in this paper is whether students maintain
personal, idiosyncratic features, which might reflect L1 principles, when
taught how to write academic texts in English or whether there is an obvi-
ous change towards an English academic writing style as observed by
Swales, who states that
[w]e are faced in effect with a growing linguistic and rhetorical monopoly
and monoculture against which we need to consider offering ‘cultural rain-
forest’ arguments of the following type: “Insofar as rhetorical practices em-
body cultural thought patterns, we should encourage the maintenance of va-
riety and diversity in academic rhetorical practices – excessive
standardization may counteract innovation and creative thought by forcing
them into standard forms” (Mauranen 1993: 172). (Swales 2000: 67)
For this purpose I will use the QSUM method, which has proved very fruit-
ful for the analysis of authorship in the framework of forensic linguistics.
In the next section I will explain the QSUM method in as much detail as
necessary and illustrate it with one of the student texts under scrutiny. This
will be followed by the actual analyses of texts written by the same stu-
dents at the beginning and at the end of a course in academic writing as
well as different texts written by native speaker academics. In the final

section I will try to interpret the findings and to establish the possible con-
sequences with respect to academic writing.
2. The QSUM method
2.1. Analysing for authorship
The cusum (cumulative sum) technique4 is a statistical method used to

establish the cumulative frequency of a variable as the sum of all preceding
frequencies, starting from the lowest value of the variable. It is a technique
widely used for instance in stock surveillance and quality control or as a
tool for feedback about performance in clinical medicine (cf. Chang and
McLean 2007). In linguistics the cusum technique is one of the stylometric
methods used in the attempt to ascribe authorship to written as well as spo-
ken texts, often used for forensic purposes.
The search for an objective method of authorship attribution has quite a
long history. As early as 1851, Augustus de Morgan (1851) suggested to
apply statistical methods to the Epistles to the Hebrews in order to deter-
mine whether they were written by the Apostle Paul. He claimed that if the
average length of words was much the same as the one in the other Epis-
tles, this might be a strong indication that Paul was the author, or at least
that all the Epistles were written by the same person.
Even though this might appear to be a rather simplistic and unscientific
method, it paved the way for stylometrics. De Morgan’s method did not
concentrate on overt individual language habits such as favourite words
and phrases, which above all are rather infrequent in relation to the overall
length of a given text, but on frequent, regular features authors use subcon-
sciously. Yule (1944) in his Statistical Study of Literary Vocabulary aug-
mented de Morgan’s method by adding elements of word distribution such
as sentence length with respect to word length and the ratio of the various
parts of speech within a given text.
In the 1960s, when computers became readily available, there was
growing interest in applying such techniques to larger corpora and to im-
prove the methods.5 Without going into detail we notice that from a lin-
guistic point of view there had been a constant shift away from semantics
and the assumedly author-specific choice of words towards the subcon-
scious but highly individual preference for specific syntactic structures.
2.2. Andrew Q. Morton’s QSUM method
It was in the 1960s when Andrew Q. Morton, a Church of Scotland minis-

ter and a computer specialist, started to apply cusum tests to language as a
method for authorship analysis. When he was convinced that this technique
was applicable to any type of written and spoken texts, apart from being
simple to use and easy to understand, he published the Qsum Plot (Morton
and Michaelson 1990), introducing his method and illustrating it with a
variety of text, mainly from modern literature. He was able to show that an
author’s writing style remains personal and consistent, not only within a
given text, but across different genres of writing and over time, even if the
author had deliberately changed her/his style.
The QSUM method is so convincing that it soon became recognised as
a valid technique in legal settings and “Morton’s expertise was to be
sought in many sensitive legal cases where ‘confessions’ or admission
statements were at issue” (Farringdon 1996: 15). The QSUM method is
thought to provide an authorial “fingerprint” similar to voice prints or real
fingerprints. Its success in forensic linguistics is due to its simplicity on the
one hand and reliability and applicability even in relatively short texts, and
thus being apt for analyses of alleged suicide letters, forged wills, forced
confessions, manipulated interviews, etc.
2.3. How to design and interpret QSUM charts
As mentioned above, the QSUM method is easy to carry out and even eas-
ier to interpret, as a cusum chart is basically a graphical representation of
the features under investigation. As sentence length, i.e. the number of
words per sentence, is the first stylistic aspect of relevance, I will explain
the design of the QSUM chart given in Table 1, which is based on one of
the students’ papers analysed in section 3.6
Every row in Table 1 represents one sentence of the text under investi-
gation and is labelled with the respective sentence number. In the second
column we have the words per sentence (wps) as well as the average sen-
tence length. The third column gives the sentence length deviation (sld),
i.e. the difference between the average sentence length of all the sentences
and their respective sentence length. For the fourth column the cumulative
sum of all the sentence length deviations (qsld) has to be calculated by
adding the sld of all the previous sentences. Columns 5 to 7 are not rele-
vant for the time being, but will be used below.
Table 1. QSUM chart of Text 1, Student A (A1)

(23lw + q(23lw +
# wps sld qsld 23lw + ivw
ivw)d ivw)d
1 19 -7.88 -8 9 -5.65 -6
2 20 -6.88 -15 12 -2.65 -8
3 26 -0.88 -16 14 -0.65 -9
4 8 -18.88 -35 3 -11.65 -21
5 42 15.12 -19 23 8.35 -12
6 18 -8.88 -28 10 -4.65 -17
7 40 13.12 -15 20 5.35 -12
8 9 -17.88 -33 4 -10.65 -22
9 27 0.12 -33 15 0.35 -22
10 45 18.12 -15 26 11.35 -10
11 23 -3.88 -19 11 -3.65 -14
12 22 -4.88 -24 15 0.35 -14
13 23 -3.88 -27 15 0.35 -13
14 51 24.12 -3 28 13.35 0
15 28 1.12 -2 11 -3.65 -4
16 35 8.12 6 21 6.35 3
17 21 -5.88 0 12 -2.65 0
Average 26.882 14.65
Key:
# = sentence number wps = words per sentence
sld = sentence length deviation qsld = cusum of sld
So far this method seems to be very similar to de Morgan’s (1851) tech-

nique of counting word length with the only difference that instead of
working with the bare values of the variable, they are put in relation to all
the preceding values and the overall average. In the graphical representa-
tion with the sequence of sentences on the abscissa and the qsld (column 4)
on the ordinate as given in Figure 1, we can see that the feature sentence
length is not a very reliable parameter for analysing for authorship. A vi-
able cusum chart would display a levelled graph with marginal deviations.
It is thus obvious that in addition to sentence length, other features have
to be found. Although Forsyth and Holmes (1996) in their paper on fea-
ture-finding deplore the fact that the choice of textual features, which is a
crucial determinant for success, is left to the intuition of the analyst, rather
than being based on an objective method for finding the most suitable fea-
tures in any given text, I suggest that we proceed with the ones suggested
cusum of sentence length 10
0
1 3 5 7 9 11 13 15 17
-10
-20
-30
-40
sequence of sentences
Figure 1. Cusum graph of sld of Text 1, Student A (A1)
by Morton (in Farringdon 1996). Firstly, there is no need for any back-
ground knowledge external to the texts being analysed and secondly they
have proved very successful, as can be seen in all the different applications
illustrated in Farringdon (1996).
As a second aspect of author-specific language use, occurrences of
other features or ‘habits’ within each sentence have to be counted. Morton
showed that an analysis combining words consisting of two or three letters
(23lw) and words with an initial vowel (ivw) proved to be very fruitful, at
least for English texts. The habit 23lw is a clear focus on function words,
as most of them consist of two or three letters and as the number of content
words shorter than four letters is very small. Based on absolute figures
given in Farringdon (1996: 45) the two- and three-letter words account for
0.2% and 2% of all words respectively, including the function words. Their
frequency in a text, however, is rather high. Among the 25 most frequent
words in the British National Corpus, only two words (that and with) are
longer than 3 letters. Furthermore, by concentrating on function words, the
focus shifts to syntactically significant elements. Vowel-initial words on
the other hand are very numerous in the English language, as a lot of them
are Latin prefixed. A rough count in the Oxford Advanced Learner’s Dic-
tionary (Hornby 1995) reveals that about 20% of the lexical entries start
with a vowel. Furthermore, by focusing on these two habits, we get a ratio
of nearly 50% of all the words occurring in a text, which is an ideal situa-
tion.
Considering the same text (A1) as above, we have entered the number
of habits per sentence in column 5 (of Table 1 above), the respective devia-
tions in column 6 and the cusum values in column 7. Rather than plotting
the cusum values into a separate graph, we integrate them into the graph
given in Figure 1, thus creating a representation of a QSUM chart, where
sentence lengths and the habits are put into correlation (Figure 2).
40
55 30
35 20
15 10
q23lw+ivw
qsld
0
-5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 -10
-25 -20
-45 -30
-65 -40
qsld q23lw+ivw
Figure 2. QSUM Chart of Text 1, Student A (A1)
In this chart we see the deviation from the average for the number of words
per sentence (qsld) as well as the habit being analysed within each sentence
(q23lw+ivw). If these two lines track each other closely enough, it is an
indication of the integrity of the text and of its homogeneity. The elegance
of the QSUM method is that reading such a chart requires no mathematical
knowledge, although the respective distances between the two lines could
easily be calculated and used as input for statistical analysis.
The very close match of the two curves thus shows that the text ana-
lysed is clearly consistent. Student A, at this stage in her development, has
found her own personal style, although she is a non-native speaker. Fur-
thermore, the whole text is free from any external influences, i.e., it was
written by one and the same writer. As we know that this is actually the
case, the QSUM method has once more passed the test. Farringdon (1996,
cover) claims that “[e]ach person’s QSUM ‘fingerprint’ retains consistency

across his or her written and spoken utterance and across different genres.”
In a number of analyses she further shows that the writer’s age does not
affect this consistency either.
In order to see whether this is also true for non-native speakers and
whether the active teaching of writing skills has any influence, positive or
negative, we will analyse a number of papers written by students attending
academic writing skills courses at the English departments of the Universi-
ties of Berne and Zurich.
3. Analyses of student papers
If Farringdon’s (1996) claims are valid, the teaching of writing skills would
have no influence on the personal writing style of the students attending
courses in which they are taught how to write academic texts. It must be
noted once more that the notion of ‘style’ in this context has nothing to do
with ‘style’ in the literary sense. In other words, we are not interested in
“[t]he characteristic manner of literary expression of a particular writer,
school, period, etc.” but in the “[f]eatures pertaining to the form and mode
of expression of a text” (Hornby 1995: s.v.). The results of my analyses
will therefore either refute the QSUM method or question the influence of
writing skills courses with respect to style. It goes without saying, how-
ever, that such courses are a valid part of the syllabus, whatever the results
may be, particularly if the participants perceive them as useful for their
studies and their academic work.
3.1. Methodology
For the present study I have randomly chosen 22 students in the English
departments of the Universities of Berne and Zurich. All of them attended
a course taught by different lecturers in which they had to write essays on
literary or linguistic topics at regular intervals. As a longitudinal study
would have gone beyond the scope of this paper, I decided to concentrate
on their very first texts and on the ones written at the end of the course,
that is on two texts for each student with a gap of 8 months in-between.
All the students are (very) advanced non-native speakers of English
with levels ranging from B2 to C2, according to the Common European
Framework of Reference for Languages. Their respective mother tongues
are mostly (Swiss) German, French or Italian, but also Spanish, Russian,
Swedish, and Croatian. Most of them had just started their studies at uni-
versity level, after having between four and seven years of two to three
lessons a week of tuition in general English at grammar school level.
Although I asked the students for permission to analyse their essays,
they were not informed about the purpose of the analysis or about the fea-
tures under scrutiny. Thus even if some of them might have edited their
texts with respect to grammatical or orthographic mistakes, this wouldn’t
have had any effect on the analysis.
Furthermore, the essays had to be slightly edited by myself. So names,
dates, numbers, references and especially quotations were marked as such
and analysed as being one word. For further details about preparing texts
for a QSUM analysis and reediting them, the reader should turn to Farring-
don (1996: 20–24).
3.2. Analyses and results
In order to illustrate how the QSUM method works, we have already ana-
lysed the first student paper in section 2 above. The first text by student A
(A1) was her very first essay at the beginning of the course. Although her
mother tongue is French, the analysis showed that her personal style when
writing in English is very homogeneous, i.e. it shows a uniform character
throughout with respect to the ratio of habits and sentence lengths. Such a
homogeneity “is taken to mean that the sample ... is indeed by the one
writer” (Farringdon 1996: 7) without any borrowings or insertions from
other sources. It might of course be interesting to compare her French and
English writing, but this would go beyond the scope of this study, as it
would imply having to identify common habits for both languages.
In all the student essays under scrutiny, a very similar picture emerged,
irrespective of what the mother tongue of the author was. We thus can as-
sume that their English, or rather their respective interlanguage, shows
idiosyncratic but consistent features. By way of further illustration let us
consider the QSUM chart given in Figure 3, which shows the analysis of
the first paper written by a student with Italian background (B1).
It can be seen very clearly that the charts given in Figure 2 and Figure 3
are both homogeneous but quite different. A critical reader unfamiliar with
QSUM charts might argue that it is obvious that the number of habits in a
sentence is always dependent on sentence length and that these charts are
therefore relatively meaningless. In order to counter this objection, we can
use the so-called sandwich method, which is a reliable means for assessing
whether a text of unknown origin is written by an assumed author or not.
For our purpose let us put text A1 between the first and the last 15 sen-
tences of text B2 and carry out the QSUM analysis. The resulting chart
given in Figure 4 displays a clear mismatch of the two curves, which is
proof that the intermittent text is of different origin.
100 60
75
50 35
q23lw+ivw
25
10
qsld
0
-25 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 -15
-50
-40
-75
-100 -65
qsdl q23lw+ivw
Figure 3. QSUM Chart of Text 1, Student B (B1)
100 60
75
50 35
q23lw+ivw
25
10
qsld
0
-25 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 -15
-50
-40
-75
-100 -65
qsdl q23lw+ivw
Figure 4. QSUM Chart of Sandwich (B1 – A1 – B1)

Thus far, the QSUM method has proved reliable and we can proceed with
the analyses of papers written at the end of the writing skills courses. If
Morton, Farringdon and their colleagues are correct in claiming that an
author’s personal style is consistent across different genres, we will get the
same picture: QSUM charts of the individual papers as well as sandwich
tests with text written by the same student at the beginning and at the end
of the course will prove homogeneous. This would of course also be an
indication that the students are resistant to changes in their individual writ-
ing style and that teaching genre-specific linguistic features of academic
writing would appear to be a fruitless undertaking.
However, much to my surprise, the results were quite different from
what was to be expected. The QSUM charts of all the texts written by the
22 students at the end of their courses show the same picture as given in
Figure 5, which represents the QSUM chart of the second text of the same
student B (B2).
120 75
95 50
70
q23lw+ivw
45 25
20
qsld
-5 0
-30 -25
-55 1 5 9 13 17 21 25 29 33
-80 -50
-105
-130 -75
qsdl q23lw+ivw
Figure 5. QSUM Chart of Text 2, Student B (B2)
As can be seen, the analysis of text B2 shows a lack of integrity and of

homogeneity in the text, although quotations and references had been ex-
cluded from the test. As indicated above, there might be two reasons: Ei-
ther personal writing is not consistent across different genres or the stan-
dards for academic writing taught in the courses are so rigid that they
displace some individual features.
In order to test whether the latter is the case, let us have a look at the
differences between personal and academic writing of two professionals of
English mother tongue. For this purpose I analysed a personal letter (C1,
D1) and a piece of academic writing (C2, D2) each, produced by two lec-
turers responsible for the courses and from whom I was able to obtain a
personal text. In order to avoid the conclusion that their academic writing
styles had influenced the students so strongly that they follow their mas-
ter’s voice, I have deliberately chosen tutors who were not directly in-
volved in the teaching of students A and B, whose text analyses are shown
above. They have, however, taught some of the other 20 students, whose
writing showed the same characteristics.
As was to be expected, the analysis of the personal letters (C1, D1)
showed a very high degree of consistency and I am convinced that com-
parison with other texts written by the same authors would show that they
have their own individual styles and that the QSUM technique is a valid
method to demonstrate this.
The QSUM chart of the academic text (C2, D2), however, showed a
completely different picture, as can be seen in Figure 6.
100
70
75
50 45
q23lw+ivw
25 20
qsld
0 -5
-25 1 6 11 16 21
-30
-50
-75 -55
-100 -80
qsdl q23lw+ivw
Figure 6. QSUM Chart of an academic text written by a professional (Text C2)
Apparently, academic writing has become so strongly standardised that it

does not leave much room for individuality. It seems that the authors of
academic texts feel forced to use constructions which are normally very
rare in their personal writing. What can for instance readily be noticed in
the texts C1 and C2 is the fact that the passive is a rare feature in the au-
thor’s personal writing, with only 2 occurrences, whereas in the academic
text excerpt of roughly the same length, it occurs 11 times. For a detailed
analysis of the passive in academic writing, see Wanner’s (in press) case
study in Deconstructing the English Passive.
4. Interpretation and conclusion
Although the present study is based on a rather small number of texts and
relies exclusively on the QSUM method, we can see a clear tendency that
academic texts differ from other text types with respect to their homogene-
ity. According to Farringdon (1996: 7) such heterogeneity “is clear evi-
dence of its being mixed utterance. What we have here is a borrowing or
insertion from other sources within the text.” However, as we know that
none of the texts lack integrity and that obvious insertions such as quotes
have been excluded from the analyses, we have to assume that academic
writing is strongly influenced by a genre specific standard. Swales (2004:
62–63) refers to Devitt’s (1997) article ‘Genre as a language standard’,
where she “argues that just as language standards provide rules of linguis-
tic ‘etiquette’ (punctuation, usage, etc.), so genres also have their associ-
ated ‘etiquettes’ ... [which] are not absolute but are conceived in terms of
what is socially and rhetorically appropriate.” Nevertheless, Swales goes
on to point out that “linguistic and generic ‘standards’ still permit a fair
amount of choice ... because not everything is controlled or controllable.”
The author of ‘It’s a steal’ (New Scientist 2007) is not aware of the
genre specific etiquette when he writes that “[e]loquent language has never
been the strong point of academic papers, so it’s somewhat ironic that
some scientists are lifting clever turns of phrase ... from other published
papers in a bid to sound more articulate.” He then goes on claiming that
“[m]ost culprits are people whose first language is not English”, which is
in strong contrast to Swales (2004: 52), who states that “[t]he difficulties
typically experienced by NNS academics in writing English are ... au fond
pretty similar to those typically experienced by native speakers.”
In her survey of the reviewing process of an applied linguistics journal,
Belcher (2007) also states that language use was not the most salient prob-
lem in international scholars’ submissions. She reports a case where a re-
viewer commented on a paper written by a native speaker of English that
“[i]t might be useful for a NS of English to read the text just to disentangle
some of the sentences.” However, the author’s disadvantage was not an

issue of language competence but the fact that he was an off-network
scholar with little access to scholarly publications, i.e. with little experi-
ence with the genre of academic writing.
Our analysis supports Swales’s claim and gives an explanation for
Belcher’s example by showing that for native as well as non-native speak-
ers academic writing differs considerably from other genres. Whereas the
QSUM method as applied to the non-genre-specific texts, i.e. the first es-
says of the students and the private writings of members of staff, showed
individual, yet consistent idiosyncratic features, academic writing seems to
be depersonalised. This does not mean, however, that they would adhere to
a common linguistic standard even if it were possible to tell what such a
standard might be. The only assumption that can be made is that the stan-
dard, genre-specific overall structure of academic papers as outlined by
Turabian (2007), Gibaldi (2003) and others has a strong influence on the
actual language used in academic papers.
Devitt (1997: 54) stated that “[o]nly when we understand genres as both
constraint and choice, both regularity and chaos, both inhibiting and en-
abling will we be able to help students to use the power of genres critically
and effectively.” One interpretation might thus be that the constraints are
imposed by virtue of genre-specific requirements and that choice is re-
flected in the individual writing style.
Although academic writing style is not taught explicitly but emerges in-
directly from the overall design of academic writing, we can draw the con-
clusion that the lecturers do a very good job in preparing their students for
a professional career. Furthermore, it might be hoped that students outside
the linguistics departments will discover the value of genre-specific lan-
guage courses and their relevance in the globalised world of academia,
where not only English as such, but English for academic writing has be-
come the Lingua Franca.
Acknowldegements
The author would like to thank Franz Andres Morrissey, Karl-Heinz

Gottschalk, Karin Grünwald, Miriam Locher and Brigitte Statzer-Zimmerli
for their valuable comments and helpful suggestions made on the first draft
of this paper.
Notes
1. According to the Common European Framework of Reference for Languages.

2. There are no courses in academic writing in any of the other languages, neither
at Zurich nor Basle. At Basle one single course in English academic writing (5
x 3 hours) is offered.
3. All the quotes in this paragraph are taken from different online course descrip-
tions.
4. Cusum is the term I use when referring to the neutral statistical technique,
whereas QSUM is its application for the analysis for authorship as developed
by Andrew Q. Morton (cf. Farringdon 1996).
5. For a more detailed account, see Farringdon (1996, chapter 1).
6. For a detailed step by step explanation of the QSUM method, see Farringdon
(1996, chapter 2).
References
Bailey, Stephen
2003 Academic Writing. A Handbook for International Students. London:
Routledge.
Baugh, L. Sue
1993 How to Write Term Papers and Reports. Lincolnwood: VGM Career
Horizons.
Belcher, Diana D.
2007 Seeking acceptance in an English-only research word. Journal of
Second Language Writing 16 (1): 1–22.
Bussmann, Hadumod
1996 Routledge Dictionary of Language and Linguistics. London:
Routledge.
Chang, W. R. and I. P. McLean
2007 CUSUM: A Tool For Early Feedback About Performance?
http://www.edu.rcsed.ac.uk/lectures/lt29.htm. (10 Nov., 2007)
de Morgan, Augustus
1851 Letter to the Revd. W. Heald, 18 August 1851. In Memoir of Augus-
tus de Morgan, Morgan, Sophia Elizabeth (1882), 214–216. London:
Longman; Green and Co.
Devitt, Amy J.
1997 Genre as a language standard. In Genre and Writing, Wendy Bishop
and Hans Ostrum (eds.), 45–55. Portsmouth: Boynton / Cook.
Farringdon, Jill M.
1996 Analysing for Authorship. A Guide to the Cusum Technique. Cardiff:
Forsyth, Richard S. and David I. Holmes
1996 Feature-finding for text classification. Literary and Linguistic Com-
puting 11 (4): 163–174.
Gibaldi, Joseph
2003 MLA Handbook of Writers of Research Papers. Sixth Edition. New
York: The Modern Language Association of America.
Hamp-Lyons, Liz and Ben Heasley
1987 Study Writing. A Course in Written English for Academic and Pro-
fessional Purposes. Cambridge: Cambridge University Press.
Hornby, Albert Sidney
1995 Oxford Advanced Learner’s Dictionary. Fifth edition. Oxford: Ox-
ford University Press.
Jordan, Robert R.
1999 Academic Writing Course. Third edition. Harlow: Longman.
Morton, Andrew Q. and Sidney Michaelson
1990 The Qsum Plot. Edinburgh: University of Edinburgh Press.
New Scientist
2007 It’s a steal. New Scientist, 31 March 2007, 193 (2597): 7.
Pirie, David B.
1985 How to Write Critical Essays. London: Routledge.
Sager, Juan C.
1982 Multilingua – Introduction and editorial programme. Multilingua 1
(1): 4–8.
Swales, John M.
2000 Language for specific purposes. Annual Review of Applied Linguis-
tics 20: 59–77.
Swales, John M.
2004 Research Genres. Exploration and Application. Cambridge: Cam-
Swales, John M. and Christine B. Feak
2004 Academic Writing for Graduate Students. Essential Tasks and Skills.
Second Edition. Ann Arbor: University of Michigan Press.
The Chicago Manual of Style
2003 The Chicago Manual of Style. 15th Edition. Chicago: The University
of Chicago Press.
Torr, Andrew
2007 Introduction to academic writing. Online course description at
http://www.sprachenzentrum.uzh.ch/angebot/kurse/sprachen/detail.ph
p?kursnr=1639&sprachnr=5. (15 Nov, 2007)
Tribble, Christopher
1996 Writing. Oxford: Oxford University Press.
Turabian, Kate L.
2007 A Manual for Writers of Research Papers, Theses, and Dissertations.
7th Edition. Revised and edited by Wayne C. Booth et al. Chicago:
The University of Chicago Press.
Wanner, Anja
in press Deconstructing the English Passive. Berlin: Mouton de Gruyter.
Watts, Richard J.
1987 Editorial. Multilingua 6 (1): 3–5.
Yule, G. Udny
1944 The Statistical Study of Literary Vocabulary. Cambridge: Cambridge
University Press.
Chapter 15
Linguascaping Switzerland:
Language ideologies in tourism
Adam Jaworski and Ingrid Piller
1. Language ideologies and media representations of tourism
The area of sociolinguistic research known as ‘language ideologies’, i.e.,

the study of a set of discourses “on language which represents a coherent
set of beliefs about language, a language, a language variety, language use,
language structure, etc.” (Watts 2001: 299) has for the past decade pro-
duced a number of insightful studies placing language at the centre of vari-
ous social, cultural and political debates (see, for example, Schieffelin,
Woolard and Kroskrity 1998; Blommaert 1999; Jaffe 1999; Kroskrity
2000; Johnson and Ensslin 2007). These studies are usually based on the
examination of competing discourses on language in particular countries,
locations, communities, and so on, in overt texts, written or spoken, evalu-
ating different aspects of language use from a variety of perspectives, such
as those of government officials, educators, media organizations, ‘ordi-
nary’ individuals, as well as other self- and other-appointed ‘experts’, in-
cluding linguists. The ideologization of language, however, does not only
take place through explicit metalanguage. Ideas about, beliefs of and atti-
tudes toward language with regard to patterns of prestige and standardness,
displays of authority and hegemony, acts of subversion and contestation,
orientation to the aesthetic dimension of code, and so on may be articulated
explicitly or made manifest in communicative practice, and they are always
linked to specific contexts of use and the speaker’s/writer’s sociocultural
knowledge (see, e.g., Woolard and Schieffelin 1994; Kroskrity 2004). In
this sense, all language use displays various degrees of reflexivity, or meta-
level which is inextricably linked with the indexing of specific language
forms as ideologically charged with respect to the relative power position
and identity of their speakers (or writers) (cf. Coupland and Jaworski 2004
for an overview). One recent example demonstrating language ideological
work through code choice in a bilingual setting is Alexandra Jaffe’s (2007)
study of language representation and use on Corsican radio and television.
302 Adam Jaworski and Ingrid Piller
Jaffe demonstrates how the choice of a minority language (Corsican), and

the degree to which it is spoken in media broadcasts indexes their audi-
ences as ‘Corsican’, whether a bounded and homogeneous speech commu-
nity through hypercorrect, Corsican monolingual, authoritative news radio
broadcasts, or a community undergoing a language shift in more relaxed,
mixed-language use in light radio entertainment programmes. Jaffe also
demonstrates how the representation of Corsican–French bilingual educa-
tion in a TV documentary is achieved, somewhat paradoxically, by the
editorial process of erasing in the final version of the programme most of
the French language usage by pupils and teachers to normalize, or authen-
ticate the use of Corsican as an academic register.
In an earlier study (Jaworski et al. 2003), we have examined the uses
and representations of languages other than English in tourist destinations
featured in British TV holiday programmes. In this work, we follow Deb-
orah Cameron’s claim that one of the central tasks of critical analyses of
language practice is to reveal how institutions like the broadcast media (cf.
Silverstein’s 1998 notion of ‘centering institutions’) produce and repro-
duce linguistic norms, and then to consider “how these norms are appre-
hended, accepted, resisted and subverted by individual actors and what
their relation is to the construction of identity” (Cameron 1990/1999: 62).
Watts (1999: 84) raises a similar sentiment when he states: “Since any
language ideology is constructed from mythical accounts of language use
and language structure, it is important to locate examples of those accounts
when observing the social practices of everyday life.”
Our earlier study reveals that languages other than English in TV holi-
day programmes function in four main categories: (1) expert talk (e.g.,
guided tours, explanations, instructions); (2) service encounters (e.g., pur-
chasing foodstuff in shops or markets); (3) phatic communion (e.g., ex-
changing greetings, thanking – often as part of a service encounter); and
(4) naming and translating (e.g., providing labels for local concepts, arte-
facts or dishes, translating local place names). However, further inspection
of metapragmatic comments about host languages made by programme
presenters, suggested an underlying set of beliefs about these languages:
that although tourists may have ‘fun’ trying them out in travel destinations,
they are primarily employed to index local people as out-group members in
relation to the British travellers (here, presenters and the implied tourists,
i.e., TV audiences), and a useful semiotic resource in creating a lingua-
scape of the travel destination, an authenticating and exoticizing strategy
relying on the display of host languages alongside iconic sites, ‘typical’
dishes, artefacts, activities, etc. Whether using a local language (other than
Linguascaping Switzerland: Language ideologies in tourism 303
English) addressing the TV viewers or in face-to-face interactions with

local people, presenters rarely venture beyond acts of language crossing,
i.e., using (snippets of) languages or varieties of language which the
speaker cannot legitimately claim membership of (see Rampton 1995,
1998). By appropriating host languages in this way, the presenters in the
holiday programmes never claim to be ‘really’ French, Italian, Kenyan, or
Fijian. On the contrary, more often than not, they emphasize their differ-
ence from their hosts. For example, by often sending themselves up as
incompetent host language speakers they position themselves firmly as
British (English-speaking) tourists. Different from the hosts and aligned
with the (implied) tourists in front of the TV screens, their ultimate aim is
to create a sense of community of (British-international) tourists with their
viewers, only playfully and fleetingly enacting a different linguistic, ethnic
or national identity. Through these programmes, and venturing into host
languages, they reassure their compatriots about the safety of foreign travel
in these destinations: they can remain British even outside of Britain, find-
ing the comforts of ‘home away from home’ and ways of getting by with
no or little foreign linguistic skill. Using host languages may be useful and
fun, but the low-level proficiency required on holiday does not threaten
anyone to having to become ‘someone else’. Not unlike Rampton’s discus-
sion of linguistic crossing creating new cross-ethnic patterns of identifica-
tion, crossing in the holiday programmes may be regarded as creating new
cross-national or inter-national allegiances, which allow one to maintain a
preferred, conservative national identity while momentarily venturing out-
side of one’s own national boundary. Even if one ends up in Italy, France,
Kenya or Fiji becoming Italian, French, Kenyan or Fijian is no more than
an optional pretend play.
Such play with local languages and orienting predominantly to the TV
audiences by the presenters leaves local people depicted in the programmes
with little control over their own subject-positioning or footing in relation
to the presenters and TV viewers, as their role in the participation frame-
work is constantly changing from the ratified hearer to ratified eavesdrop-
per, onlooker, or bystander, and object of the (implied) tourist gaze (Urry
2002). In Goffman’s (1981) terms, theirs is truly “subordinated communi-
cation” – in an interactional and also in a sociopolitical sense.
These mediatized representations of staged encounters in tourism are
indicative of the degree to which language has come to be ritualized and
commodified in tourism performances – what Jack and Phipps (2005) refer
to as the “languaging and translating” of the tourist landscape. By referring
to linguascapes and linguascaping, however, we mean also to reveal the
very particular deployment and devaluing of local languages as both back-

drop and ludic resource. It is in this way, that, as powerful ideological me-
diators, TV holiday programmes and their style-setting presenters promote
a regime of touristic and intercultural truth about local languages which
construct subjectivities for hosts and identities for viewers (cf. Foucault
1980; Mellinger 1994). Specifically, host languages are appropriated as a
primary identity resource by which visitors may construct themselves as
tourists. Furthermore, it is through their playful, transient crossings into
local languages that they further position themselves as “cosmopolitan
internationals” – not in the sense of their being culturally engaged with, or
embracing of, local people (cf. Hannerz 1996), but rather with respect to
their appeals to the elite caché of global citizenship. These are people
freely traversing national boundaries but staying firmly rooted in their mu-
tual identification as (British) nationals.
Just as tourists pre-visualize before their holidays the famous landmarks
to be visited, local ‘types’ to be encountered, leisure activities to be en-
gaged in, etc., so is the linguistic situation in the destination scrutinized
and assessed, even if only through the most banal, rhetorical question: ‘Do
they speak English there?’ Much of the tourist texts consulted before the
holiday provide just this information. Most tourist guidebooks published in
the UK, for example, include language glossaries, information on local
languages, the ability of locals to speak English or, more rarely, other tour-
ist languages, the desirability of tourists speaking some local language,
usually limited to a few phatic phrases, and other sorts of ‘facts’ on the use
and ecology of language in the tourist destination. Such descriptions, inevi-
tably conceived of as acts of linguistic differentiation are socially con-
structed and ideologically embedded practices (Gal and Irvine 1995; Irvine
and Gal 2000).
The newspaper travelogue is a well-established and popular genre of
tourism writing. In the UK, for example, many weekend editions of news-
papers, both those traditionally referred to as ‘quality’ / ‘broadsheet’ as
well as the ‘tabloids’, carry travel sections full of stories and accounts of
different types of holidays by travelling journalists, assorted ‘celebrities’,
and – increasingly – the reading and travelling public. These texts regularly
involve three major types of metalinguistic information with regard to the
destinations they describe: (1) naming of site-specific objects and concepts
in a local language (e.g., indigenous people, dishes, traditions, animals,
plants, artefacts, modes of transport, etc.); (2) quoting local people (e.g., in
interaction with the writer in English, or in another language, with the Eng-
lish translation present or absent); (3) metapragmatic comments about local
languages and hosts’ use of English; (4) describing holidays involving

foreign language learning.
In the following analysis of extracts from UK travelogues focusing on
Switzerland as a travel destination, we examine how the communicative
practices and the Swiss linguistic situation at large are reported as indexes
of particular types of people and Switzerland as a particular type of travel
destination. We particularly focus on the three semiotic processes of the
ideologization of language proposed by Judith Irvine and Susan Gal (2000;
see also Gal and Irvine 1995): iconization (attribution of an inherent con-
nection between a linguistic form or linguistic variety and social groups or
activities; a transparent and dependable representation of the specific and
distinctive qualities of the group), fractal recursivity (transposition of a
meaningful opposition salient at one level of social and linguistic organiza-
tion onto another level, e.g., projecting intra-group contrasts onto inter-
group contrasts, or vice versa); and erasure (a simplification of the socio-
linguistic field through which some sociolinguistic phenomena, and the
social actors involved in them, are rendered invisible, e.g., imagining a
social group or a language as homogeneous by disregarding its internal
variation).
2. Switzerland as a tourist destination in the UK weekend travel

sections
Our analysis is based on an ‘opportunistic’ sample of 23 travelogues pub-

lished in three UK national, middle-class-oriented, weekend national pa-
pers: The Sunday Times, The Guardian and The Independent published
between 1999–2006. The predominant focus of the articles is on winter
holidays in the Alps, particularly reporting and giving advice on winter
sports (mainly skiing), food and drink, accommodation and transport. Four
articles differ (slightly) from this pattern: one article features a report on a
‘walk’ across the Bernese Alps during summertime; two articles choose
non-Alpine locations, these being reports on the attractions of Zürich and
Lucerne; and another article also stands out: it is a feature on Romansch in
the canton of Graubünden (Milnes 2001; see Sources below for an over-
view).
Various countries and regions around the world have developed specific
genres of tourism for which they are known internationally (Urry 2002).
Probably most famously, the UK seems to have morphed into a giant ‘heri-
tage site’ attracting international tourists seeking to gaze at ‘history’ and
‘tradition’. Although not exclusively, France is typically associated with

romantic (Paris) and gastronomic tourism, Italy with cultural tourism and
Thailand with sex tourism. As reflected in our data, in the context of the
British media representations of tourism, and probably conforming to the
stereotype and also patterns of actual British tourist behaviour, Switzerland
appears to be associated predominantly with Alpine sports holidays. British
tourism to Switzerland has a comparatively long history as Tissot (2000;
2003) demonstrates: it is British travellers who actually “discovered” the
Alps as a tourist destination (rather than a hindrance or a place of hardship
and even terror to be avoided). Even today, British tourists are amongst the
top 5 source countries for tourists to Switzerland (No author 2006). One
consequence of the traditionally very strong presence of British tourists
and British institutions in Swiss tourism is the ubiquity of English in Swiss
tourism contexts (see Piller 2007 for an overview). Although English is not
an official language in Switzerland it is often called the fifth national lan-
guage (see, e.g., Watts and Murray 2001). The most widely-known socio-
linguistic characteristics of Switzerland are its diglossia in the German-
speaking region (Ferguson 1959) and its official quadrilingualism. The
official languages are (Standard) German, French, Italian and Ro-
mansch (in order of frequency) although the most widely used lan-
guages are probably German, French and Italian, with English being
widely used as the ‘first second language’.
2.1. Exotic linguascape – Romansch
Not unlike in travel stories about other destinations, where relatively lesser
spoken languages are used (e.g., Wales, the Basque Country, Malta, and so
on), articles about Switzerland often mention the smallest of its national
languages, spoken as a first language by approximately 0.5 percent of the
population – Romansch.
Extract 1 (Source: Kennedy 2002)
If Switzerland is the Greta Garbo of European states (its federal slogan
should be ‘We want to be alone’), then the Engandine [sic] must be its most
isolated outpost. Nearly four hours by train from Zurich airport, and 1,650
metres above sea level, it’s a collection of medieval villages and small farm-
ing communities stretched out across a vast Alpine valley.
Romansch, that curios hybrid of French, German and Latin, is the cen-
tral language of this canton. To add to the sense of seclusion, the Engandine
remains one of the least developed corners of the country, free of the high-
rise developments and commercial glitz that so characterise the usual ski-
resort scene. St Moritz is its notable exception. […]
Extract 2 (Source: Milnes 2001)

Mountains lock in languages, just as surely as they conserve old beliefs.
Consider that in one corner of western Europe, women did not get the vote
until 1990.
The location was the Romansch-speaking villages in the canton of
Graubünden in eastern Switzerland. Question: Where in the western world
was motorised transport not legalised until 1925? Answer: you’re getting the
picture.
So it was easy for me to assume in advance that the struggle to keep
the Romansch language, spoken by just 0.6 percent of the Swiss population,
from dying out was just another example of King Canute-type behaviour;
trying in vain to hold back the tides of language change.
As I arrived in the pristine Alpine town of Chur, capital of
Graubünden, the chilly mountain air acted as an antidote to the British traf-
fic fumes lodged in my lungs. There was an appealing bleakness about the
icy peaks looming over the little town. Once the heart of the Romansch-
speaking world, Chur is still the seat of Lia Rumantscha (the central organi-
sation for the promotion of the Romansch language and culture) and of Ro-
mansch radio and TV, so I was disappointed to find the German language
leaping out at me from billboards, signs, books and magazines. And despite
straining my ears for a foreign tongue, only the sound of Swiss-German
floated over coffee and cakes in the crowded cafés.
Fortunately, in my faux-rustic hotel, the young Latin-looking recep-
tionist, who was chatting to her mate in a language that sounded like hurdy-
gurdy Italian, instructed me where to go for the “real” Romansch experi-
ence. Tracing a line up the Engadine with her finger, she explained, “All
these villages used to be Romansch, but now they’re Swiss-German. But if
you go here or here.” She pointed to the area around Bad Scuol near the
Austrian border and around Ilanz and Disentis to the west, “everything is
Romansch; the kindergartens, the schools, the banks – you have to speak
Romansch to live there.”
I asked her where I could find interesting literature about Romansch
culture and history. “Well that’s the problem,” she said wistfully.
A sign for guests on the desk stated that the five major dialects of Ro-
mansch were all spoken in this hotel; Sursilvan (from the area of Surselva,
“above the wood”), Sutsilvan (from Sutselva, “below the wood”), Vallader,
Puter and Surmiran. The problem, I discovered, was that there was no single
written Romansch language.
I picked my way down icy pavements to the library of the Lia Ru-
mantscha where, wading through scholarly tomes on the subject, I learnt that
Rumantsch Grischun was invented around 1982 to address this problem.
Unfortunately, less than half the Romansch-speaking population support its
existence, since it is natural to no one.
Leaving Chur behind, I travelled west by train through the steep val-
leys of Surselva, sandwiched between the Glarner Alps to the north and the
Andulamassiv to the south, heading towards Andermatt. Once I’d got over
marvelling that it was possible for trains to run on time even when there was
several feet of snow, I started to understand how this diversity of dialects
came about.
Today, the bright-red train of the Rhätische Bahn stops every 15 min-
utes or so, as a cluster of snow-capped roofs comes into sight. Small towns
such as Ilanz, Trun and Disentis are linked to the outside world by rail, road
and global communications, even in midwinter. But in the days before mo-
torised transport and TV, these places must have been completely cut off for
much of the year, linked only by hazardous mountain-passes.
The train conductor passed the time by giving me examples of words
that vary from valley to valley; so “snow” in Romansch, I learnt, could be
“neiv”, “nev” or “naiv”. Ever since Rome invaded Rhaetia and vulgar
Latin got mixed in with the local Rhaetish, resulting in Romansch, dialects
have developed differently in separate secluded valleys, meaning that today
there is no united Romansch front to fight the battle for language survival.
Stopping in Ilanz, I attempted to get directions to the regional museum
of Surselva in the hope of finding out more, but was told disapprovingly that
Romansch was “being lived, not displayed”, and that anyway the museum
was shut. I was told this, of course, in Swiss-German.
But in fact, finding Romansch “being lived” didn’t seem to be such an
easy task. No doubt in some of the less-visited villages, white-bearded old
men and women with weathered faces still thrived, and the colourful Alpine
ceremonies I’d read about for taking the cows up and down the mountain
still punctuated the calendar. But in Disentis, when I asked about “cumegn”,
the traditional village voting assembly held every three years on 1 May in
which voters turn out in costume to elect their public officials, I was told
that the ceremony would sadly not happen in future. In the past, every vil-
lage had its own voting system, and it was this that enabled political conser-
vatism to thrive.
Today, insularity is gradually giving way to internationalism; Disentis
is offering holidaymakers the opportunity to build and sleep in their own ig-
loo; and in Ilanz the ancient cattle-market and old alleyways are now com-
plemented by an internet café. As communications have finally broken down
the isolation of different communities, the rural life and customs are evolv-
ing to suit the growing tourist industry.
Nowhere is this clearer than in Bad Scuol in the lower Engandine,

where, though tourism has not yet reached the mammoth proportions of St
Moritz, thermal spas and skiing mean that the town now has 3,000 beds for
visitors. Here, as in the glitzier resorts of the Upper Engadine, the old cus-
tom of “schlitteda”, in which boys sledge from village to village taking a girl
with them, has now become a kitsch tourist activity, with sledges replaced
by horses and carts in the summer. Alongside genuine traces of the past,
such as the benches outside houses, where residents still sit to exchange gos-
sip with their neighbours, are fake frescos painted on modern hotels, offer-
ing visitors the clichéd Switzerland of their imaginations.
In the year 1000, just about everyone in the current geographical area
of Graubünden would have spoken Romansch in some form. But gradually,
over the years, the language has been pushed aside – by German-speaking
settlers, the emigration of Romansch-speakers to Austria and the US, and,
more recently, by tourism and global communications. “The trouble is”, one
of the mountain guides told me “that when young people leave the villages
to work in towns, they don’t come back to the country.”
The receptionist in Chur had followed the same path as many of her
contemporaries, away from the country and the language she grew up with.
Switzerland is renewing financial and legal efforts to foster the Romansch
language and culture, but the evidence so far is that the growth of tourism is
inseparably linked to the decline of Romansch. So does it matter if Ro-
mansch disappears? In theory, one language is enough for the whole world,
let alone one canton. But as I was blasted by English advertising in Zürich
Airport, I felt a wave of resistance to the loss of diversity. Perhaps, I started
to think, a little Romansch conservatism was not such a bad thing after all.
Then again, perhaps I’d just spent too much time in Graubünden.
To begin with, it is worth noting that each of these extracts is characterized

by factual inaccuracies that serve to exoticize Romansch even further: En-
gadin, for instance, is not actually a canton as Extract 1 asserts – it is a
region in the canton of Graubünden. Women won the right to vote only in
1990 not in a Romansch-speaking area as Extract 2 claims but in the Ger-
man-speaking canton of Appenzell-Innerrhoden.
As is clear from the above extracts, mention of Romansch comes with
comments of ‘isolation’, ‘seclusion’, ‘underdevelopment’, and (traditional)
rural life. The principle of fractal recursivity is manifested here, by extend-
ing the ‘remoteness’ and other ‘peculiarities’ of the language to the geo-
graphical and social characteristics of the community. But just as in Extract
1, Romansch seems to be mentioned ‘for colour’, to add a little exoticism
to the skiing experience which the article focuses on, Extract 2 is a lengthy
(1220 word) feature centring wholly on the language and the Romansch-
speaking areas of Graubünden and Engadine. The same tropes of relative

isolation and backwardness are exercized, but here they are coupled with
nostalgia for (possibly) disappearing linguistic and cultural diversity under
the pressures of tourism and globalization, and the view of Romansch, and
the way of life associated with it, as ‘authentic’. The tourist quest for au-
thenticity (MacCannell 1989) is clearly a top priority for the author, and as
far as socio-linguistic authenticies are concerned, she finds them alongside
a number of different criteria (cf. Coupland 2003) such as
- Romansch indexing individual authentic speakers, e.g., ‘the young

Latin-looking receptionist’, or the pedagogically-inclined ‘train conduc-
tor’;
- Romansch being an attested and attestable, naturally occurring language
in ‘the area around Bad Scuol near the Austrian border and around Ilanz
and Disentis to the west’, and in the form of different dialects which
‘vary from valley to valley’;
- Romansch being a fully owned, unmediated language, even if it is only
declared as such by a local resident: ‘Stopping in Ilanz, I attempted to
get directions to the regional museum of Surselva in the hope of finding
out more, but was told disapprovingly that Romansch was “being lived,
not displayed”, and that anyway the museum was shut. I was told this,
of course, in Swiss-German.’ (but was the author attempting to speak to
the local resident in Romansch or even Swiss-German?); and, finally,
- Romansch indexing authentic cultural membership – ‘the young Latin-
looking receptionist’ again: ‘everything is Romansch; the kindergartens,
the schools, the banks – you have to speak Romansch to live there’;
even though some traditional forms of cultural life are disappearing
‘cumegn’, or becoming commodified and stylised: ‘schlitteda’.
Thus, even though still holding on to the status of an icon of the region
and part of the local linguistic landscape (cf. the mention of the sign in
stating that ‘the five major dialects of Romansch were all spoken in this
hotel’), the language seems to be giving way to one of a more prominent
official language of Switzerland ([Swiss] German) and the globalizing use
of English (cf. ‘English advertising in Zürich Airport’). This imbalance in
the use of Romansch and other languages of Switzerland (including Eng-
lish) is reflected in the remaining travelogues we analyse below.
2.2. Multilingual linguascape – German, French and Italian
Despite Switzerland’s multilingualism as a nation consisting of several

distinct linguistic areas, the majority of its population is said to remain
monolingual. For example, quoting Dürmüller (2001), Demont-Heinrich
(2005: 73) claims that “a majority of its citizens, with the exception of the
speakers of what some have called ‘small’ languages of Italian and Ro-
mansch, are in fact monolingual”. It is the peculiar tension between state
multilingualism and individual monolingualism that remains one of the
defining features of the Swiss ethnolinguistic identity (Demont-Heinrich
2005; Watts 2001). British press travelogues featuring Switzerland draw on
this fact by slipping in the display of the rather predictable snippets of
German, French and Italian words and phrases. As in other publications of
this type, they predominantly draw from the semantic domains of food,
drink, and other aspects of the locale. In our sample, we have found several
examples of this sort of language crossing, reproduced in Extracts 3–6
below, mostly but not exclusively dealing with the culinary semantic field.
Extract 3 (Source: Scott 2001)
Specialities here include warm chicken salad with wild mushrooms, pot au
feu and pancakes with berries.
[…]
Situated in a tiny hamlet of little mazot huts, Zum See has its own for-
mal terrace with Matterhorn views, but in busy periods – which seems like
every day in springtime – extra tables (sometimes simply comprising slates
on oil drums) are sited in sunny corners amid the huts. Max Menning is an
accomplished cook, and his home-made pasta is superb. Many people phone
ahead on their mobiles to reserve slices of Apfelstrudel and other confec-
tioneries. Max’s wife, Greti, exudes a serene calm as the ambience gets ever
more frenetic.
Extract 4 (Source: Anonymous 2005)

For rustic Alpine atmosphere: really, almost anywhere that sells vin chaud
or schnapps will do – provided the view’s good. Last season, our favourite
was the very rudimentary Holzerbar, at the side of the Tschuggen piste in
Grindelwald.
Sure, the gluhwein was good enough, but what really mattered was the
weather, the scenery (care of the north face of the Eiger) and the skiing: the
Tschuggen is one of the best high-speed carving tracks we know.
Extract 5 (Source: Bray 2000)

On the slopes by the Klein Matterhorn, it was even better – long, flattering
stretches of dry powder reaching down to Plan Maison, above Cervinia,
over the invisible frontier in Italy. We lunched off capaletti and tagaliatelle
ai funghi porcini and skied down, no passports needed, enjoying the sun-
shine from the south and the enthusiasm of Italians up for Sunday from the
Aosta Valley.
Extract 6 (Source: Brown 2006)

German-speaking Lucerne is a smooth, near-silent 50-minute train ride from
the airport at Zürich.
[…]
The name (spelt Luzern in German) probably derives from the Celtic
lozzeria, meaning “a settlement on marshy ground”, but the town dates its
origins from the mid-eighth century, when its Benedictine monastery is be-
lieved to have come under the control of the Alsatian Abbey of Murbach.
Lucerne’s ecclesiastical past keeps watch over the modern town from its
many spires and bell-towers, and the verdigris-laden onion domes of the
Jesuit Church, where several of this year’s spiritually-inclined concerts will
take place.
You might also try a little lip-smacking walnut schnapps. […]
Many venues will also warm guests with a Kaffee fertig: a coffee laced
with schnapps.
All the forms quoted above, pot au feu, mazot huts, Apfelstrudel, vin
chaud, schnapps, gluhwein, capaletti, tagaliatelle ai funghi porcini, Kaffee
fertig (see also “neiv”, “nev” or “naiv”, in Extract 2), show considerable
variation in the use of the scare quotes, italics, or no special marking for
the non-English words and phrases. This seems to reflect the uncertainty of
the texts’ authors whether to treat these forms as foreign citations, loan-
words, or fully assimilated English-language forms. Collectively, alongside
overt comments about languages spoken in particular towns or areas (see
Extract 6), they appear to add to the iconization of Switzerland as a for-
eign-language and multilingual tourist destination.
2.3. Safe linguascape – English
However, in our data, by far the most frequent language to represent inter-
actions between tourists (writers) and local people is English. Whether
English was the original language of the interaction, which is most likely,
or whether it is an unacknowledged translation is never clear. As has been

mentioned, most features in our sample present the attractions of winter /
skiing holidays, with only occasional mention of non-Alpine resorts. The
Swiss people represented in these stories are quite limited in type, mostly
(named or unnamed) tourism operators such as owners or managers of
restaurants, hotels, skiing businesses, guides, and so on. One of them – a
named ‘igloo supervisor’ – in Extract 7 below appears in dialogue with the
author of the article:
Extract 7 (Source: Newson 2006)
‘Now,’ said Rahel Zürcher, the igloo supervisor, as she gave me my 8am
cup of tea, ‘do you want to catch the railway back down the mountain, or
shall we go sledging?’
Of course, we took the less sensible option.
Other English-speaking Swiss (and foreign) people encountered in our data

are other Swiss and international, always English-speaking tourists. In ex-
tract 8, the two tourists are a Swiss male skier and an Italian female train
passenger. They are both ‘othered’ somewhat through the author’s mild
ridicule of their attire, taste and intellect. However, as they are members of
the privileged ‘leisure class’ (Veblen [1912/1994], the damage of such
prejudicial language, in our view, is relatively minor:
Extract 8, continuing from Extract 1 (Source: Kennedy 2002: 5)
‘This is where the jet set come in’, Gerhard told me.
Gerhard was a middle-aged gent decked out in one of those stretch-
Lycra tracksuits that tends to accentuate every bulge in the human body (and
that led me to wonder: why do the taciturn, uber-controlled Swiss go in for
these cartoonish super-hero outfits?). As we stood there talking, a Gulf-
stream jet made its approach towards the nearby runway.
‘It’s probably Naomi Campbell,’ Gerhard said, with a mischievous
smile. ‘It’s only a 15-minute drive from the airport to St Moritz, so all the
beau monde land here.’
‘And how far is it by skis?’ ‘Another 13km.’
I paid the equivalent of three quid for a train ticket to Zuoz, and stood
for most of the way back, gazing out at the terrain I’d covered earlier that
day. By rail, the journey took 20 minutes. A seat finally opened up and I
flopped into it – right next to a chic Italian woman in regulation electric-red
parka and overpriced eyewear.
‘You look happy,’ she said.
‘That’s because I’m not skiing any more,’ I said, then explained that it
had taken me five unstinting hours to haul myself from Zuoz to St. Moritz.
‘Five hours by skis?’ she said, aghast. ‘You must be insane. The train
is so much faster.’
‘Really?’ I said.
The English spoken by the authors, hosts and other tourists in these and
other extracts creates a sense of cosmopolitanism of Switzerland. The un-
problematic use of English as a lingua franca indexes the tourists and hosts
as international, globalized, well-educated and wealthy. Given that the
newspapers in which these travelogues appear aim at elite, middle-class
UK readership who arguably identify themselves with this sort of self-
image, seems to lend support to this interpretation. The topics of the con-
versations and of the descriptive parts of many extracts are also frequently
focused on ‘global’ topics such as ‘designer’ brands, royalty, ‘celebrity’,
elite lifestyle. In Extract 8 above, this is achieved through the mention of
the approaching Gulfstream jet, Naomi Campbell and St Moritz as an ap-
parent magnet for ‘the beau monde’ in general. In other articles, readers are
exposed to typical lists of available ‘designer’ shops located in various
towns and resorts, e.g., ‘Prada, Gucci, Louis Vuitton, Dolce & Gabbana
and Bulgari are only the start of it’ (Newsom 2003). These signifiers of
extreme luxury lifestyle, ‘iconic’ media personality, and the crossing into
French (in the predominantly Romansch/German speaking area) reinforce
the status of Switzerland generally and St Moritz in particular as a glamor-
ous and elite tourist destination.
English-speaking hosts and tourists are not the only resource used in
these travelogues to create a sense of familiarity with Switzerland for the
British readers. Continuing the recognizable trope of high-profile portrayal
of the Swiss ski resorts, the authors repeatedly mention other ‘celebrities’
frequenting their slopes, exclusive hotels, bars, and discos such as British
royals and an assorted list of mostly British and some ‘international’ celeb-
rities: Prince Charles (and his family), the Duchess of York (with prin-
cesses Beatrice and Eugenie), George Clooney, Claudia Schiffer, and many
others. As a sure sign of Switzerland being internationalized, albeit from a
decidedly ‘Anglo’ perspective, references to the British presence in Swit-
zerland past and present are common. For example, Extract 9 plays on a
commonly acknowledged ‘appropriation’ of Switzerland by British tourists
going back to the Victorian era:
Extract 9 (Source: Brown: 2006)
It was the English who began the vogue for Swiss tourism, my guide Gabby
tells me as we strike out on a bracing hike around the lake, and Lucerne’s
popularity snowballed after a long stay by Queen Victoria in 1868.
The strong presence and ‘Anglicization’ of Switzerland appears to con-

tinue today, not surprisingly drawing on the imagery of traditional British
upper-class lifestyle:
Extract 10 (Source: Mills 2005)
Other recent developments include the acquisition (six years ago) of the Ho-
tel de la Paix by Stephen Purdew, the owner of the famous Champney’s
health spa chain. Purdew has spent over £1 million furnishing the hotel in
English country style. He’s named the rooms after counties, fitted tartan car-
pets and en suite whirlpool tubs. In typically tricky Champéry style, it’s a
corporate-bookings-only operation. The owner insists it is a labour of love
rather than a serious business venture.
Overall, the newspapers create an image of a safe English linguascape

through easy English conversations with locals and other English speakers,
being surrounded by international, commercial signage and more localized
but resolutely English linguistic landscape, e.g., “…The Pipe, a snow-
boarders’ hang out, which advertises ‘groovy music, chillin’ sport videos
and wild party nights’” (Bray 2000), and Anglo-themed hotels (see Extract
10). To invoke Gal and Irvine’s (1995) terminology again, all of these
metadiscursive representations ideologize language by recursively trans-
posing the international status of English (and French) to represent Swit-
zerland as cosmopolitan and globalized, and by erasing local languages
from the scene creating a sense of Switzerland as a familiar winter holiday
destination.
3. Conclusion
Tourism has become a key practice of our age and tourism discourses con-
stitute one of the key ways of producing knowledge about places. In this
chapter, we have explored the linguistic knowledge about Switzerland that
is produced and reproduced in one such tourism discourse, British newspa-
per travelogues. We observed three aspects of the language make-up of
Switzerland that were discussed in those travelogues: to begin with, the
smallest national language, Romansch, received a disproportionate amount
of attention. It is the high level of attention in itself that serves to exoticize
Romansch and additionally Romansch is exoticized by being presented as
an aspect of nature – linked to the seclusion and isolation of the disparate
mountain valleys where it is spoken, linked to speakers as phenotypes (“a
Latin-looking receptionist”), and linked to political units that may seem
exotic to the British reader (e.g., the Engadin as canton) or to historical

events that demonstrate backwardness (e.g., the late date when women
gained the right to vote). The larger national languages of Switzerland –
German, French, Italian – are also exoticized but to a relatively lesser de-
gree and through one linguistic practice only, namely the use of local terms
for local cuisine, landmarks or cultural events – a practice that can be ob-
served relative to Romansch, too, of course. All these practices serve to
render Switzerland linguistically special, “exotic” and a place that is
clearly marked as “not home”. It is also noteworthy that the writers have
completely failed to see any languages they did not expect to see, namely
the various immigrant languages of Switzerland. Given the high number of
immigrants working in the Swiss tourism sector, it is unlikely that this
oversight is due to a lack of opportunity but rather to a lack of expectation
of those languages. In the same way that “all the dialects of Romansh”
were spoken in the hotel that the writer of Extract 2 visited, typical Swiss
hotels would also have staff who are speakers of Croatian, Portuguese,
Serbian, or Spanish (these are immigrant languages in which training mate-
rials for tourism workers are typically produced; see, e.g.,
http://www.quality-our-passion.ch).
We can thus observe standardization in our data relative to what counts
as touristically legitimate languages of Switzerland: the official languages
only – the travelogues do not in any way deviate from the official version
of the Swiss linguistic landscape. Another aspect of standardization that
can be observed in the travelogues is the normalization of English: by not
identifying the language in which encounters between the writer and locals
took place or identifying it as English, English is naturalized as the lan-
guage of tourism to Switzerland, and, of course, beyond – Switzerland is
hardly an exception in this regard. Tourism discourses are yet another dis-
course where the ubiquity of English is brought into existence.
In sum, tourism as a practice carries with it its own form of linguistic
standardization: to begin with, it is official versions of what counts as a
language and which languages are spoken where and by whom that become
widespread knowledge through travelogues and other tourism genres. Sec-
ond, English is naturalized as the international lingua franca spoken by
“everyone” – everyone who matters to the tourist that is. What becomes
normalized is the cliché on the one hand and the convenient on the other –
two forms of “nothing” – forms devoid of specific content – that Ritzer
(2007) sees as characteristic of cultural globalization. The languages of
tourism – be they local languages or English – have been standardized to
become non-languages, devoid of anything that makes them special.
Acknowledgement
The research for this paper was in part supported by funding from the
Leverhulme Trust (Grant No. F/00 407/D) to the Centre for Language and
Communication Research, Cardiff University, for a larger project on Lan-
guage and Global Communication. In part it was also supported by funding
from the Swiss National Fund (project number 108608) for a larger project
on Languages, identities and tourism: Towards an understanding of social
and linguistic challenges in Switzerland in the context of globalization
(grant holders: Ingrid Piller and Alexandre Duchêne).
Sources
Anderson, Alf
2002 Park and ride. The Guardian. 23 February 2002/ pp. 16–17. [Loca-
tion: Swiss Alps]
2004 The right kind of snow. The Guardian. 13 March 2004. pp. 12–13.
[Location: Swiss Alps Grindelwald].
Anonymous
2005 Gluhwein snifters. The Sunday Times. 6 November 2005. p. 27. [Lo-
cation: Swiss Alps].
Ashdown, Paddy
2002 If this is faking it, then I’m all for it. The Independent. 3 February
2002. p. 19. [Location: Swiss Alps].
Bray, Roger
2000 Round the Horn. The Guardian. 23 December 2000. pp. 8–9. [Loca-
tion: Swiss Alps / Zermatt].
Brown, Helen
2006 Water music. The Independent. 4 March 2006. pp. 10–11. [Location:
Engadine / St Moritz].
Cox, Jonathan
2003 It’s party time. The Sunday Times. 22 June 2003. p. 2. [Location:
Lucerne, Lucerne International Music Festival, August 14th to Sep-
tember 20th].
Garbett, Will
2001 Europe’s country trails. The Guardian. 4 March 2001. p. 9. [Loca-
tion: Bernese Alps – Feature on ‘Walks on the wild side – from the
Alps to Madeira’s banana canals’].
Holford, Nicky
2000 White weekends. The Guardian. 25 November 2000. pp. 10–11.
[Location: Swiss Alps – A ‘winter break’ feature on Italy, France,
and Switzerland].
2002 Emotional peaks. The Guardian. 9 February 2002. p. 13. [Location:
Swiss Alps / Verbier].
Kennedy, Douglas
2002 Get into the grove. The Sunday Times. January 6, 2002. p. 5. [Loca-
tion: Engadine / St Moritz].
Mills, Simon
2005 Skiing as it used to be. The Guardian. 12 November 2005. pp. 6–7.
[Location: Swiss Alps / Champéry].
Milnes, Anthea
2001 There’s nothing like a bit of Romansch. The Independent. 23 June
2001. p. 5. [Location: Graubünden].
Newsom, Sean
2003 Moritz for mere mortals. The Sunday Times.26 January 2003. p. 8.
[Location: St Moritz].
2004 Bargain time on the piste. The Sunday Times. 15 February 2004. p.
10. [Location: Swiss Alps].
2005 White frontiers The Sunday Times. 2 October 2005. p. 21. [Location:
Swiss Alps].
2006 A place to chill. The Sunday Times. 19 February 2006. pp. 24–25.
[Location: Swiss Alps].
Rufford, Nick
2002 A ski pass to the stars. The Sunday Times. 8 December 2002. p 8.
[Location: Swiss Alps].
Scott, Alistair
1999 Klosters – fit for a prince? The Sunday Times. 12 December 1999. p.
7. [Location: Swiss Alps / Klosters]
2000a Apartment life for the smart skier. The Sunday Times. 5 November
2000b High life: The best bars in the Alps. The Sunday Times. 31 December
2001 The high art of lunch. The Sunday Times. 18 February 2001. p. 9.
[‘10 top alfresco eateries’ in the Alps].
Warwick, Samantha
2002 The ultimate blind date. The Guardian. 27 April 2002. p. 9. [Loca-
tion: Zürich].
References
Blommaert, Jan (ed.)

1999 Language Ideological Debates. Berlin: Mouton de Gruyter.
Cameron, Deborah
1990/1997 Demythologising sociolinguistics. In Sociolinguistics: A Reader
and Coursebook, Nikolas Coupland and Adam Jaworski (eds), 55–
67. Basingstoke: Macmillan.
Coupland, Nikolas
2003 Sociolinguistic authenticities. Journal of Sociolinguistics 7 (3): 417–
431.
Coupland, Nikolas and Adam Jaworski
2004 Sociolinguistic perspectives on metalanguage: Reflexivity, evaluation
and ideology. In Metalanguage: Social and Ideological Perspectives,
Adam Jaworski, Nikolas Coupland and Dariusz Galasinski (eds), 15–
51. Berlin: Mouton de Gruyter.
Demont-Heinrich, Christof
2005 Language and national identity in the era of globalization: The case
of English in Switzerland. Journal of Communication Inquiry 29 (1):
66–84.
Dürmüller, Urs
2001 The presence of English in Swiss universities. In The Dominance of
English as a Language of Science: Effects on Other Languages and
Language Communities, Ulrich Ammon (ed.), 389–403. New York:
Mouton de Gruyter.
Ferguson, Charles A.
1959 Diglossia. Word 15 (2): 325–340.
Foucault, Michel
1980 Power/Knowledge. New York: Pantheon.
Gal, Susan and Judith T. Irvine
1995 The boundaries of language and disciplines: How ideologies con-
struct difference. Social Research 62 (4): 967–1001.
Goffman, Erving
1981 Forms of Talk. Philadelphia: University of Pennsylvania Press.
Hannerz, Ulf
1996 Transnational Connections: Culture, People, Places. London:
Routledge.
Irvine, Judith T. and Susan Gal
2000 Language ideology and linguistic differentiation. In Regimes of Lan-
guage. Paul V. Kroskrity (ed.), 35–83. Santa Fe, New Mexico:
School of American Research Press.
Jack, Gavin and Alison Phipps

2005 Tourism and Intercultural Exchange: Why Tourism Matters. Cleve-
don: Channel View.
Jaffe, Alexandra
1999 Ideologies in Action: Language Politics on Corsica. Berlin: Mouton
de Gruyter.
2007 Corsican on the airwaves: Media discourse, practice and audience in
a context of minority language shift and revitalization. In Language
in the Media, Sally Johnson and Astrid Enslin (eds.), 149–172. Lon-
don: Continuum Press.
Jaworski, Adam, Crispin Thurlow, Sarah Lawson and Virpi Ylänne-McEwen
2003 The uses and representations of host languages in tourist destina-
tions: A view from British TV holiday programmes. Language
Awareness 12 (1): 5–29.
Johnson, Sally and Astrid Ensslin (eds)
2007 Language in the Media. London: Continuum.
Kroskrity, Paul V.
2000 Language ideologies in the expression and representation of Arizona
Tewa ethnic identity. In Regimes of Language: Ideologies, Polities,
and Identities, Paul V. Kroskrity (ed.), 329–359. Santa Fe, New
Mexico: School of American Research Press.
2004 Language ideologies. In A Companion to Linguistic Anthropology,
Alessandro Duranti (ed.), 496–517. Oxford: Blackwell Publishing.
MacCannell, Dean
1989 The Tourist: A New Theory of the Leisure Class. Berkeley, Califor-
nia: University of California Press.
Mellinger, Wayne M.
1994 Toward a critical analysis of tourism representations. Annals of Tour-
ism Research 21 (4): 756–779.
No author
2006 Swiss Tourism in Figures. Berne: Swiss Tourism Federation & Fed-
eral Statistical Office.
Piller, Ingrid
2007 English in Swiss tourism marketing. In Wildern in luso-austro-
deutschen Sprach- und Textgefilden: Festschrift zum 60. Geburtstag
von Erwin Koller [Roughing it in the Linguistic and Textual Wilds of
Portuguese, Austrian and German: Festschrift for Erwin Koller on
the Occasion of his 60th Birthday], Cristina Flores and Orlando
Grossegesse (eds.), 57–73. Braga, PT: Cehum – Centro de Estudos
Humanísticos.
Rampton, Ben
1995 Crossing: Language and Ethnicity among Adolescents. London:
Longman.
1998 Language crossing and the redefinition of reality. In Codeswitching

in Conversation, Peter Auer (ed.), 290–317. London: Routledge.
Ritzer, George
2007 The Globalization of Nothing 2. London: Sage.
Schieffelin, Bambi B., Kathryn A. Woolard and Paul V. Kroskrity (eds.)
1998 Language Ideologies: Practice and Theory. New York: Oxford Uni-
versity Press.
Silverstein, Michael
1998 Monoglot ‘standard’ in America: Standardization and metaphors of
linguistic hegemony. In The Matrix of Language: Contemporary
Linguistic Anthropology, Donald Brenneis and Ronald K.S. Macau-
lay (eds), 284–306. Boulder, Colorado: Westview Press.
Tissot, Laurent
2000 Naissance d’une industrie touristique: Les Anglais et la Suisse au
XIXe siècle. [Birth of a Tourist Industry: The English and Switzer-
land in the 19th century.] Lausanne: Payot.
Tissot, Laurent (ed.)
2003 Construction d’une industrie touristique aux 19e et 20e siècles:
perspectives internationales [Development of a Tourist Industry in
the 19th and 20th Centuries: International Perspectives]. Neuchâtel:
Ed. Alphil.
Watts, Richard J.
1999 The ideology of dialect in Switzerland. In Language Ideological
Debates, Blommaert, Jan (ed.), 67–103. Berlin: Mouton de Gruyter.
Theory, Nikolas Coupland, Christopher Candlin and Srikant Sarangi
(eds.), 297–320. London: Pearson Educational.
2001 Die fünfte Landessprache? Englisch in der Schweiz [The Fifth Na-
tional Language? English in Switzerland]. Berne: Vdf.
Veblen, Torsten
1912/1994 The Theory of the Leisure Class. New York: Penguin Books.
Urry, John
2002 The Tourist Gaze. 2nd edition. London: Sage.
Woolard, Kathryn and Bambi B. Schieffelin
1994 Language ideology. Annual Review of Anthropology 23: 55–82.
Chapter 16
The rules of “Denglish”
Elke Hentschel
1. Introduction
As early as in 1996, a colleague teaching computer sciences asked for my

advice with regard to linguistic problems in her students’ MA-theses. The
problems in question were due to the integration of English terms into the
otherwise German text, which led to a language mix often called “Deng-
lish” (from deutsch ‘German’ and English). The result were sentences like
the following:
(1) Die einzelnen Steckplätze der Backplane müssen die Möglichkeit bieten, die
Daisy-Chain durchzuschleifen, falls in den Steckplatz ein Bus-slave-Modul
eingesetzt werden soll.
‘The individual slots of the back plane have to offer the possibility of loop-
ing through the daisy chain, if a bus slave module is to be inserted into the
slot.’
(2) Bei einer dritten Lösung, dem Bus-Snooping (“Bus-Schnüffeln”) oder Bus-
Watching (“Bus-Beobachtung”), beobachtet die Cache-Steuerung den Bus
hinsichtlich der Speicherzugriffe anderer Master.
‘In the third solution, that of bus snooping (German translation in brackets)
or bus watching (German translation in brackets), the cache control watches
the bus with regard to memory access by other masters.’
These examples show two different ways of dealing with the evolving
problems: In example (1), the English expressions (highlighted) were inte-
grated into the construction and treated like German words (which in-
volves, among other things, the assignment of grammatical gender). The
author of example (2), while basically using the same strategy, additionally
gave a translation of the English expressions into German.
Sentence (1) contains three Anglicisms: Backplane, Daisy-Chain and
Bus-slave-Modul.1 The last example is the most interesting one insofar as it
is a hybrid form: the last element is German (the corresponding English
324 Elke Hentschel
word would have to be spelled module). As to Bus, the first part of this
composition, the written form of the word happens to be identical in both
languages. This leads to pronunciation problems when reading the com-
pound aloud: should it be pronounced [aUr rkdHu] as in the standard Eng-
lish pronunciation (cf. Jones 2006 s.v. bus), or [aTrrkdHu], following the
standard German pronunciation (cf. Mangold 2005, s.v. Bus)?
In example (2), the author tries to deal with the language mix problem
by adding translations in brackets. It is, however, interesting to note that he
does not do so with every English word in the sentence: while bus-
snooping and bus-watching were considered worthy of a translation and
thereby marked as foreign terms in need of explanation, this was not the
case with cache (cf. Cache-Steuerung) and master. The author is obviously
quite accustomed to them and treats them as if they were German words.
What is more, he puts the translations into quotation marks, whereas the
foreign terms themselves stay unmarked, and are adapted to the rules of
German orthography by the use of capital letters.
The examples illustrate Anglicisms as a problem in computer sciences
in German speaking countries, where countless English technical terms
have to be translated or otherwise integrated into German. But computer
sciences are by no means the only field where this kind of problem arises.
Due to the growing internationality of sciences in general, Anglicisms can
be found in almost every scientific field, from medical to nuclear physics
(cf. e.g., Röhrenbeck 1988; Schmitt 1985). But not only there: indeed, they
have become a part of everyday language. Teenage language (cf. e.g., Zi-
fonun 2000), or “adspeak” (cf. e.g., Schütte 1996), just to name two out of
many examples, are full of them (see also Pittner 2001 for further exam-
ples). The reason why the following survey focuses on examples taken
from the area of every day computer usage is that in this field, German
alternatives to the English expression are often hard to come by. Speakers
do not use them in order to achieve a certain stylistic effect, but simply
because it is difficult to find an adequate German expression.
The above examples illustrate several central problems that arise when
English words are integrated into German: There are phonetic / phonologi-
cal, morphological, syntactic and last but not least orthographic questions
to be solved. In the following text, problems of the first kind – for instance
the question of whether the first two syllables of Bus-slave-Modul should
be pronounced [aUr rkdHu] or [aTrrkdHu] – shall not be dealt with. Neither
shall syntactic issues like the adoption of English sentence patterns and
government rules into German (e.g., the accusative construction etwas
erinnern instead of prepositional sich an etwas erinnern under the influ-
The rules of “Denglish” 325
ence of English to remember something). Instead, orthographic and mor-

phological problems and their intersection shall be focused upon. Morpho-
logical problems have to be dealt with in any case – whether the items in
questions are used in writing or in speaking. On the other hand, the ortho-
graphic solution chosen by the language users often gives insight into their
unconscious analysis of a word.
Moreover, morphology and orthography, taken together, offer the best
possible view on the development of new standards. In the given context,
the term “standard” is not meant in the sense of prescriptive norms, which
in most cases do not yet exist, but refers to naturally evolving rules. When
users cannot look up a word in a dictionary or grammar book, they do not
make random choices, but use the rules that they feel should be applied in
the case given. These more or less spontaneous decisions are mirrored best
by the language use on Internet pages.
The following analysis is based on empirical data collected in 2003 and
then again three years later, in 2006, in order to illustrate the development
in users’ strategies over the last years. The orthographical and morphologi-
cal integration of Anglicisms can be explored together by examining verbal
participles. Here, in addition to the development of orthographical stan-
dards, the use of the participle marker -ge- leads to the necessity to apply
German morphological rules to foreign words (cf. also Pittner 2001: 232)
and mirrors the evolving morphological standards.
2. Some general observations on spelling and inflection
2.1. Spelling conventions
When it comes to spelling, one might be inclined to think that the central
problem lies only in the question of whether to use the English spelling and
thereby marking the word as foreign, or to integrate it into the German
orthographic system and thereby adopting it more closely into the language
(for integrational processes of this kind as mirrored by dictionaries, cf. e.g.,
Langner 1995; Lee 1996). However, things are far from being that simple.
What we are dealing with can be illustrated by the following list of tokens
taken from a voluminous corpus of Swiss German chat logfiles, most of
them documenting local language varieties.2 The forms in Table 1 all rep-
resent adaptations of the English verb to chat and the nouns chat and chat-
ter.
326 Elke Hentschel
Table 1. Adaptations of the English verb to chat and the nouns chat and chatter
ch@ttet(3rd sg) chattä cheten Chett
Chät chätta chetta
Chat chättä chette
Chat chättät (3rd Sg) chetu
chata chatte
chatä chätte gechattet (part)
chäta chatten gechatut (part)
chätä chätten gichatter (part)
chätät (part*) chätter gichattet (part)
chate Chatter gichattut (part)
chäte chattest (2nd sg)
chaten chättet (3rd sg) schatto
chäten chättet (part) Tchätt
Chater chattet (2nd pl) tschät
chätet (2nd pl) chattet (3rd sg) tschätä
chatet (part) chattet (part) Tschätt
chatet (3rd sg) chatti (1st sg) tschätte
chatisch (2nd sg) chattisch (2nd sg) Tschätter
chato Chatto tschättet (part)
chätö Chattu tschetä
Chäts (pl) Chättu Tschett
Chatt chattut (3rd Sg) tschettu
Chätt Chatu cheten Chett
Chatt Chätu chetta
chatta chatusch (2nd Sg) chette
ch@ttet(3rd sg) chatut (3rd Sg) chetu
*part = participle
It turns out that the adaptation of the spelling does not occur in toto, but
undergoes several individual steps and shades. This can be seen all the
more clearly in this case because many of the chatters used their regional
dialect, which encouraged them to a greater amount of creativity in spelling
and made the integration process more transparent. In principle, one can
distinguish three elements:
– The affricate of the onset [sR] can be rendered either by ch (English) or

by tsch (German spelling). Sch occurs only once and is therefore diffi-
cult to evaluate; it might well be a typing error.
– The nucleus vowel, pronouncedZæ\in the original language, in Ger-
man mostly rendered as [D] or [d], can be spelled as a, ä or e. The
choice might partly depend on the local pronunciation.
– The coda [t] can be rendered as t or tt. In both languages, the gemina-
tion can be used to mark the preceding vowel as short.
As a result, for the German infinitive alone, the following 12 combinations

are possible (Table 2). The three forms marked with an asterisk, however,
did not occur in the corpus.
Table 2. The possible orthographic realizations of to chat in German
chaten chatten *tschaten tschatten
chäten chätten *tschäten tschätten
cheten chetten *tscheten tschetten
Obviously, the simple dichotomy “original spelling” versus “adaptation

into the German system” does not lead very far. It is the adaptation process
itself that causes the problems and leads to inconsistencies. Should pho-
netic or phonological principles prevail? Should the shrinking phonological
stock of standard German, where short [D] is partly being replaced by [d],
be taken as basis, or rather the richer inventory of the local varieties? And
last but not least: how to integrate the morphological changes into the
spelling, which invariably occur during the adaptation process?
Since it is clearly not possible to address all these questions in this pa-
per, the focus shall in the following lie on the spelling of past participles.
In the case of regular (“weak”) verbs, their historically related forms are
very similar in both languages and differ only in the last consonant: while
English participles end in -d, their German counterparts end in -t. When an
English verb is integrated into German, the participle will be formed by a
prefixed ge- and a suffixed -(e)t, yielding forms like ge[chat]et.3 Now the
spelling of the coda is a simple dichotomic difference – it has either to be d
or t –, easily investigated and used as a possible indicator of how the or-
thographic standardization of loanwords develops.
2.2. The inflection of verbal participles
The correct use of German verbs with prefixes depends on several factors,
but the general rule is: If the prefix carries the word accent, it has to be
separated from the stem for finite present and preterite forms, infinitives
with zu and participles. In the latter cases, the morphemes -ge- (participle)
and -zu- (infinitive) are infixed between the verbal prefix and the stem (cf.
Hentschel and Weydt 2003: 53f.). Especially in the case of infinitives, the
328 Elke Hentschel
rules are not always completely straightforward, so sometimes native

speakers themselves are not sure which form to use (cf. Becker and Peschel
2003). With participles, however, there are but few cases of doubt, which
is why they make good examples for the following analysis.
Verbal prefixes that can be separated normally belong to a class that
simultaneously serve as prepositions.4 If an English verb shows a similar
structure, i.e., if it consists of a verbal stem with a prefixed preposition like
in to downgrade, undergo, upload etc., the underlying structure is in most
cases intuitionally transparent to German speakers; the languages are, after
all, very closely related. Now if a verb with this kind of structure is
adopted and completely integrated into German, it should show the same
features as a German verb, which means that the treatment of the prefix
should follow the rules of German grammar. The formation of the partici-
ple can therefore be seen as an indicator of the degree of the integration,
which in this case always also means the growing standardization of its
forms. Some of the correspondent forms have even found their way into the
most recent edition of the German standard dictionary Duden – a diction-
ary originally meant for spelling, but used by many people as a kind of
reference in all cases of doubt – where they are listed with separated parti-
ciples (cf. Duden 2006 s.v. downloaden, updaten, upgraden). No problem,
therefore – or is there?
3. Empirical findings
The following analysis will take into account empirical findings collected
in February 2003, and in May and July 2006, when the use of certain ver-
bal participles by German speaking Internet users were investigated. The
verbs in questions were the German adaptations of to forward, to upload/to
download, to telnet, to update and to ftp (the abbreviation stands for file
transfer protocol). In both data collections, all possible variations of the
past participles – which can be inflected for case, gender and number –
were searched for, and the number of hits was recorded. It is true that this
method can in some cases lead to double recording. However, since the
same search machine (Google) was used, and the numbers found are not
used absolutely, but only in relation to one another, this possible source of
error lies well within the range of what is acceptable. The order in which
the findings are presented in the following text mirrors the frequency of the
forms investigated from least to most frequent.
3.1. File Transfer Protocol (ftp) as a German verb
Ftp was initially only the abbreviation for a transmission system on the
Internet, the so-called file transfer protocol. The same abbreviation ftp (in
lower case letters) is at the same time the command with which the proto-
col is called up. This led first to an English verb to ftp, and then to its Ger-
man counterpart ftpen [Desdodi?m], both used to describe the process of
transferring data by ftp. The protocol is not used by all Internet users, and
therefore the frequency of ftpen and its participles is not very high. What is
most interesting about it is the fact that the predominant majority of the
users are very much aware of the unusualness of the derivational pattern. In
2003, 103 of 143 findings (72%) showed an explicit marking: either the
form was hyphenated (ftp-en), or an apostrophe was used in order to mark
the morpheme boundary (ftp’en). Even the majority of the 40 cases without
marked morpheme boundary showed morpho-derivational awareness inso-
far as the abbreviation is spelled with capital letters (FTPen) or quotation
marks are used for the derivated verb (“ftpen”). Only very few unmarked
cases were recorded. In 2006, 425 of 695 findings (61%) were hyphenated
or presented with an apostrophe. Among the other cases, the frequency of
unmarked use had clearly grown to an approximate fourth. However, the
upper case spelling FTPen was clearly still the majority.
The following examples show the use of the verb with hyphen, apostro-
phe, upper cases and/or quotation marks in context. They are presented
with an English translation in single quotation marks, which is as idiomatic
as possible. The verb in question is highlighted in bold:
(3a) Warum sollte man zum FTP-en ein kryptisches CLI-Tool verwenden (...)
‘Why should one use a cryptic CLI tool in order to ftp’
(3b) Es gab eine Anfrage, woher man PGP Software ftp-en koennte.
‘Someone asked from where one could ftp PGP software.’
(3c) FTP Voyager macht FTP’en so einfach wie (...)

‘FTP Voyager makes ftp-ing as easy as …’
(3d) Die Benutzer können ausschließlich über den Webbrowser ftp’en.

‘The users can only ftp by using the web browser.’
(3e) Will man also “von aussen” einen FTP auf sein Mathe-Account machen,
sollte man merkur an-ftp-en.
330 Elke Hentschel
‘So if one wants to make an ftp “from the outside” to one’s mathematics’ ac-
count, one should ftp to merkur.’
(3f) Dateien zum FTPen sucht man am besten über PDLOOK des RZ der Saar-
Uni.
‘It is best to search for files for ftp-ing with PDLOOK of the RZ of Saar Uni-
versity.’
(3g) Um Dateien zu “FTPen” genügt meist ebenfalls der Webbrowser.

‘In order to “ftp” files, usually the web browser is sufficient, too.’
The infinitive forms that could be recorded were:

simple infinitives: ftp-en, FPT-en, “FTPen”, ftp’en, FTP’en
nominalized infinitive: zum FTPen
infinitive with prefix: an-ftp-en
In 2003 there were only 4 cases of the participle to be found, all of them
part of a perfect tense form and therefore without ending, with upper cases
as marker: geFTPt. Example (3h) illustrates this usage:
(3h) Hab mir mal das Handbuch geFTPt.
‘Just ftp-ed the manual.’
In 2006, there were 17 cases of geFTPt, again all of them part of a perfect
tense, and no attributive forms could be found. The majority showed upper
case spelling, but there were three occurrences of geftpt, too. Hyphens and
apostrophes were not used.
One most interesting case of a composition with an English verbal pre-
fix (up-) plus the German affixes ge- and -t could be registered in 2003 as
well as in 2006 as part of a passive construction. In this case, hyphenating
was the manner of choice in order to mark the morpheme boundaries:
up-ge-ftp-t. Example (3i) is a case in point:
(3i) Diese Datei ist manchmal nicht vollständig up-ge-ftp-t worden (...)
‘This file sometimes has not been “ftp-ed up” completely …’
3.2. Telnet(t)en
The English verb to telnet is based on the identical Unix- or Linux-

command telnet which leads to a certain type of connection between two
computers on the Internet. The German verb offers a choice of spelling in
the infinitive, where it can be found as telneten or telnetten. With around
200 (in 2002/3) respectively 640 (in 2006) recordings the verb occurs rela-
tively rarely. As in the case of ftp, telnetting might be an activity used only
by a minority of Internet users, and even with those, it tends to be replaced
by the newer and more secure connection via ssh (secure shell).5
However, the results of three queries, the first one in November 2002,
the second one in January 2003, and the third one in July 2006, are surpris-
ingly consistent. In all cases, the spelling with double t and without any
additional markers was clearly preferred by 73 to 74% of the users, and
only a small minority of between 4 and 6% found it necessary to mark the
morphological boundary between the English stem and the German infini-
tive affix. This might be due to the fact that the syllabic structure as well as
the spelling of the word are completely compatible with the German lin-
guistic system. Table 3 shows the findings in detail.
Table 3. The use of to telnet in German
recorded in telnetten telneten telnet-en getelnetted getelnettet
2002 152 (73%) 44 (21%) 11 (5%) 0 0
2003 157 (74%) 47 (22%) 9 (4%) 1 1
2006 476 (74%) 129 (20%) 37 (6%) 2 12
The participle of this verb could theoretically be one of the following eight
variants: getelnet(t)et, getelnet(t)ed, telgenet(t)et and telgenet(t)ed. In 2003,
only two of them occurred, each of them used once: getelnettet, getelnet-
ted. In 2006, there were 12 findings for getelnettet and 2 for getelnetted
with the English spelling of the participle suffix. Again, telgenet(t)et and
telgenet(t)ed did not occur. Obviously, German speakers do not accept tel-
as a separable prefix, and the verb therefore cannot be split.
3.3. Forwarden
The English verb to forward has been transferred to German only in com-
puter contexts, meaning the forwarding of e-mail. The infinitive forwarden
yielded 2,510 findings in 2003, and as many as 77,900 in 2006. The verb
332 Elke Hentschel
cannot yet be found in the standard German reference book Duden (2006).
Since the first syllable for is phonetically identical with the German sepa-
rable verbal prefix vor, there are two morphological as well as two spelling
alternatives for the participle: geforwardet, geforwarded, forgewardet and
forgewarded. The results of the two queries are shown in the following
Tables 4 to 7. The percentages given refer to all forms found in the two
investigations, i.e., 742 in 2003 and 35,750 in 2006. For illustration, you
can find examples after every table.
Table 4. geforwardet
Year findings %
2003 381 51%
2006 16,600 44%
(4a) Ein Port kann immer nur zu einem bestimmten Client geforwardet werden.
‘A port can only be forwarded to a specific client.’
Table 5. geforwarded
Year findings %
2003 329 44%
2006 17,500 46%
(4b) (…) dass Mails an diese Adresse noch ein weiteres Mal geforwarded werden.
‘… that mails are forwarded one more time to this address.’
Table 6. forgewarded
Year findings %
2003 17 2%
2006 2,130 6%
(5a) (...) User-Mails, die auf den Palmlounge POP3 Account forgewarded
wurden (...)
‘… user mails that were forwarded to the Palmlounge POP3 Account …’
Table 7. forgewardet
Year findings %
2003 14 2%
2006 1,520 4%
(5b) (...) weil (...) keine Mails forgewardet wurden.

‘... because no mails were forwarded.’
It is clear that the predominant majority of speakers do not consider for-

warden a separable verb. A slightly growing minority of 10% (6% of them
spelling it with -d and 4% with -t), however, insert the prefix of the partici-
ple between for- and -ward. Strangely enough, this strong integration into
the German class of separable verbs does not automatically lead to a pref-
erence of the German spelling in the participle ending; on the contrary, it
seems that these speakers even prefer the English spelling, which does not
fit well with the German morphology. But since separated participles are
but a small minority, the data cannot be interpreted too strongly.
The non-separating majority of users are not too sure about the spelling
of the participle either. It is surprising to see that the German spelling that
had been used in 51% of the non-separated verb forms in 2003, is only
preferred by 44% of the users of this variety in 2006. Far from becoming
better integrated, as one might have expected especially in view of the
increasing frequency of its use, in terms of spelling, the form seems rather
to have become slightly less integrated, and the evolving standard de-
creased.
In light of this, it is of course most interesting to investigate if the spell-
ing in -ed is only an interference from English, or if it is “meant” that way.
In German pronunciation – which seems the most likely choice in view of
the German prefix ge- – the otherwise phonological difference between /t/
and /d/ does not apply in the coda, since final stops are always voiceless.
Both spellings would therefore represent [t]. In the position before a vowel
in the following syllable, however, the distinction between the voiced and
the voiceless consonant is clearly articulated. If spelling the participle with
-d is only due to the habitual English spelling of the word and the fact that
it does not mark a pronunciation difference in this position, it should cease
to occur in the case of participles inflected for gender, number and case.
Therefore, the occurrence of inflected variants of all four forms has been
investigated. The results were as follows (Tables 8 to 10):
Table 8. geforwardet, inflected forms
-en -e -er -es -em sum
2003 53 61 6 13 0 133
2006 1,380 801 53 33 61 2,328
334 Elke Hentschel
(6a) Die Vergabe von geforwardeten Subdomains und Verlinkung erfolgt (...)
‘The distribution of forwarded subdomains and the linking occurs …’
(6b) Die Anzeige schon anderweitig geforwardeter Nachrichten (...)

‘The indication of mails that have already been forwarded elsewhere …’
Table 9. geforwarded, inflected forms

2003 2 1 1 0 0 4
2006 27 36 1 0 0 64
(7a) (...) danke für die geforwardede Antwort.

‘... thanks for the forwarded answer.’
(7b) (...) besonders wenn die geforwardeden > rauszulöschen sind.

‘... especially if the forwarded > have to be deleted.’
Table 10. forgewardet, inflected forms

2003 0 1 0 0 0 1
2006 86 62 2 0 1 151
No findings of forgewarded in an inflected form could be recorded in ei-

ther year. The summary of the findings is shown in Table 11:
Table 11. Summary of participles derived from forwarden

inflected forms on –t % inflected forms on -d %
2003 134 97% 4 3%
2006 2,479 97% 64 3%
Although the number of inflected forms is much smaller than that of forms
without endings, the tendency is quite clear, and it remained unchanged
over the years: If there was a morphological marking that influences the
pronunciation of the word, 97% of the users preferred the German spelling.
3.4. Upgraden
Upgrading seems to be something the Internet users do more often than

forwarding – or at least they use the verb referring to this process far more
often: the Internet research resulted in 30,200 (2003) and 1,830,00 findings
(2006) for the infinitive. It is therefore not surprising that this verb can also
be found in the most recent edition of the Duden (Dudenredaktion 2006
s.v. upgraden), where its forms are given with upgradet (third person sin-
gular present) and hat upgegradet (third person singular perfect).6
As with forwarden, again it is the participle, its inflection and spelling
that are of interest. The following findings could be recorded for partici-
ples without ending:
Table 12. Non-inflected participles of upgraden

findings 2003 % 2003 findings 2006 % 2006
upgegradet: 809 45% 40,800 43%
upgegraded: 578 32% 25,900 28%
geupgradet: 277 15% 10,900 12%
geupgraded: 153 8% 16,500 18%
forms on -t, sum 1086 60% 51,700 55%
forms on -d, sum 731 40% 42,400 45%
-ge- inserted, sum 1387 76% 66700 71%
ge- prefixed, sum 430 24% 27400 29%
Although the German spelling prevails from the beginning, it cannot be

said that it was strengthened over time – on the contrary, it even lost a little
of its percentage. 45% of all users still prefered the English spelling. A
similar phenomenon can be stated for the non-separated versus the sepa-
rated, i.e., fully integrated forms: the latter also dropped slightly in use.
Although the numbers per se are not significant, it is still astonishing to see
that the verb has not become better integrated in the last three years.
As to the spelling, it is again interesting to see what happens when the
participle is inflected for case and number. The research yielded the fol-
lowing findings:
Table 13. upgegradet, inflected forms

2003 127 58 16 3 7 211
2006 3,190 4,470 657 1,210 150 9,677
336 Elke Hentschel
(8a) (...) die bei upgegradeten Systemen falsch war.

‘... that was wrong in the case of upgraded systems.’
(8b) Ein ‘upgegradeter’ Palm synct sich nur mit ebenfalls ‘upgegradetem’ Desk-
top.
‘An upgraded Palm synchronizes only with a likewise upgraded desktop.’
Table 14. upgegraded, inflected forms

2003 0 1 0 0 0 1
2006 114 63 22 0 0 199
(9a) (...) mit voll upgegradeden Rittern (...)

‘... with fully upgraded knights ...’
(9b) (...) upgegradedes Lichschwert (...)

‘... upgraded light sword ...’
Table 15. geupgradet, inflected forms

2003 3 0 0 0 0 3
2006 587 880 46 85 24 1,622
(10a) Jason’s überraschende Rückkehr in der geupgradeten Super Duper (...)

‘Jason’s surprising return in the upgraded Super Duper ...’
(10b) Und dann soll dir ein einzelner geupgradeter Berserker im Vorbeigehen die
komplette Army wegrölzen ? ! ! ?
‘And then a single upgraded berserk is to casually slag off your complete
army?!!?’
Table 16. geupgraded, inflected forms

2003 0 0 0 0 0 0
2006 3 36 0 0 0 39
(11a) ... ist die geupgradede version dann auf der max cd?
‘... is the upgraded version to be found on the max cd?’
(11b) (...) kommt die KI mit haufenweise geupgradeden Truppen daher.

‘... the AI comes along with heaps of upgraded troups.’
Table 17. Inflected participles of upgraden, summary

inflected forms on -t % inflected forms on -d %
2003 214 99.5% 1 0.5%
2006 11,299 98% 238 2%
These findings seem to sustain the hypothesis that spelling the participle
with -d is but an interference from English orthography, which almost
completely vanishes as soon as inflected forms are used, where this spell-
ing is adverse to the pronunciation.
3.5. Updaten
Among the verbs investigated here, updaten with 58,100 (2003) respec-
tively 2,510,000 (2006) occurrences of the infinitive is the next in fre-
quency. Again, this verb has been adopted into the Duden (Dudenredaktion
2006 s.v. updaten), and as with upgraden, the forms are given only for the
third person singular present and perfect, where the participle is rendered
as upgedatet.
The language users themselves, however, were not so sure. The results
presented in Table 18 are the numbers for participles without further in-
flection.
Table 18. Non-inflected participles of updaten
findings 2003 % 2003 findings 2006 % 2006
upgedatet 24,000 52% 434,000 34%
upgedated 9,390 20% 248,000 19%
geupdatet 7,160 15% 288,000 22%
geupdated 5,740 12% 325,000 25%
forms on -t, sum 31,160 67% 722,000 56%
forms on -d, sum 15,130 33% 573,000 44%
-ge- inserted, sum 33,390 72% 682,000 53%
ge- prefixed, sum 12,900 28% 613,000 47%
It is surprising that the “Germanized” forms, be it the spelling of participles

on -t, be it the insertion of the morpheme -ge-, did not become more fre-
quent over the three years of the observation period, but on the contrary
rather lost their dominance. Especially in the case of the separation, the
percentage fell from as much as 72% in 2003 to a mere 53% in 2006.
338 Elke Hentschel
Again, the inflected forms have been searched for and analyzed. Since
updaten is more frequent than the verbs that have been dealt with so far,
the number of inflected forms was far larger, too. Table 19 gives a sum-
mary of upgedatet and upgedaded, while Table 20 shows the findings for
the prefixed forms geupdatet and geupdated.
Table 19. -ge- inserted (upgedatet/d), inflected

2003 1,690 99% 20 1%
2006 105,430 99% 701 1%
(12a) Überaus praktisch, so ein alljährlich upgedateter Berlin-Stadtplan.

‘Exceedingly practical, such a yearly updated city map of Berlin.’
(12b) Die Version 6.0 wurde auf einem komplett upgedatetem OS aufgebaut.
‘Version 6.0 was built on a completely updated OS.’
(12c) Tunes 3 mit ‘smart playlists’ (...) und natürlich upgedateder ipod.
‘Tunes 3 with ‘smart playlists’ … and of course an updated Ipod.’
(12d) Vielleicht kann man da was mit nem upgedatedem Treiber fixen?
‘Maybe it can be fixed with an updated driver?’
Table 20. ge- as a prefix (geupdatet/d), inflected

2003 451 99% 4 1%
2006 21,486 97% 634 3%
(13a) (...) zur Übertragung neuer oder geupdateter Seiten (...)

‘... in order to transfer new or updated pages ...’
(13b) Ein täglich geupdatetes Informationsportal (...)

‘A daily updated information portal ...’
(13c) (...) mit der geupdateden Version (...)

‘... with the updated version ...’
(13d) (...) mit geupdatedem Kernel (...)

‘... with an updated kernel ...’
Obviously, the use of the inflected form again interferes with the -d-
spelling. This effect is altogether very strong, slightly more pronounced in
the case of the Germanized morphology with infixed -ge- (upgedatet/d).7
3.6. Uploaden and downloaden
In 2003, there were 1,130,000, and in 2006, as many as 50,400,000 find-

ings for the infinitive downloaden, while there were only 30,700 (2003)
and 3,570,000 (2006) for uploaden. The difference in frequency mirrors, of
course, the difference in use: people download something from the Internet
more often than they upload. This fact seems to have influenced the editors
of the Duden, too, where only downloaden is recorded (Dudenredaktion
2006: s.v. downloaden), which is treated in exactly the same way as up-
daten and upgraden. Although uploaden, as the figures from the Internet
queries show, is much more frequent than updaten or upgraden, it has not
yet found its way into the dictionary.
Since both verbs share the same stem load, the results of the two que-
ries are pooled and presented in Tables 21a and 21b, which show the oc-
currences for participles without case/number-marking endings. Now this
is certainly an astonishing result: instead of becoming more integrated,
these verbs seem to become less so the more often they are used. In the
case of uploaden, the usage of the participle with infixed -ge- dropped by
18 points from 97% to 79%, but still stayed in the majority. In the case of
the more frequent downloaden, however, the frequency dropped from 68%
to a mere 26%, while the prefixed version reached the astonishing value of
74%. As there were already more than 40,000 hits in 2003, and there were
more than 800,000 in 2006, any errors due to too small a database can be
ruled out. In addition to this, these findings are in line with the ones con-
sidered so far, where the number of cases was considerately smaller, as
was the percentage loss of the Germanized forms; but the same tendency
was visible there, too.
340 Elke Hentschel
Table 21a. Non-inflected participles of uploaden

findings 2003 % findings 2006 %
upgeloadet 18,000 87% 128,000 61%
upgeloaded 2,080 10% 37,100 18%
geuploadet 418 2% 16,700 8%
geuploaded 301 1% 26,700 13%
Σ -t 18,418 89% 144,700 69%
Σ -d 2,381 11% 63,800 31%
Σ -ge- 20,080 97% 165,100 79%
Σ ge- 719 3% 43,400 21%
Table 21b. Non-inflected participles of downloaden

findings 2003 % findings 2006 %
downgeloadet 6,640 38% 17,900 3%
downgeloaded 5,240 30% 124,000 23%
gedownploadet 3,900 22% 255,000 47%
gedownloaded 1,710 10% 140,000 26%
Σ -t 10,540 60% 272,900 51%
Σ -d 6,950 40% 264,000 49%
Σ -ge 11,880 68% 141,900 26%
Σ ge- 5,610 32% 395,000 74%
As for the inflected forms of the participles, the following numbers could
be established:
Table 22. -ge- inserted (upgeloadet/d; downgeloadet/d), inflected forms

up down
inflected % inflected % inflected % inflected %
forms on forms on forms on forms on
-t -d -t -d
2003 1,047 99% 10 1% 1,903 98% 39 2%
2006 13,096 99% 191 1% 37,818 99% 559 1%
(14a) Hier finden Sie auch Ihre downgeloadeten Spiele.

‘You’ll also find your downloaded games here.’
(14b) (...) wenn man mit dem Downgeloadetem nicht zufrieden ist?
‘... if one is not content with what has been downloaded?’
(14c) Entpacken Sie das downgeloadede ZIP-Archiv (...)

‘Unzip your downloaded Zip archive…’
(14d) (...) einen Computer mit downgeloadeder Software (...)

‘... a computer with downloaded software ...’
(14e) Ein von Dritten upgeloadetes Sample (...)

‘A sample that has been uploaded by a third party ...’
(14f) (...) nach jedem fertig upgeloadetem Bild ein Update (...)
‘... an update after every completely uploaded picture ...’
(14g) Manche Daten, wie upgeloadede Images etc. (...)

‘Some data, such as uploaded images etc. ...’
(14h) (...) auf der sich dann die upgeloadeden Dateien befinden (...)
‘... where the uploaded files can be found ...’
Table 23. ge- as prefix (geuploadet/d; gedownloadet/d), inflected forms

up down
inflected % inflected % inflected % inflected %
forms on - forms on forms on - forms on
t -d t -d
2003 64 98% 1 2% 628 98% 10 2%
2006 3,198 99% 30 1% 38,111 97% 1,048 3%
(15a) Ein einmalig ebenfalls gedownloadetes Programm überprüft (...).

‘Another program, downloaded as well, controls ...’
(15b) (...) Filesystem mit gedownloadeten PDF-Dateien (...)

‘... a file system of downloaded pdf files ...’
(15c) (...) keinerlei Garantie für die hier gedownloadede Software!

‘... no guarantee whatsoever for the software downloaded from here!’
(15e) Die hier gedownloadeden Referate und Hausaufgaben (...).

‘The presentations and homework downloaded from here ...’
(15d) (...) auf jedes geuploadete Bild (...)

‘... on each uploaded picture ...’
(15e) (...) bei der von diesem Modell geuploadeten Software (...)
‘... with the software uploaded by this model ...’
342 Elke Hentschel
(15f) (…) der von meiner Platte geuploadeden Files (…)

‘... of the files uploaded from my hard disk ... ’
At least with regard to the orthography of inflected forms, a relatively sta-

ble standardization seems to have developed.
4. Conclusion: The (non-)emergence of new standards
Considered in toto, the data collected in this survey lead to the astonishing
conclusion that the integrative process underlying the morphological and
orthographical standardization of Anglicisms into German is much less
straightforward than has been supposed so far (cf. e.g., Eisenberg 2001).
The following tables 24–28 show a recapitulation of the findings for a total
of 228,157 (2003) respectively as many as 4,804,165 (2006) past partici-
ples, inflected as well as uninflected ones. Only the results for telnet(t)en
have been omitted, since there were only 16 participles found altogether,
which cannot be considered a representative sample.
Let us first consider the case of forwarden. As has already been stated
above, the users prefer the German spelling and thereby mirror the assimi-
lation of the word itself into the German language, whenever there is an
inflectional ending to the participle. However, even if we take all partici-
ples together, the trend does not go in the direction of a better integration.
On the contrary, it shows a growing use of the English spelling – even
though, on the other hand, the morphological integration as mirrored by the
use of -ge- as an infix shows a slight increase. Both changes should not be
considered as significant per se, but taken together, we get the disturbing
picture of a growing insecurity rather than the impression that the users
take the use of these forms for granted.
Table 24. forwarden
2003 % 2006 %
forms on -t, sum 532 61% 21209 53%
forms on -d, sum 347 39% 19,084 47%
-ge- inserted, sum 32 4% 3,801 9%
ge- prefixed, sum 847 96% 36,492 91%
Upgraden shows a somewhat similar picture, although in this case, the

majority remained where it was three years before. However, instead of
becoming more pronounced and thereby mirroring the more and more fre-
quent use of the forms as becoming a matter of course, the orthographic as

well as the morphological integration again decreased by 4% and 7%.
Again, this is not exactly what one would have expected.
Table 25. upgraden
2003 % 2006 %
forms on -t, sum 1300 64% 62999 60%
forms on -d, sum 732 36% 42,638 40%
-ge- inserted, sum 1599 79% 76,576 72%
ge- prefixed, sum 433 21% 29,061 28%
In the case of updaten, the facts are even more against expectation. The
morphological integration decreased distinctively (by 17%), and the ortho-
graphical one still shows a loss of 9%:
Table 26. updaten

2003 % 2006 %
forms on -t, sum 33,301 69% 848,916 60%
forms on -d, sum 15,154 31% 574,335 40%
-ge- inserted, sum 35,100 72% 788,131 55%
ge- prefixed, sum 13,355 28% 635,120 45%
Finally, in 2006 the verbs uploaden and downloaden show as much as 40%
less of morphological integration than they did in 2003, mirrored by a raise
from a minority of 17% to a majority of 57% of the forms using the prefix.
There is a smaller but with 17% still very noticeable decrease in terms of
orthography, too.
Table 27. up/downloaden

2003 % 2006 %
forms on -t, sum 32,600 78% 509,823 61%
forms on -d, sum 9,391 22% 329,628 39%
-ge- inserted, sum 34,959 83% 358,664 43%
ge- prefixed, sum 7,032 17% 480,787 57%
Although the tendencies to be observed were not identical with all verbal
forms that have been collected, the overall tendency is quite clear, as sum-
marized in Table 28:
344 Elke Hentschel
Table 28. All verbs considered

2003 % 2006 % difference
forms on -t, sum 67,733 73% 1,442,947 60% –13%
forms on -d, sum 25,624 27% 965,685 40% +13%
-ge- inserted, sum 71,690 77% 1,227,172 51% –26%
ge- prefixed, sum 21,667 23% 1,181,460 49% +26%
sum of forms collected 93,357 2,408,632 +2480%
When inserting English verbs with accented prefixes similar to German

ones into a German text, the tendency of the language users to treat the
participle-forming element -ge- as an infix rather than a prefix did not start
out equally strong with all verbs. However, in all cases the same develop-
ment can be observed: with increasing frequency the integration of these
forms, instead of becoming stronger, became distinctively weaker. The
same, if somewhat less pronounced phenomenon can be observed with the
orthographic findings. Again, the orthographical integration of the partici-
ple, expressing itself in the use of German -t instead of English -d, did not
increase, but decrease with the growth of forms used. This decrease oc-
curred notwithstanding the fact that some of the verbs in question
(downloaden, upgraden, updaten) have found their way into the German
reference book Duden (2006), where they are rendered as completely inte-
grated (i.e., with infixed -ge- and spelled with -t). The fact that the ortho-
graphical de-integration is not stronger is only due to the fact that with
inflected forms, the users tend to write the -t- they speak anyway (since, as
has been mentioned before, German has no voiced stops in the coda), so
the influence of the English role model, so to speak, is less pronounced in
these cases.
These results are puzzling, to say the least. Standardization, or so it
seems, can neither be foreseen nor does it follow the rule that increasing
frequency leads to higher integration (and thereby standardization). Obvi-
ously, additional, so far unknown factors play an important role in the de-
velopment of new standards for foreign and loan words.
Notes
1. The additional question of whether such compounds should be hyphenated – a

question that also relates to the following examples bus-snooping and bus-
watching, shall not be dealt with here. For a more detailed discussion of this
point cf. Augst (1992).
2. I wish to thank Beat Siebenhaar for entrusting me with the data. For a more
complete analysis of these, see Siebenhaar (2003).
3. In standard German, verbs with unseparable prefixes as well as those ending
on -ieren do not take ge-. In Alemannian dialects like the ones in the Swiss
examples given above, the prefix ge- may sometimes be omitted with other
verbs, too (cf., e.g., Marti 1985: 132 for details).
4. This is why many linguists prefer the term “verb particle” in order to refer to
them. The prefix ein- is, however, an exception to this rule, since it does not
occur independently. This points to the fact that these morphemes are not as
independent as they may seem at first sight, which is why the term “prefix”
rather than “verb particle” is used in this article.
5. Again, the abbreviation is a Unix command. A search machine query in 2003
brought 18 findings for a German verb ssh-en, all of them morphologically
marked by a hyphen. In July 2006, there were already 368 cases of unmarked
use (sshen) along with 193 hyphenated forms or forms with an apostrophe
(ssh-en, ssh’en).
6. The past tense is omitted, since it is rarely used in modern colloquial German;
but according to the participle, it should probably rather be gradete up than
upgradete.
7. One user showed two different formations of the participle within the same
text (both of them spelled with -t), one directly following the other: Zuerst
aber noch zu den neuen und geupdateten Artikeln. Upgedatet sind (...) Obvi-
ously, s/he could not decide which should be the standard form.
References
Augst, Gerhard
1992 Die orthographische Integration von zusammengesetzten Anglizis-
men. Sprachwissenschaft 17: 45–61.
Becker, Tabea and Corinna Peschel
2003 ‘Wir bitten Sie das nicht misszugeneralisieren’. Sprachverhalten in
grammatischen Zweifelsfällen am Beispiel trennbarer und nicht-
trennbarer Verben. In Sprachliche Zweifelsfälle. Theorie und Em-
pirie / Grammatical uncertainties. Theoretical and empirical as-
pects, Wolf Peter Klein (ed.), 85–104. (Linguistik online 16 (4),
www.linguistik-online.com/16_03).
Dudenredaktion (eds.)
2006 Duden. Die deutsche Rechtschreibung. Mannheim: Dudenverlag.
Eisenberg, Peter
2001 Die grammatische Integration von Fremdwörtern. Was fängt das
Deutsche mit seinen Latinismen und Anglizismen an? In Neues und
346 Elke Hentschel
Fremdes im deutschen Wortschatz: Aktueller lexikalischer Wandel,

Gerhard Stickel (ed.), 183–209. Berlin/New York: de Gruyter.
Hentschel and Weydt
2003 Handbuch der deutschen Grammatik. Third edition. Berlin/New
York: de Gruyter.
Jones, Daniel
2006 Cambridge English Pronouncing Dictionary. Peter Roach, James
Hartman, and Jane Setter (eds.). 17th edition. Cambridge: Cambridge
University Press.
Langner, Heidemarie
1995 Die Schreibung englischer Entlehnungen im Deutschen: Eine Unter-
suchung zur Orthographie von Anglizismen in den letzten hundert
Jahren, dargestellt an Hand des Dudens (Theorie und Vermittlung
der Sprache 23). Frankfurt a. M.: Lang.
Lee, Jinhee
1996 Die graphematische und morphologische Integration von Fremd-
wörtern im Deutschen: Untersuchungen anhand von Wörterbüchern
des 19. und 20. Jahrhunderts. (Dissertation Universität Erlangen).
Mangold, Max
2005 Das Aussprachewörterbuch. Duden. Vol. 6. Sixth edition. Mann-
heim: Dudenverlag.
Marti, Werner
1985 Berndeutsche Grammatik für die heutige Mundart zwischen Thun
und Jura. Bern: Francke.
Pittner, Karin
2001 Deutsch – eine sterbende Sprache? In Beiträge zu Sprache und
Sprachen 3. Vorträge der 6. Münchner Linguistik-Tage, Karin Pitt-
ner and Robert J. Pittner (eds.), 229–237. München: Lincom Europa.
Röhrenbeck, Ulrike
1988 Dottern, dippen oder strippen: Anglizismen in der deutschen Medi-
zinersprache. München: Medikon.
Schmitt, Peter
1985 Anglizismen in den Fachsprachen: Eine pragmatische Studie am
Beispiel der Kerntechnik. Heidelberg: Winter.
Schütte, Dagmar
1996 Das schöne Fremde: Anglo-amerikanische Einflüsse auf die Sprache
der deutschen Zeitschriftenwerbung. Opladen/Wiesbaden: West-
deutscher Verlag.
Siebenhaar, Beat
2003 Sprachgeographische Aspekte der Morphologie und Verschriftung in
schweizerdeutschen Chats. In Chatforschung/Chat analysis, Elke
Hentschel (ed.), 125–140. (Linguistik online 15 (3), www.linguistik-
online.com/15_03).
Zifonun, Gisela
2000 Grammatische Integration jugendsprachlicher Anglizismen. Der
Deutschunterricht 4: 69–79.
Part III
Issues on politeness and impoliteness
Chapter 17
(Im)politeness in English as Lingua Franca
discourse
Juliane House
1. Introduction
In his influential social model of politeness, Richard Watts has pointed to

the importance in politeness research of also “offering ways of recognising
when a linguistic utterance might be open to interpretation by interlocutors
as ‘(im)polite’” (2003: 143). He has linked his notion of “politic behav-
iour”, which participants construct as appropriate to an ongoing interac-
tion, to Bourdieu’s (1985) concept of “habitus”, such that politic behaviour
will be consistent with the disposition of the habitus in accordance with the
relevant social context. In this paper I want to apply these seminal ideas to
a context which has also been one of the many interests of Richard Watts
(see Watts 2001): English as a lingua franca (ELF) as it is increasingly
used in interactions between members of different linguistic and cultural
backgrounds.
A very large part of international communication today is conducted in
a lingua franca, and the majority of lingua franca interactions worldwide
take place in English. Over the past decades, English has consistently
grown in international importance and is now well established in areas
such as politics, business negotiations, cultural and scientific events, and
wherever members of smaller, “exotic”, “less commonly taught” languages
have occasion to talk to members of larger language communities, who see
no reason to acquire and converse in the “smaller language”. English is the
language whose spread has been furthered by recent unprecedented devel-
opments in information and communication technologies, and by the recent
economic developments towards globalisation and internationalisation. For
example, although in business mergers an incoming company may not have
its headquarters in an English-speaking country, as was the case for in-
stance with Daimler-Chrysler, such joint ventures typically adopt English
as their working language which, in turn, promotes a local need for Eng-
lish.
352 Juliane House
ELF interactions are interactions in English between members of two or

more different language communities, for none of whom English is the
mother tongue. Not only are such interactions already extremely frequent
today, they will also increase exponentially in the future. Given this fore-
cast, it seems necessary to examine such interactions more closely. In what
follows I will report on three case studies which I have conducted in the
past 15 years or so, which deal with ELF interactions. I will look at this
work with the research question in mind of whether one can say with rea-
sonable certainty that ELF speakers are – judged by a native English
speaker norm – impolite speakers, i.e. whether and if so how ELF speakers
construct (im)polite behaviour in interactions, i.e. behaviour that is not
deemed to be appropriate in a given situation.
2. Three Case Studies of ELF talk
2.1. Case Study 1
The data for the first case study (see House 1999) is taken from a larger
corpus of ELF data collected inside a PhD study by Agnes Lesznyak
(2004) under my supervision. It was elicited during an international stu-
dents’ meeting in the Netherlands, in which a simulation was enacted in-
side the conference program. The task set to participants in this game was
to arrive at a decision about which of the member states of the European
union qualify on various criteria to take part in the third phase of the Euro-
pean Economic and Monetary Union. Participants included speakers of
Dutch, Hungarian and German. The group discussion was taped and tran-
scribed. The data is quasi-authentic and was elicited via participant obser-
vation. As reported by Lesnyak, and as can be gleaned from the tapes, par-
ticipants appeared to forget quickly that a taped simulation was being
enacted.
I analysed a 30-minute segment of this data using the categorial frame-
work developed by Willis Edmondson and myself (e.g. Edmondson 1981;
Edmondson and House 1981). The analysis was guided by the working
hypothesis that misunderstandings in ELF communication are conceptual,
i.e. derive from different culture-conditioned knowledge schemata. This
would appear to be the premise hiding behind most studies of intercultural
communication including much of my own work. However, this hypothesis
was not confirmed. Instead, the most common communicative “distur-
bances” appeared to be simply inherent in the nature of group interaction,
(Im)politeness in English as Lingua Franca discourse 353
and all minor local routine disruptions, which, interestingly, appear not to
be perceived as problems, are allowed to stand. Compare extract (1):1
(1) 1 D2: Germany should be in (?) no question about that at the
2 D5: do we (?)
3 G: (vocalisations - trying to get the floor)
4 D1: we all agree
5 G: the
6 D4: yeah I
7 D5: but we have to talk about that like if you have a debt
and it’s (?) the right direction
8 D1: yeah yeah okay
9 H2: yeah but if you look at France for example it’s the
same phase that’s the same phase
10 D1: okay but (?)
11 H2: because (?) so and it’s just the question of time
12 D6: yeah but it’s still (?)
13 D1: okay wait (?) maybe it’s smart and (?) useful now to discuss if
there is a trend towards the
14 D2: but there’s (?) one direction of the debt
15 H2: no but
16 D1: no but in general (?) because you’ve also mentioned
France
17 H2: but (?) we can also say of Germany that which is im
more important that uh they should be in by any any means or or
that their debt is a little bit over and shows the wrong di direction
(0.2) we are to decide this I think so if we say that its its
18 D5: yeah but we
19 D6: although
20 D5: can we make a remark on it so we say Germany has
to be in we have to be sure
21 D1: yeah so the main reason
22 D5: as long as it’s getting up all the time
In extract (1) the interactants seem to be geared towards achieving a certain

level of “normality” and consensus. If there are communicative inadequa-
cies in the ELF interaction, these are certainly never foregrounded, but
rather ignored. The talk is “robust”, interactants heed the “let-it-pass” prin-
ciple, to use Firth’s (1996) terminology, and the repeated interruptions and
non-sequiturs, e.g. in turn 12 to13, 15 to 16, or 19 to 20 do not lead to open
conflict. However, we can see that interactants frequently engage in non-
aligned, “parallel talk”, they “play their own game”, without seeming to
register what their interlocutors are actually saying. There also seems to be
354 Juliane House
a lack of mutual orientation. Mutual orientation is, however, the most basic
social alignment between speaker and hearer. Hence in this specimen of
English as lingua franca interaction, hearers are not treated as true co-
participants. Often participants seem to act neither as responders nor as
repliers (in Goffman’s (1976) sense), but only as initiators (cf. House
[1996] for similar findings with German learners of English as a foreign
language).
A further characteristic of this data is the interactants’ failure to appro-
priately preface disaffiliative action. Interactants prefer to use raw nega-
tion, rejection and disagreement, undiplomatically shunning delay, mitiga-
tion and other strategic means of smoothing their negative reactions to their
interlocutor’s previous move. Interpersonally active gambits such as ap-
pealers or cajolers as well as discourse strategies such as pre-exchanges,
disarmers, sweeteners, imposition minimizers (as defined in Edmondson
and House 1981) are markedly absent from this ELF data. Instead, many
turns are either devoid of any introductory lubricating element – such as
receipts (I see), starters (well), or represents (repeating a previous turn’s
wording): Turns are often simply prefaced with and, but, no, or so, I think
or I don’t think. The nakedness of the resulting turn initiations is particu-
larly obvious in (2):
(2) 1 D1: okay you were also discussing Austria because if we are uh meas-
uring Germany with
2 H3: what about Italy
3 D1: so some criteria (?) for Austria (2sec)
4 G: but uh
5 D4: no I think it’s good one that the I think it’s a good reason she said
6 D1: no but
7 D4: that the no because Austria I uh I think there is not a real
8 D1: no no
9 D4: argument their argument it’s just their their policy I think it’s a
good one
10 D5: but I don’t think it’s for Germany
11 D2: I I don’t think we have to give reasons for sticking to the admis-
sion
12 D1: no I
13 D5: I think I think maybe a good point for Germany too because it’s
reconstructing Central and Eastern Europe
14 H2: yeah but
15 D1: no but Germany already
16 D6: we already mentioned that...
Can we now say that the features of ELF talk observed in (1) and (2) above
(as well as other features discussed in House 1999) point to the interac-
tants’ lack of politeness? If we look upon politeness with Watts as a dis-
cursive concept, we would for instance need to take account of partici-
pants’ management of turn taking. Here we can see that there is very little
recognition of sensible points of transition and of the necessity of aligning
one’s behaviour as speaker with the behaviour of one’s hearer, the result
being an under-attuning of individual turns that leads to, or is the outcome
of, a lack of mutual responsibility for the ongoing talk as a collective un-
dertaking. Since recipient design in this data is remarkably “rough”, we
might reasonably expect conflict or mutual antagonism to be the outcome
of exchanges such as the ones in data extracts (1) and (2), as no attempt is
made to pre-empt or avoid conflict. However, as the interactants’ behav-
iour is apparently heavily overlaid by the let-it-pass principle, open conflict
is not allowed to break out. Given this successful conflict-avoidance be-
haviour, we might discard the impoliteness verdict and rather attempt to
explain ELF interactants’ behaviour with reference to the English Native
Norm Irrelevance Hypothesis, which simply says that in the context of
ELF communication, L2 (i.e. native English) linguaculture-specific linguis-
tic behaviour is perceived to be interactionally and communicatively ir-
relevant. The let-it-pass principle becomes, therefore, both a hearer-
oriented communicative strategy, and also a self-defensive mechanism.
The irrelevance of L1 linguaculture-specific norms means, in a nutshell,
that face is saved all round: it is thus less a sign of impolite behaviour but
rather a strategy of politeness in the sense that the behaviour of these ELF
speakers seem to be appropriate to this communicative situation character-
ised by a group of co-participants with widely differing linguistic and cul-
tural backgrounds.
ELF interactants do not seem to seek to adjust to some real or imaginary
L2 native speaker norms. Rather, they seem to simply act as individuals.
Following Henri Tajfel’s (1981) continuum of interpersonal and group
identity, and his considerations of group influence and group stereotyping
versus interpersonal and individual feelings and behaviour, I would posit
for the ELF data a focus on interpersonal and in particular individual con-
cerns rather than on group identity. However, this opposition is too sim-
plistic: the group identity dimension is realised in the participants’ willing-
ness to allow other participants equal individual freedom – hence the “let it
pass” phenomenon. Instead of acting as speakers of Hungarian, German, or
Dutch learning and using English as a foreign language, interactants are
individual ELF speakers united in something as vague, fluid and immate-
356 Juliane House
rial as the “community of ELF speakers”, a sort of “intersociety”, which is

always constituted anew in any on-going talk. This community of ELF
speakers can also be described as a “community of practice” (cf. House
2003) in the sense of Wenger (1998) and Eckert and McConnell-Ginet
(1992) – a very appropriate notion in that the constitution of a community
of practice is governed by a joint purpose – to communicate efficiently in
English as the agreed and chosen language of communication without,
however, heeding or being constrained by English native norms.
The English Native Norm Irrelevance-Hypothesis is directly connected
with another explanatory hypothesis which is influenced by Emmanuel
Levinas’ (1961) theory of conversation, in particular his notion of the con-
versationalist as self-oriented speaker and its interpretation by Katarzyna
Jaszolt (1996). According to Levinas, talk-in-interaction takes place not on
the basis of a shared socio-cultural background, but rather directly from
one individual to another. Interaction is interpreted as a relationship be-
tween conversational partners which is doubly dynamic and cannot be de-
scribed using the metaphor of a dance, where one partner leads and the
other one follows. Rather, interaction is a process in which each participant
does his or her own dance, and meaning is created in between them. As an
alternative to the dancing metaphor, communication can be seen as a self-
centred affair, with the speaker leaving the hearer freedom of interpreta-
tion, the freedom to create assumptions. This “selfishness” in talk, which
has been downplayed in mutuality-based theories of conversation (cf. e.g.
Markova et al. 1995) seems to be particularly marked in ELF interaction,
which is why I call the second explanatory hypothesis of ELF speakers’
behaviour the Self-Centred Hypothesis. The self-centred behaviour dis-
played by ELF speakers can be interpreted as an attempt to save their own
face and improve their own interactional performance. But despite this
interactional behaviour, ELF speakers still seem to be engaged in success-
ful relational work in that they manage to “create some form of common
understanding among themselves” (Watts 2001: 255). Their behaviour is
not perceived by co-participants as inappropriate, rather their focus on self
can be seen as shared, strategically employed, and situationally appropriate
politic behaviour.
2.2. Case Study 2
The second case study (House 2002) essentially confirms the results of the
first one described above. The data for this study comes from an ongoing
study of ELF talk conducted at the University of Hamburg. Subjects in this

30 minute ELF interaction are four students at the University of Hamburg
who were asked to interact on the basis of a trigger in the form of a text
and also commented ex post facto on their own productions. They come
from Germany, Korea, Indonesia and China.
The analysis of this data shows a phenomenon reminiscent of the “self-
centredness” found in case study 1 above, which I have here called: Pursu-
ing one’s own agenda. It can be characterised as follows: three of the inter-
actants, the three Asian participants, clearly pursue their own agenda
throughout the 30 minute interaction. (Only the German speaker, Brit, dif-
fers markedly from this discourse behaviour in that she repeatedly tries to
enter into an argumentative dialogue attempting to ‘problematise’ issues by
inserting critical and doubting questions). The Asian participants’ private
agenda is highlighted by the re-occurrence of a limited number of key
words (see below). This striving after one’s own agenda makes the entire
interaction resemble a set of parallel monologues rather than a dialogue –
or rather a tetralogue – because each participant simply follows his or her
individual macro-theme. The result of such self-oriented pursuance of
one’s personal theme is an under-attuning of individual turns, which leads
to, or is the outcome of, a lack of demonstrated responsibility for the ongo-
ing talk as a collective achievement. Whatever coherence is achieved in the
discourse is only recognisable in the individual paths taken by each par-
ticipant.
Data excerpt (3) exemplifies the ‘monologic tracks’ followed by the
three Asian participants in the discourse. Many non-sequiturs that occur
are due to the interactants’ insistence to pull the talk, by hook or by crook,
towards their particular pet topic. In data excerpt (3), we can see how
Mauri’s turns (for reasons of space I have selected only four of Mauri’s
responding moves here) all centre around the economic reason for the
dominance of English, English being the most important global business
language. Mauri’s key word or leitmotif is ‘business’ (half of Mauri’s 24
turns are devoted to this leitmotif):
(3) Wei: I think English is easy to learn than for example German or
Chinese. (3 sec)
Mauri: Yes I think it is about the important meaning of the erm of the
English language is is (2 sec) this is erm business language is
very important but do you agree then that all the people of the
world that they eh would speak English?
----------------------------------------------------------------------------------------------
358 Juliane House
Wei: No they don’t have problem like French French people I think
erm (2 sec) in China student he likes speak learn English and
European languages.
Mauri: They had to they have to they have to learn English because
this is getting more and more important in business world so if
you cannot speak English so you’re lost.
----------------------------------------------------------------------------------------------
Wei: They don’t have tradition to learn sorry to learn German per-
haps Japanese a little bit tradition (laughs) but not German
Mauri: Yes the importance of language meaning depends on business
erm (2 sec) issues (2 sec) the more important the business is-
sue the more you have to learn this language because all the
people use this language if you cannot speak in this language
you lost and you have to so I think it’s begins erm of course
with the colonialism I think too because the history of this de-
velopment how the language in very early period
----------------------------------------------------------------------------------------------
Joy: Yes I think so and non-native norms because there are (2 sec)
a lot of norms language norms and I think it’s very difficult in
reality to understand people from the other countries other
culture backgrounds
Mauri: Also language erm (3 sec) the languages develop itselves so
maybe if you don’t stay in the countries so you cannot erm,
you cannot erm (2 sec) get with this development I think so if
you just use this language English as business language so it
maybe you can survive you can make a deal you can make ne-
gotiation with economic partners but erm
The idiocratic type of coherence between speakers’ turns documented in

data excerpt (3) seems to be the result of speakers’ (monomaniacally) fol-
lowing their very own agenda. This ‘mental pre-occupation’ is indexed
linguistically as explained in what follows.
Participants hardly use any interactively and interpersonally active gam-
bits, i.e. discourse markers and discourse particles. Especially indicative is
the dearth of gambits surrounding turn change, although there are also
markedly few turn-internal gambits – with one remarkable exception:
Represents, a type of gambit used to repeat to oneself either a substantial
part of, or the whole of, the previous move enacted by the interlocutor. The
major functions of a Represent are, as its name suggests, to “re-present”
the previous speaker’s move in order to aid the present speaker’s working
memory in his comprehension and production processes, to provide textual
coherence, signal acceptance and uptake, request confirmation and indicate

to alter that one does not intend to “steal” his turn. Here is an example of
such a Represent:
(4) Brit: And if erm things like Nigerian English, Indian English which
is a sort of variety in itself it should be respected
Mauri: Should be respected
Presumably in an attempt to make up for their failure to use gambits as

conventional devices for lubricating turn changes (by using such turn-
initial uptakers as hmm, yes, well, I see and so on) speakers resort to realis-
ing an abundance of utterance-initial conjunctions such as and and but that
provide less interpersonal, propositional linkage between utterances, which
is reminiscent of the findings in case study 1 above. Consider excerpt (5):
(5) Joy: And you mean that English (2 sec) is really getting important or
taken for the education because the grammar is syntactical erm the
grammar is very [easy]
Wei: [is ea]sy is very easy
Brit: But the pronunciation (2 sec) and also the relationship between
what you read and what you write and what you pronounce so that
is the problem
Joy: I don’t know I don’t know erm whether the English is very [easy].
Brit: [I don’t] think so either. (all laughing) That is sort of propaganda to
push English forward.
Joy: Yes I think so I think so
Mauri: And you feel it is very easy but it is not easy
Especially marked in this ELF talk is the interactants’ idiosyncratic man-

agement of turn-taking which makes speakers’ explicit accounting for their
interlocutors’ expectations and needs (i.e. the discourse’s recipient design)
become rather “rough” as there is little recognition on the part of ELF in-
teractants of sensible points of transition from one speaker to the next
speaker, and the necessity of aligning one’s behaviour as speaker with the
behaviour of one’s hearer. In other words: interactants just start talking
instead of waiting for the best point at which to “jump in” and bring for-
ward their next argument. They appear not to be able to wait for and/or to
project a suitable point of transitional relevance in the non-native language.
Instead of assuming co-responsibility for the turn taking machinery by
demonstrating linguistically that as current speaker one takes (verbally
explicit) account of what one’s interactant has just said or is expecting
360 Juliane House
from oneself at this point in the discourse, participants patently engage in

‘their own thing’ – oblivious of the co-text, so to speak, as we saw in ex-
cerpt (1) .
In sum, both the English Native Norm Irrelevance Hypothesis and the
Self-oriented Hypothesis suggested in case study 1 above seem to be con-
firmed in this study. For this sort of ‘selfishness’ in talk – which has been
downplayed in many theories of conversation (from Grice’s (1975) and
Sperber and Wilson’s (1995) overridingly influential Cooperative and
Relevance Principles to Leech’s (1983) many maxims), but which seems to
be particularly marked in ELF interaction – a mutuality-negating theory of
conversation such as Levinas’ and Jaszcolt’ views described above seems
to be more adequate.
With regard to the question of whether and how ELF talk can be called
(im)polite, we can only repeat the conclusion drawn from case study 1:
Despite their essentially self-oriented behaviour, ELF speakers manage to
establish an appropriate level of relational work, saving both their own and
alter’s face.
2.3. Case Study 3
The third case study is an examination of the behaviour of the gambit you
know in ELF interaction (cf. House 2007). You know has been described in
most of the literature as a basically interpersonal, intersubjective or other-
oriented marker with a potential for signalling politeness. In my own study
I hypothesised that you know would be used differently by ELF interac-
tants: less interpersonally, more self-referenced, highlighting both formula-
tion difficulties and coherence relations in speakers’ own turns.
The data for this study consists of 13 informal ELF conversations be-
tween university students, age range 20–35, each consisting of 4 interac-
tants with a great variety of different L1 backgrounds and an advanced
competence in English. The interactions lasted approximately 30 minutes,
amounting to a total of 6 1/2 hours of recorded talk. The conversations
were triggered by a short provocative article. Participants were also sub-
jected to retrospective interviews, where they were asked to listen to the
recordings, examine the transcription, reflect on and comment on their use
of you know.
The results of this study indicate that (1) you know occurs with much
greater frequency in so-called “considered talk” phases than in phatic
opening, closing and small talk phases; (2) the more fluent a speaker is, the
more he or she will make use of you know and (3) there is a surprisingly
frequent co-occurrence of you know and the conjunctions but, and, be-
cause. When you know co-occurs with these conjunctions, it appears to act
as a re-inforcing or focussing strategy making the connection expressed by
the conjunctions more salient. You know is thus a highlighting, or focus-
sing device, making more salient the adversative, causal and additive rela-
tions of the conjunctions but, because, and and.
Extract (6), in which participants discuss ELF in German universities is
an example of you know co-occurring with but in the sense described
above:
(6) H: No matter how many people speak in the university they some of them
speak really well English but you know the real life it’s different and
you have to learn English
S: yes (ehm)(1s)
A: this institute where you’re working at is this the only possibility to erm to
learn English…
A closer look at how the conjunctions but, because and and function in
their co-occurrence pattern with you know shows that these conjunctions
most frequently signal “externally operating relations”, i.e. those located in
the phenomena that constitute the external context of what is being said
(i.e. not the socio-communicative process that constitutes the speech event
in the forms of interaction between speaker and hearer, which would be
internal) (Halliday and Hasan 1976: 241–243). In Martin’s (1992) ap-
proach such external relations are by and large oriented ideationally (refer-
entially), whereas internal relations are oriented to genre in dialogic, inter-
personal modes (cf. also the similar distinction by Redeker (1990) into
ideational and pragmatic relations, where I would claim that the ELF users
primarily opt for co-marking ideational relations with you know). My claim
here is that in ELF talk you know in its catalyst functions for the conjunc-
tions but, and and because tends to mark primarily ideational relations.
The fact that you know tends to co-occur with conjunctions that signal
experiential relations rather than addressee-related ones can be interpreted
as confirming my hypothesis that you know is used in ELF talk primarily as
a device to highlight those relations signalled by the conjunctions but, be-
cause and and. Such a use is primarily speaker-centred: the speaker uses
you know with a highlighting function (cf. Miller and Weinert 1998). In
terms of the textual functions theme and rheme and given and new, as
viewed in systemic-functional grammar, we might further hypothesise that
362 Juliane House
in the use of you know there is no relationship to anything “given” in the

sense that there “is a point of contact with what you know” (Halliday 1970:
163). Rather there is a linkage with the “theme” in the sense of “this is the
heading to what I am saying”, you know is used as a coherence producing
discourse marker, signalling adversative, causal and additive linkage rela-
tions. You know is used in ELF discourse as a booster or highlighter of
discourse relations.
Further, I found that you know is also used on its own as a relational
phrase, i.e. speakers use you know to indicate implicit conjunctive relations
of addition (subcategorised as elaboration, extension, expansion in the
Hallidayan sense), contrast, opposition, concession and as causal relations
without the co-presence of the cue words and, but and because, such that
the total co-occurrence of you know and coherence relations with or with-
out overt cue phrase amounts to over 80% of all instances of you know in
this ELF data!
Extract (7) is an example of the use of you know as a relational phrase
with no co-presence of another relational phrase, i.e. you know alone sig-
nals a relation, in this case an adversative relation of contrast. Participants
talk about body weight and superslim models.
(7) JY: we never I mean they never talked about or we never talked about how
thin or skinny they were
AZ: hmm
JY: But somehow we still HAD girls who you
know were suffering from anorexia or bulimia they got problems with
their own bodies
AZ: hmm (very softly)
JY: It wasn’t it wasn’t that time I mean ..you know we didn’t ha/have uhm
those girls we didn’t have these/those girls like you know who are
HAILED …
But you know was also found to function in this ELF data as a coherence
marker in a different sense: it is used whenever the speaker is momentarily
“incoherent”, when s/he hesitates, cannot find the right words, fumbles for
the appropriate formulation, and tries to repair his or her misstep using you
know as what Schourup (1985) has called an “evincive”, i.e. a signal re-
vealing the presence of his or her planning difficulties. In such cases, you
know occurs in mid-utterance and also inside nominal, verbal and adverbial
groups, acting at a more local, micro level. This fumbling use of you know
also occurs with great frequency in this ELF data: some 53% of all occur-
rences of you know.
Taken together, these finding show that in ELF talk you know – despite
the overt presence of you as either 2nd person singular or plural personal
pronoun in this construction – does not seem to address any co-participant
or even any group of participants, and no response from the addressees is
expected or given. The original meaning of you know is clearly no longer
virulent, you know is here primarily used to help speakers process and plan
their own output, and to link spans of discourse. The marker you know in
this ELF data has little to do with hearer deixis or a second person perspec-
tive, it has nothing to do with mutual engagement and an interpersonal
perspective. You know is here no more and no less than a strategy where
the only account a speaker takes of the listener would be when s/he en-
gages in what Halliday has called “bullying the listener” (1970: 163) – i.e.
presenting something on the surface as given when it is in fact new (Halli-
day 1967: 211).
The use of you know in this ELF data can also be explained as a case of
de-grammaticalisation or as Watts (2003: 176) has called it as pragmati-
calisation, i.e. the reverse process of grammaticalisation: the effect of prag-
maticalisation being a bleaching of the propositional content of a linguistic
expression to such an extent that it no longer functions as contributing to
the truth value of a proposition but begins to function as a marker indicat-
ing “procedural meaning” in verbal interaction. Watts has called such
pragmaticalised expression EPMs, expressions of proceduralised meaning
– a characterisation which also fits well with my analysis of the function of
you know as routine formula in this ELF data (cf. also Baumgarten and
House [2007] for comparable findings with another EPM).
So you know seems to be used in ELF talk as an expression of proce-
duralised meaning – NOT predominantly to signal some joint activity, or to
indicate incrementing common ground. It does not serve primarily as a
marker of politeness, solidarity, camaraderie, explicating concern for al-
ter’s face wants and face needs. Rather, it seems to be the case that these
ELF speakers, in the interests of procedural, discourse organisational pur-
poses, use you know to monitor their own progression in discourse – a per-
fectly ordinary activity.
As in both case studies 1 and 2 discussed above, ELF speakers, who do
not aim at native English competence, are too concerned with their own
discourse production to be primarily interpersonally oriented. The English
Native Norm Irrelevance Hypothesis and the Self-Centred Hypothesis are
thus confirmed in this third case study as well.
364 Juliane House
3. Conclusion
Can we now conclude on the basis of the evidence of the results of the
three case studies presented above that ELF interactants’ behaviour can be
judged as “impolite”? Following Watts ideas of politic behaviour and the
importance of relational work (cf. also Locher and Watts 2005), one might
say that given the linguistic capital available to ELF speakers for manipula-
tion (Watts 2003: 160), they probably cannot help the type of self-oriented
behaviour conforming to a new ELF norm which my analysis has revealed.
But more than that: ELF speakers apparently engage in such interactional
behaviour for their own and alter’s benefit, to help their own production, to
make the talk robust, to tolerantly let others’ oddities pass. Clearly, such
behaviour cannot be called inappropriate or impolite as it is shared by in-
teractants, and does not lead to conflict and misunderstanding. The ELF
data examined in the three case studies discussed in this paper show that
ELF speakers’ interactional behaviour proceeds within the framework of
politic behaviour, with no adverse effect being noticed by participants.
Notes
1. D, G, H = Dutch, German, Hungarian native speaker respectively. (?) = difficult

to identify, (0.5) = pause of 0.5 seconds
References
Baumgarten, Nicole and Juliane House

2007 Subjectivity in native and non-native discourse. Paper given at the 6th
International Symposium on Bilingualism, Hamburg, June 2007.
Bourdieu, Pierre
1985 The genesis of the concepts of “habitus” and “field”. Sociocriticism 1
(2): 11–24.
Edmondson, Willis
1981 Spoken Discourse. A Model for Analysis. London: Longman.
Edmondson, Willis and Juliane House
1981 Let’s Talk and Talk About It. A Pedagogic Interactional Grammar of
English. München: Urban & Schwarzenberg.
Eckert, Penelope and Sally McConnell-Ginet
1992 Think practically and act locally: Language and gender community-
based practice. Annual Review of Anthropology 21:461–490.
Firth, Alan
1996 The discursive accomplishment of normality. On ‘Lingua Franca’
English and conversation analysis. Journal of Pragmatics 26 (3):
237–260.
Goffman, Erving
1976 Replies and responses. Language in Society 5: 257–313.
Grice, H. Paul
1975 Logic and conversation. In Syntax and Semantics, Peter Cole and
Jerry L. Morgan (eds.), 41–58. New York: Academic Press.
Halliday, Michael Alexander Kirkwood
1967 Notes on transitivity and theme, part 2. Journal of Linguistics 3 (1):
37–81.
1970 Language structure and language function. In New Horizons in Lin-
guistics, J. Lyons (ed.), 199–244. Harmondsworth. Penguin Books.
Halliday, Michael Alexander and Kirwood Ruqaiya Hasan
1976 Cohesion in English. London: Longman.
House, Juliane
1996 Developing pragmatic fluency in English as a foreign language. Rou-
tines and metapragmatic awareness. Studies in Second Language Ac-
quisition 18 (2): 225–252.
1999 Misunderstanding in intercultural communication: Interactions in
English as lingua franca and the myth of mutual intelligibility. In
Teaching and Learning English as a Global Language, C. Gnutz-
mann (ed.), 73–93. Tübingen: Stauffenburg.
2002 Communicating in English as a lingua franca. In EUROSLA Year-
book 2, S. Foster-Cohen (ed.), 243–261. Amsterdam: Benjamins.
2003 English as a lingua franca: A threat to multilingualism? Journal of
Sociolinguistics 7 (4): 556–578.
2007 (Inter)subjectivity in English as a lingua franca discourse: The case
of you know. Paper given at the 10th International Pragmatics Con-
ference, Göteborg, July 2007.
Jaszczolt, Katarzyna
1996 Relevance and infinity: Implications for discourse analysis. Journal
of Pragmatics 25 (5): 703–722.
Leech, Geoffrey N.
1983 Principles of Pragmatics. New York: Longman.
Lesznyak, Agnes
2004 Communication in English as an International Lingua Franca.
Hamburg: Books on Demand GmbH.
Levinas, Emmanuel
1961 Totalité et infini. The Hague: Nijhoff.
366 Juliane House

search 1 (1): 9–34.
Markova, Ivana, Carl Graumann, and Klaus Foppa (eds.)
1995 Mutualities in Dialogue. Cambridge: Cambridge University Press.
Martin, Jim
1992 English Text. System and Structure. Amsterdam: Benjamins.
Miller, Jim and Regina Weinert
1998 Spontaneous Spoken Language. Syntax and Discourse. Oxford:
Clarendon Press.
Redeker, Gisela
Ideational and pragmatic markers in discourse structure. Journal of
Pragmatics 14 (3): 367–381.
Schourup, Lawrence
1985 Common Discourse Particles in English Conversation. New York:
Garland Publications.
Sperber, Dan and Deirdre Wilson
1995 Relevance. Communication and Cognition (Second Ed.). Oxford:
Blackwell.
Tajfel, Henri
1981 Human Groups and Social Categories. Cambridge: Cambridge Uni-
versity Press.
Watts, Richard
Theory, Nikolas Coupland (ed.), 297–320. London: Longman.
Wenger, Etienne
1998 Communities of Practice. Cambridge: Cambridge University Press.
Chapter 18
How to be impolite: Rating offensive strategies
Mercedes Viejobueno, Carol G. Preston and

Dennis R. Preston
1. Introduction
Politeness theories (e.g., Lakoff 1973, 1989; Leech 1983; Brown and Lev-
inson 1987 [1978]) have focused on how communicative strategies are
used in “…establishing and/or maintaining in a state of equilibrium the
personal relationships between the individuals of a social group […] during
the ongoing process of interaction” (Watts 1992: 50). Set against this equi-
librium is impoliteness, which Culpeper, Bousfield and Wichmann (2003:
1545) define as “the use of strategies that are designed to attack face and,
thereby cause social conflict and disharmony”, and Beebe (1995) gives the
following definition of rudeness:
Rudeness is defined as a face-threatening act (FTA) which violates a so-
cially sanctioned norm of interaction for the social context in which it oc-
curs. It is only rudeness if it receives insufficient redressive action to miti-
gate its force or, of course, if it does not occur in a context such as intimacy
or emergency, that would negate the need for redressive action. Conse-
quently, it causes antagonism, discomfort or conflict and results in some dis-
ruption of the social harmony. (Beebe 1995: 159)
Austin (1990: 279) argues that impoliteness is characterized by face attack
acts, i.e., “communicative acts which are injurious to the hearer’s positive
or negative face, and are introduced in a situation which could have been
avoided, but where their inclusion is perceived by the hearer to be inten-
tional”.
Others, however, show that such cooperative behavior is not always the
norm and that impoliteness1 is not unusual in everyday interaction. Speak-
ers so frequently violate the putative social norm of politeness that one
may ask if politeness in a positive sense is a norm at all. Kienpointner
(1997: 251) points out that rudeness cannot be regarded as “the marked,
abnormal and irrational counterpart of politeness”. That impoliteness is a
“universally occurring phenomenon” and that “systematic, rule-governed
368 Mercedes Viejobueno, Carol G. Preston and Dennis R. Preston
rudeness is not absent in ordinary conversation” (1997: 256) suggest that a

descriptive framework is necessary, but those that have been proposed are
incomplete.
1.1. Problems with impoliteness models
Several models, taking Brown and Levinson (1987 [1978]) as their point of
departure, propose impoliteness superstrategies. All of them, however,
consider positive and negative impoliteness on-record but fail to account
for that distinction in off-record attacks, and none has shown convincingly
how impoliteness superstrategies should be ordered with regard to degree
of offense.
In particular, there are inconsistent claims in the literature about the of-
fensiveness of off-record (i.e., ironic or sarcastic) strategies. Following
Brown and Levinson’s treatment of irony as a face-saving strategy, Dews
and Winner (1995) and Dews, Kaplan and Winner (1995) claim that irony
performs a muting function that results in less offense, but other studies
(Colston 1997; Toplak and Katz 2000; Okamoto 2002; Huang 2004) report
that irony is perceived as being more offensive.
1.2. Objectives of this study
The first goal of this work is to offer a taxonomy of impoliteness super-

strategies that accounts for all categories of face attack. We suggest that
such attacks can be oriented to either aspect of face when performed off-
record. Most of the examples given in the literature involve such attacks
oriented to the hearer’s positive face, for example:
(1) You are so mature!! (Said to a person who has been behaving in a very
childish way.)
But there are also examples of sarcastic attacks on negative face (Austin
1990: 289):
(2) No, no – go ahead. White carpet is boring and the red spots really improve
it. (Where the carpet is new, and the hearer has just spilled red wine on it.)
We will distinguish attacks directed to the hearer’s positive and negative

face within the off-record impoliteness superstrategy. Following Culpeper
How to be impolite: Rating offensive strategies 369
(1996) in using the term sarcasm2 for off-record in general, we will use the
term positive sarcasm to refer to an off-record attack on the hearer’s posi-
tive face and negative sarcasm to refer to an off-record attack on the
hearer’s negative face. These impoliteness superstrategies are illustrated in
Figure 1.
Positive impoliteness
On-record
Negative impoliteness
Face Attack
Positive Sarcasm
Off- record
Negative Sarcasm
Figure 1. Superstrategies for performing a face attack
The second major goal of this work is to suggest how these different super-
strategies should be ordered with respect to degree of offense. In doing so,
we also hope to address the controversy about the function of sarcasm or
off-record attacks.
Politeness theories have generally claimed that off-record strategies ful-
fill a face-saving function. According to Brown and Levinson, by doing a
FTA off-record “…the actor leaves himself [sic] an ‘out’ by providing
himself with a number of defensible interpretations; he cannot be held to
have committed himself to just one particular interpretation of his act”
(1987: 211). This face-saving function has been widely supported by other
researchers. Barbe (1995: 10) argues that “... ironic criticism provides a
means to save face for both the speaker and addressee, neither of whom has
to acknowledge the possible criticism when challenged”. Jorgensen (1996:
616) claims that sarcasm may be thought of as softening the threat to the
hearer’s face.
Dews and Winner (1995: 4) extend the idea of sarcasm as a face-saving
strategy into the tinge hypothesis:3 “The evaluative tone of the literal mean-
ing of ironic utterances automatically colors (or tinges) the hearer’s percep-
tion of the intended meaning”. In the case of ironic insults (i.e., saying
something positive to mean something negative), the positive literal mean-
ing tinges and weakens the negative intended meaning (4), although other
studies have suggested that sarcasm is, in fact, harsher. We hope this study
will shed some light on these differing interpretations.
Finally, by investigating two different cultural groups (in the United
States and Argentina), we hope to contribute to an understanding of how
some perceptions of impoliteness may be general and others specific, al-
though we are aware of the limitations of that goal imposed by selecting
only two largely European based groups.
2. Experimental design
We conducted a web-based experiment carried out in two stages, in which

native speakers of American English and Argentine Spanish were asked
about their perceptions of negative and positive face attacks done directly
or indirectly (sarcastically). Since we were also interested in finding out if
there is a relationship between the perception of a face attack and the de-
gree of intimacy between the participants in a conversation, social distance
was introduced as an independent variable.
2.1. The description task
The goal of this first task was to elicit descriptors from actual folk respon-
dents to be used in the rating scales in the second part of the experiment.
By doing so, we felt that the participants in the second task would rate the
stimulus data using terms that they had intuitions about and that were part
of the folk communities’ ordinary means of evaluation (e.g., Preston 1999:
363). It is possible that some inconsistent results found in the literature
about the degree of offense of on-record versus off-record superstrategies
were due to judgments based on terms that the participants did not nor-
mally use to describe the situations presented to them.
Fifty-five native Spanish speakers in Argentina of various professional
backgrounds and seventy-five native English speakers who were Michigan
State University undergraduate students served as subjects for this first
step. They were recruited through personal contacts in Argentina and
through linguistics classes in the USA by the first author. The responses
were given on a website version of the task. The questionnaire consisted of
four conversations between two people, in each of which one person is the
target of a face-attack by the other. This attack was directed towards the
hearer’s positive or negative face so that two of the four conversations
ended in positive face attacks, and the other two in negative ones. These
attacks were done either on-record or off-record. Finally, the participants in
the conversations were presented as being on either close or distant terms
with one another, resulting in the following eight types:
Table 1. The eight conversation types used in the questionnaire
Positive face attack Negative face attack
Close Distant Close Distant
relationship relationship relationship relationship
Direct (on-record) 9 9 9 9
Sarcastic (off-record) 9 9 9 9
Two versions of the questionnaire were constructed consisting of the same

four situations but with the final comment changed. The stories that ended
in a direct attack in one version of the questionnaire (e.g., I can tell that
you don’t know anything about computers) ended in an indirect one in the
other (e.g., I can tell that you really know a lot about computers). Each
participant read only one version of the questionnaire. Both versions were
written in Spanish for the Argentine participants and in English for the
American subjects; the latter is shown in Appendix A. The subjects were
asked to describe in their own words how they perceived the final com-
ment made by the speaker in each of the four conversations.
In selecting the relevant descriptors for the rating task for the second
part of the experiment, we chose the ones most frequently mentioned, only
those with three or more occurrences, and arranged them into semantically
related groups and paired opposites. For example, in English, angry, an-
noyed, mad, and irritated were grouped together as annoying and paired
with not annoying. Eventually, all those descriptors that did not belong to
any group and that had fewer than five occurrences were ignored.
Eleven pairs of scales were obtained. Some were completely provided
by the subjects (e.g., friendly–unfriendly), but for those where only one of
the descriptors was found in the data, we supplied the opposite. Some
terms used in the scales were derived from semantically related descriptors
that referred to the comment rather than to the speaker (e.g., annoying was
derived from annoyed), and others were slightly modified to avoid any
ambiguity in meaning (e.g., criticizing was used instead of critical). The
scales derived from the responses by the Spanish and American English
speakers were very similar and are shown in Table 2.
Table 2. Scales used in the rating task

Spanish English
1 Fastidioso – No fastidioso Annoying – Not annoying
2 Sincero – Falso Honest – Dishonest
3 Agresivo – No agresivo Rude – Not rude
4 Amistoso – No amistoso Friendly – Unfriendly
5 Irónico – No irónico Sarcastic – Not sarcastic
6 Gracioso – Serio Joking – Serious
7 Crítico – No crítico Criticizing – Not criticizing
8 Cortés – Descortés Polite – Impolite
9 Burlón – No burlón Mocking – Not mocking
10 Amable – No amable Nice – Not nice
11 Malicioso – No malicioso Mean – Not mean
2.2. The rating task
Fifty-six Spanish speakers in Argentina of various professional back-

grounds and sixty native English speaking American undergraduate stu-
dents from Michigan State University participated in the second part of the
study. Some had participated in the first part, and they were all again re-
cruited from personal networks of the first author in Argentina and from
linguistics classes at Michigan State University. This second task was also
carried out on a website version. The same four stories used in the first part
of the experiment were also used in this part of the experiment. Following
each story the participants were presented with the eleven scales derived
from the description task and were instructed to evaluate the final comment
on a 5-point scale – e.g., 1 (very rude) to 5 (not at all rude) or 1 (very dis-
honest) to 5 (very honest). The English version of the questionnaire is
given in Appendix B.
After performing a factor analysis and a series of ANOVAs with post-
hoc Tukey tests on the data, the eleven scales were reduced to six variables
in both English and Spanish. This was done according to the following
criteria: first, the scales had to be insignificantly different from one another
as indicated by Tukey post-hoc tests, and second, the non-significantly
different scales had to belong to the same factor group. The reduction re-
sulted in the following variables:
English
Variable 1: Dishonesty, formed by scale 24
Variable 2: Jocularity, formed by scale 6
Variable 3: Rudeness, formed by scales 1, 3, and 11
Variable 4: Mockery, formed by scales 5 and 9
Variable 5: Impoliteness, formed by scales 4, 8, and 10
Variable 6: Criticism, formed by scale 7
Spanish
Variable 1: Falsedad (dishonesty), formed by scales 2 and 65
Variable 2: Maldad (maliciousness), formed by scale 11
Variable 3: Burla (mockery), formed by scales 5 and 9
Variable 4: Agresividad (rudeness), formed by scales 1 and 3
Variable 5: Descortesía (impoliteness), formed by scales 4, 8, and 10
Variable 6: Crítica (criticism), formed by scale 7
In Spanish, scales 2 and 6 fell together, but in English they were separate.
Spanish speakers appear to think that something that is serio (serious) is
also sincero (honest) and that something that is falso (dishonest) is also
gracioso (joking), but it is difficult to explain this difference. Sarcasm is
obviously untrue in the sense that what is said is the opposite of what is
meant, and it is precisely because of this divergence between what is said
and reality that such a comment is very likely to be considered as joking or
not really meant. Thus, the scale gracioso-serio (joking-serious) seems to
be taken by the Spanish speakers to refer to speaking literally or not. On
the other hand, the fact that scales 2 and 6 remained different from each
other in English could mean that English speakers interpreted joking more
in the sense of teasing or of making fun.
A second difference between the two languages involves scale 11,
(mean–not mean). This scale was included in English variable 3 with rude–
not rude and annoying–not annoying, but formed its own variable in Span-
ish. It could be that the Spanish term malicioso has a much stronger nega-
tive connotation than mean has in English. Unfortunately, we cannot deal
here extensively with the details of the translation equivalences in this task.
Finally, the impolite–not impolite and the rude–not rude scales in both
languages remained significantly different from one another, a result sug-
gesting that conclusions about the impoliteness of utterances should not be

taken as directly implying anything about their rudeness.
3. Analysis
After the six new variables were obtained, we selected Rudeness–

Agresividad, Impoliteness–Descortesía, and Criticism–Crítica as the three
most revealing of the general impoliteness that is the object of this study.
We performed a series of ANOVAs for both the English and the Spanish
data to determine if there was a significant difference in the perception of
direct and indirect face attacks (which we shall refer to as form in what
follows), of the face attacked (positive or negative), and of the familiarity
relation between the interlocutors (close or distant).
3.1. English variables results
Rudeness: The perception of rudeness revealed significant interactions for

form by face, (F(1, 195) = 24.05, p<.0001) and form by familiarity, (F(1,
195) = 14.36, p<.0001). As shown in Figure 2, only positive face attacks
were perceived as more offensive when done directly.
Rudeness
5
4.5
4
3.5
Direct
3
Sarcastic
2.5
2
1.5
1
Postive
Positive faceface Negative
Negative face face
Figure 2. Variable 3: Rudeness (Form by Face; 1 = most and 5 = least rude)
Figure 3 shows in addition that direct face attacks were perceived as differ-
ent, but only in distant social relationships, suggesting, perhaps, that people
do not expect to be attacked by someone they do not know well. Contrary

to what other studies have concluded (Dews, Kaplan and Winner 1995),
this result suggests that the perception of rudeness of a direct or sarcastic
face attack depends in part on the social relation between the interlocutors.
Rudeness
5
4.5
4
3.5
Direct
3
Sarcastic
2.5
2
1.5
1
Close Distant
Figure 3. Variable 3: Rudeness (Form by Familiarity; 1 = most and 5 = least rude)
Impoliteness: As shown in Figure 4, there was a significant three-way in-

teraction in the perception of impoliteness, (F(1, 65) = 12.52, p<0.0008),
depending on to whom and to which face the attack is directed. The differ-
ence between making an attack directly or sarcastically was significant
only when oriented to the hearer’s positive face. No such difference was
found for the negative attacks, implying again that form alone does not
generate different perceptions of the offensiveness of the comment.
In the positive face attacks, however, direct comments were rated as
more impolite than sarcastic ones, but this difference was strongest in the
distant relationships. The lack of familiarity between the speaker and the
hearer clearly contributes to the difference in perception of the impolite-
ness of the two types. The direct attack is perceived as very impolite,
whereas the sarcastic one is not perceived as so impolite.
Criticism: Finally, criticism ratings also yielded a significant three-way
interaction of form, face and familiarity, (F(1, 65) =19.85; p<0.0001). Fig-
ure 5 shows that direct comments were rated as more criticizing than sar-
castic ones, but only for positive face attacks in distant relationships. This
difference was not significant in any of the other situations. As with the
impoliteness ratings, the interaction of positive face and distant relation
here resulted in the greatest difference between direct and sarcastic face
attacks.
Impoliteness
5
4.5
4
3.5 Direct
3
2.5 Sarcastic
2
1.5
1
Positive Negative Positive Negative
Positve face Negative Positve face Negative
face face face face
face face
Close Close Distant Distant
Figure 4. Variable 5: Impoliteness (Form by Face by Familiarity; 1 = most

and 5 = least impolite)
Criticism
5
4.5
4
3.5 Direct
3
2.5 Sarcastic
2
1.5
1 Positive Negative Positive Negative
Positve
face face face
Negative Positve
face face face
Negative
face face
Figure 5. Variable 6: Criticism (Form by Face by Familiarity; 1 = most

and 5 = least criticizing)
In summary, the perception of direct and sarcastic face attacks is always

dependent on the face being attacked, the familiarity of the person being
attacked or, in most cases, an interaction of face and familiarity.
3.2. Spanish variables results
Maldad (Maliciousness): There were significant main effects for face, (F(1,
180) = 62.68, p<0.0001) and familiarity, (F(1, 180) = 11.41, p<0.0009) in
the perception of the maliciousness of face attacks. The main effect of face
occurred because positive face attacks (M = 2.8067) were rated as more
malicious than negative ones (M = 4.0158). The main effect of familiarity
occurred because attacks in distant relations (M = 3.1534) were rated as
more malicious than those in close ones (M = 3.6691). Though no interac-
tions were found, the results continue to support the conclusions reached
above about the role of positive face and distant relationships in the per-
ception of the offense of a face attack.
There was, however, no main effect for form. Direct and sarcastic face
attacks were not significantly different with respect to malice. This could
be due to the strong negative connotation that we have already noted for
the word malicioso in Spanish. Because there was no effect for form, this
variable will not be discussed in further detail.
Agresividad (Rudeness): A significant three-way interaction resulted for
agresividad, (F(1, 60) = 6.09, p>0.0165), as shown in Figure 6. Direct face
attacks were perceived as ruder than sarcastic ones in all situations except
for negative face attacks in distant relations. Once again, however, form
was most dramatic for positive face attacks in distant relationships.
Agresividad
5
4.5
4
3.5 Direct
3
2.5 Sarcastic
2
1.5
1
Positive Negative
Positve face Negative Positive Negative
Positve face Negative
face face face face
face face
Close
Close Close
Close Distant
Distant Distant
Distant
Figure 6. Variable 4: Agresividad (Rudeness) (Form by Face by Familiarity;

1 = most and 5 = least rude)
Descortesía (Impoliteness): Figure 7 also shows a significant three-way

interaction for the perception of impoliteness, (F (1, 60) = 7.83, p<0.0069).
Direct face attacks were rated as significantly more impolite than sarcastic
ones, but only for positive face attacks in distant relationships. None of the
other situations yielded significant differences. This result strongly sup-
ports the conclusions reached above about the greater effect that a positive
face attack in a distant relation has in the perception of the impoliteness
and rudeness of the two types.
Descortesía
5
4.5
4
3.5 Direct
3
2.5 Sarcastic
2
1.5
1
Positve face Negative Positve face Negative
face face face face
face face
Figure 7. Variable 5: Descortesía (Impoliteness) (Form by Face by Familiarity;

1 = most and 5 = least impolite)
Crítica
5
4.5
4
3.5 Direct
3
2.5 Sarcastic
2
1.5
1
Positve
face face face
Negative Positve
face faceface
Negative
face face
Figure 8. Variable 6: Crítica (Criticism) (Form by Face by Familiarity; 1 = most

and 5 = least criticizing)
Crítica (Criticism): Finally, the degree of criticism of direct and sarcastic

face attacks also revealed a significant interaction between form, face and
familiarity, (F(1, 60) = 7.86, p<0.0068), as shown in Figure 8, but the only
situation that showed a difference between the two types of criticism was a
positive face attack in a distant relationship.
4. Ranking of the superstrategies in English and Spanish
One of the main goals of this experiment was to obtain a ranking of the
different impoliteness superstrategies with regard to degree of offense.
This was done by ranking the least square means of the four different su-
perstrategies using a least significant difference test.
The results show that all three variables – rudeness, impoliteness, and
criticism – were ranked differently in the two social distance relations.
These results emphasize the important effect that social distance has on the
perception of offense of a criticism.
4.1. Ranking of the superstrategies for English
Tables 3 through 5 show these rankings for English. The order of the su-
perstrategies in a close relation was the same for the rudeness and impo-
liteness variables, although impoliteness was rated harsher in every case,
suggesting that some subtle difference between the two exists. The rank-
ing for the criticism variable in a close relation, however, was quite differ-
ent. The rudeness and impoliteness attacks to positive face were ranked as
more offensive than attacks to negative face, but the ranking for criticism
resulted in the opposite order. Attacks to negative face were ranked as
more criticizing than attacks to positive face.
This difference could be due to the fact that rudeness and impoliteness
are more emotional and relate to how the hearer perceives the comment,
while criticism seems to be mainly related to content or action. Since the
attacks on negative face used in the questionnaire involve the speakers
criticizing the hearers for not doing something they were supposed to do
according to some rule or norm of behavior (e.g., smoking where not al-
lowed, littering in a national park), the degree of criticism conveyed by this
type was perhaps perceived as greater than if the speakers were criticizing
the hearers for something they did not like or merely disapproved of.
For a distant relation, on the other hand, the order of the superstrategies
was identical for the criticism and impoliteness variables. However, the
ranking for rudeness was a little different. It is not clear what causes that
different ranking, but it was still a direct attack on positive face that was
rated as most offensive of all attacks.
Table 3. Ranking of superstrategies for Rudeness
1.00 = most (lighter shading) and 5.00 = least (darker shading) impolite;
means in the same column shaded the same are not significantly different
(Tukey post-hoc)
Close Superstrategy Distant
Mean Mean
2.39 Positive face–on record 2.24
2.86 Positive face–off record 2.89
3.04 Negative face–on record 3.08
3.05 Negative face–off record 3.26
Table 4. Ranking of superstrategies for Impoliteness

(Tukey post-hoc)
Mean Mean
Table 5. Ranking of superstrategies for Criticism

(Tukey post-hoc)
Mean Mean
4.2. Ranking of the superstrategies for Spanish
Tables 6 through 8 show the rankings for Spanish:

Table 6. Ranking of superstrategies for Agresividad (Rudeness)
(Tukey post-hoc)
Mean Mean
Table 7. Ranking of superstrategies for Descortesía (Impoliteness)

(Tukey post-hoc)
Mean Mean
Table 8. Ranking of superstrategies for Crítica (Criticism)

(Tukey post-hoc)
Mean Mean
The results for the Spanish variables also show a ranking for crítica in
close relationships that was very different for the rankings for agresividad
and descortesía. It appears that the same conclusion reached about criti-
cism in English applies to Spanish too.
In distant relationships, however, all three rankings show the same or-
der of the superstrategies. This result provides strong evidence that this is
how the different impoliteness superstrategies should be ordered according
to their degree of offense for a distant relationship.
5. General discussion
The results reported here have shown that the perception of direct and sar-
castic offenses strongly depends on the face that is being attacked, the so-
cial distance between the interlocutors, or the interaction between these
two factors. The presence of significant interactions for all the variables
that relate to the offensiveness of the comments in both English and Span-
ish suggests that earlier claims that sarcasm is less harsh were not com-
pletely accurate since they are not based on a simultaneous consideration
of face and degree of familiarity.
5.1. The offense variables in English
All three English variables that relate to the degree of offense of a face
attack yielded significant interactions: rudeness resulted in two-way inter-
actions of form and face and form and familiarity, while impoliteness and
criticism yielded significant interactions of form, face and familiarity.
Notice first that no difference was found between a direct and a sarcas-
tic face attack for any of the three variables when the attack was oriented to
the hearer’s negative face. This holds true regardless of the social distance
between the interlocutors.
We suspect that the lack of difference derives from the nature of nega-
tive face attacks, since they involve less emotion than ones on positive
face. Criticism oriented to positive face offends sensibilities and feelings
by expressing dislike and disapproval, but attacks on negative face inter-
fere only with the hearer’s freedom of action and can be considered less
emotional or perhaps taken less personally. Furthermore, while attacks on
positive face include attacks on a person’s beliefs, intelligence, body fea-
tures, personal habits, etc., i.e., things that are hard if not impossible to
change, attacks on negative face, i.e., attempts to impose on the hearer’s
freedom of action, can be resisted, perhaps even ignored. Hence, the way in
which a negative face attack is framed (directly or sarcastically) does not
make a difference in the degree of offense conveyed.
Contrarily, positive face seems to play an important role in the different

perception of the offense of the two form types. Its contribution to the in-
teractions found for impoliteness is also reflected in the interaction of form
and face found for rudeness. A direct attack is perceived as ruder than a
sarcastic one only when attacking the hearer’s positive face. This is even
truer in a distant relation than in a close one, as can be seen from the inter-
action for impoliteness in Figure 4 and for rudeness in Figure 3.
These results suggest that greater social distance between the partici-
pants in the conversation increases the perception of rudeness and impo-
liteness of a direct attack. Because one feels that a person one is not closely
related to has no right to criticize, a direct attack sounds very rude.
The interaction of positive face and distant relation also seems to have
an effect on the perception of sarcasm. For the variables of impoliteness
and criticism, a sarcastic comment in a positive face attack in a distant
relation was rated as less offensive than in all the other situations.
Again we suggest that this is due to the lack of shared knowledge be-
tween the participants, resulting in the ambiguity of the sarcasm. Several
researchers (Gibbs 1986; Kreuz and Glucksberg 1989; Dews and Winner
1995; Jorgensen 1996) have stressed the importance of common knowl-
edge in the interpretation of sarcastic irony (and irony in general). In order
to understand sarcastic utterances, the speaker and the hearer must share
knowledge about the situation and the speaker’s evaluation of it.
The findings here also show that in a close relationship, where the true
intention of the sarcasm is more readily understood, there is no difference
in the degree of criticism conveyed by a direct or a sarcastic face attack
and only a very small difference in the degree of impoliteness and rude-
ness.
In general terms, the results for the three variables suggest that the so-
cial distance between participants influences the perception of the degree
of offense of a face attack. These findings contradict the results in Dews
and Winner (1995) which showed that there was no effect for familiarity in
the degree of offense conveyed by the two types.
Finally, it is worth noticing that for all three variables the most dramatic
difference was found for a positive face attack in a distant relation. It is this
interaction that contributes to the different perception of the face attacks by
either intensifying the rudeness and impoliteness of a direct comment or by
reducing the degree of offense of a sarcastic comment.
5.2. The offense variables in Spanish
In Spanish, the variables also resulted in several three-way interactions of

form, face and familiarity. Similar to the impoliteness and criticism vari-
ables in English, descortesía and crítica did not show a difference between
the two form types for negative face attacks. Hence, the same conclusions
reached about negative face above also account for the Spanish data.
Though in English a direct comment was rated as more impolite than a
sarcastic one for a positive face attack in a close relation, there was no
difference between the two types of face attack for this same situation in
Spanish, matching the criticism ratings. It was, however, again a positive
face attack in a distant relation that produced the most dramatic difference
between the two form types of face attacks. For criticism ratings, the dif-
ference between the direct and the sarcastic attacks resulted from the rela-
tively smaller degree of offense that a sarcastic attack triggered in this
situation. This result is similar to the English criticism ratings and can also
be explained by the failure to correctly interpret the sarcastic tone of the
comment due to the lack of shared knowledge between the interlocutors.
Descortesía (impoliteness) and agresividad (rudeness) ratings also
showed an effect of face: a direct attack on positive face was rated as more
impolite and ruder than a direct attack on negative face. But again, it is in a
distant relationship that a direct attack is rated as the most impolite. Thus,
for Spanish too, the social distance between the participants in the conver-
sation is a determining factor in the perception of offense.
At this point it should be noted again that the different results obtained
for impoliteness and rudeness are especially intriguing since the two con-
cepts seem to be very closely related.
As shown above, the offense variables ratings were very similar in both
languages. There is an effect of face in the perception of rudeness and im-
politeness of a direct attack. A direct attack is more offensive when attack-
ing positive face than when directed at negative face. But, most impor-
tantly, there seems to be a key effect given by the interaction of positive
face and distant relationship that contributes to the highly different percep-
tion in the degree of offense conveyed by direct and sarcastic attacks.
6. The function of ironic criticism
To return to the controversy concerning the function of sarcastic irony, the

findings here appear to be more in line with those reported by Dews and
Winner (1995) and Dews, Kaplan and Winner (1995): whenever a differ-
ence between the two types was found, sarcasm was perceived as less of-
fensive. However, the fact that all the offense variables showed at least one
situation where there was no difference in the degree of offense between
the two types casts some doubt on the off-record and face-saving nature of
sarcasm and the validity of the tinge hypothesis.
Sarcasm has generally been considered an off-record strategy because it
allows more than one possible interpretation, but, in our study, only when
it is oriented to the hearer’s positive face in a distant relation does it allow
more than one interpretation. In all other situations, sarcasm appears to
have only one possible interpretation, as the impoliteness and criticism
ratings show. Thus, the on-record or off-record nature of sarcasm seems to
depend on the face that is being attacked and the social distance between
the participants.
The differences between the two types of face attack found for the rude-
ness ratings cannot, then, be attributed to the availability of more than one
interpretation. If this were the case, there should also have been more dif-
ferences in the impoliteness and criticism ratings. It is unlikely that a sar-
castic comment will be ambiguous only when considering its rudeness but
not as regards its impoliteness or criticism.
Though the findings presented in this study may appear to support the
tinge hypothesis (whenever a difference was found between the two types,
a sarcastic attack was perceived as less rude, less impolite, and less criticiz-
ing than a direct attack), we have shown that this is not really the case. The
tinge hypothesis cannot account for the different results obtained for the
three offense variables. Thus, we have argued that the less impolite and
criticizing perception of sarcasm for a positive face attack in a distant rela-
tion results from the ambiguity that sarcasm has in those situations.
7. General ranking of impoliteness strategies
The results for the rankings of the superstrategies for the three variables
have shown that two rankings are needed: one if the relation between the
participants is close and another if the relation is distant. The ranking for
criticism in a close relationship was also different from the ones obtained
for the other two variables in both languages. This difference was ex-
plained in terms of the association of criticism with content.
Interestingly, the rankings for impoliteness and rudeness in a close rela-
tionship were also very similar in English and Spanish. And, the rankings
for a distant relationship were identical in all three variables within and
across languages with the exception of the ranking for rudeness in English,
which was only a little different.
This great similarity between the different rankings across languages
raises the possibility that the order of the impoliteness superstrategies with
respect to degree of offense has some chance of being universal. However,
this is a claim that we do not want to make at this point but prefer to en-
courage future investigation of.
The general rankings of the superstrategies with respect to degree of of-
fense in both social distance relations are presented in Table 9.
Table 9. General ranking of impoliteness superstrategies
Close Distant
Superstrategy Superstrategy
1 Positive face–on record Positive face–on record
2 Positive face–off record Negative face–off record
3 Negative face–on record Negative face–on record
4 Negative face–off record Positive face–off record
Note: the strategies are ranked in order of decreasing offensiveness so that (1)
represents the most offensive of all strategies and (4) the least offensive.
Both rankings show that a direct attack on positive face is the most offen-
sive of all the superstrategies. The main difference between the two rank-
ings is that while a positive face attack is always more offensive than a
negative one in a close relation, a sarcastic attack to positive face is ranked
as the least offensive of all attacks in a distant relation. We believe that this
difference is a result of the hearer’s failure to understand the ironic tone
and the intended meaning of the utterance.
8. Conclusions
This study was principally designed to determine the degree of offense of

different impoliteness superstrategies within the framework of a more com-
prehensive taxonomy. By doing this, we also hoped to contribute to resolv-
ing the controversy in the literature about the muting or intensifying func-
tion of sarcasm compared to direct, literal face attacks.
Our results confirm that consideration of face type and social distance is
a more adequate approach in looking at the relative degrees of offense of
direct and sarcastic attacks. Since the variables that related to the degree of
offense of the face attack resulted in significant interactions, one conclu-

sion to be drawn is that the degree of offense of directness and sarcasm
depends on the face to which the attack is oriented and the familiarity rela-
tion between the participants. This means that no general conclusions can
be reached about whether sarcasm mutes or intensifies the degree of of-
fense of a face attack unless both face and familiarity are looked at.
The findings in this study have also cast doubt on the validity of the
tinge hypothesis, at least as far as sarcastic utterances are concerned. How-
ever, there seem to be some other forms of irony that dilute the degree of
offense. There is no doubt that future research would benefit from identify-
ing what these are. Another question that requires further research is what
may be causing the different results obtained for rudeness and impoliteness
ratings. Moreover, it would also be interesting to see if positive face and
distant relationships also have a significant effect on the perception of of-
fense of the different superstrategies in other languages and cultures, espe-
cially in East Asian cultures where the notion of negative face plays such
an important role in people’s lives (Gu 1990; Matsumoto 1988; Okamoto
2002). Finally, a great deal of further cross-linguistic research will be re-
quired to test the universality of the ranking of the impoliteness super-
strategies presented here.
APPENDIX A
Description task (English)

Instructions
Below you will find four brief stories describing a situation between two partici-
pants. Each situation ends with a comment or remark made by one of the partici-
pants and directed towards the other participant. Please describe with words,
phrases and/or short sentences this last comment or remark in the space provided
below. For example, if you think the comment is friendly, simply write “friendly”;
if you think that it shows that the speaker is in a good mood, write “the speaker is in
a good mood.”
Friendly. The speaker is in a good mood.
Situation 1
Robert has just bought a new computer. After installing it and putting it to work, he
realizes that there is one program that is not responding. Since he doesn’t know
how to solve this problem, he calls his friend Mike. Mike is constantly bragging
that he knows everything about computers. Mike tries to solve the problem but
cannot do it. So, Robert says to him:
“I can tell that you really know a lot about computers!!”
“I can tell that you don't know anything about computers!!”
Situation 2
Mary and Julie are good friends. For their summer vacation, they went camping for
a week in the Glacier National Park. At the entrance of the park there was a notice
reminding them that littering was not allowed. While they were hiking on the trails,
Mary threw her empty bottle of water in the bushes. Julie, who was watching Mary,
said to her:
“You have no respect for the environment!!”
“You have such great respect for the environment!!”
Situation 3
John and Steven are coworkers. They have just started to share an office. In the
company building, smoking is not allowed. Steven, however, who is a chain-
smoker, opens the window in their office and lights a cigarette. John sees this and
says to him:
“I see you are the kind of person who follows the rules”
“I see you are not the kind of person who follows the rules”
Situation 4
Liz and Laura were assigned to work on a project by their professor. It’s the first
semester of school and they don’t know each other well. Liz is a dedicated student
and wants to do well on the project. Laura is not as good a student as Liz but she
understands the material better. They decide to each research a part of the assign-
ment and meet after one week to discuss what they have done. When Laura reads
Liz’s part, she thinks it is terrible and says:
“You did a very bad job!!”
“You really did a great job!!”
APPENDIX B
Rating task (English)

Instructions
Below you will find four brief stories describing a situation between two partici-
pants. Each situation ends with a comment or remark made by one of the partici-
pants and directed towards the other participant.
Please use the 1 to 5 scales given below each situation to rate how you would inter-
pret the remark made by the speaker.
An example is given below for your reference.
If you consider the remark to be very interesting, select 1.

If you consider the remark to be somewhat interesting, select 2.
If you consider the remark to be neither interesting nor boring, select 3.
If you consider the remark to be somewhat boring, select 4.
If you consider the remark to be very boring, select 5.
Interesting Ο Ο Ο Ο Ο Boring
1 2 3 4 5
The situations are as in Appendix A.
Notes
1. Both impoliteness and rudeness are used in the literature. For the sake of par-
allelism with the politeness literature, we adopt the term impoliteness through-
out to refer to this general type of verbal behavior, although we will use both
terms later to refer to specific variables found in this study.
2. Though sarcasm is not the only indirect way to attack a person’s face (other
indirect uses of language that attack the interlocutor include understatement,
hyperbole, and rhetorical questions), sarcasm is probably the most common.
Therefore, we use that term to refer to the off-record strategy.
3. Dews and Winner (1995) claim that the tinge hypothesis is able to accommo-
date not only ironic criticism but also the opposite, i.e., ironic compliment, the
surface-criticism-plus-underlying-compliment form of verbal irony. We com-
ment here only on their arguments for off-record attacks.
4. These scales are the ones used in the rating task in English and are shown in
Table 2.
5. These scales are the ones used in the rating task in Spanish and are shown in
Table 2.
References
Austin, Paddy
1990 Politeness revisited – the dark side. In New Zealand Ways of Speak-
ing English, Allen Bell and Janet Holmes (eds.), 277–293 (Multilin-
gual Matters 65). Philadelphia: Multilingual Matters.
Barbe, Katharina
1995 Irony in Context. Amsterdam/Philadelphia: John Benjamins.
Beebe, Leslie M.
1995 Polite fictions: Instrumental rudeness as pragmatic competence. In
Linguistics and the Education of Second Language Teachers: Ethno-
linguistic, Psycholinguistic and Sociolinguistic Aspects, James E.
Alatis, Carolyn A. Straehle, Brent Gallenberger and Maggie Ronkin
(eds.), 154–168 (Georgetown University Round Table on Languages
and Linguistics). Washington, D.C.: Georgetown University Press.
Brown, Penelope and Stephen C. Levinson
1987 Politeness: Some Universals in Language Usage. Cambridge: Cam-
bridge University Press. Original edition 1978.
Colston, Herbert L.
1997 Salting a wound or sugaring a pill: The pragmatic functions of ironic
criticism. Discourse Processes 23 (1): 25–45.
Culpeper, Jonathan
1996 Towards an anatomy of impoliteness. Journal of Pragmatics 25 (3):
349–367.
Culpeper, Jonathan, Derek Bousfield and Anne Wichmann
2003 Impoliteness revisited: With special reference to dynamic and pro-
sodic aspects. Journal of Pragmatics 35 (10–11): 1545–1579.
Dews, Shelly, Joan Kaplan and Ellen Winner
1995 Why not say it directly? The social functions of irony. Discourse
Processes 19 (3): 347–367.
Dews, Shelly and Ellen Winner
1995 Muting the meaning: A social function of irony. Metaphor and Sym-
bolic Activity 10 (1): 3–19.
Gibbs, Raymond W. Jr.
1986 On the psycholinguistics of sarcasm. Journal of Experimental Psy-
chology 115 (1): 3–15.
Gu, Yueguo
1990 Politeness phenomena in modern Chinese. Journal of Pragmatics 14
(2): 237–257.
Huang, Chun
2004 On the function of ironic criticism. Masters thesis. Department of
Linguistics and Germanic, Slavic, Asian and African Languages,
Michigan State University.
Jorgensen, Julia
1996 The functions of sarcastic irony in speech. Journal of Pragmatics 26
(5): 613–634.
Kienpointner, Manfred
1997 Varieties of rudeness: Types and functions of impolite utterances.
Functions of Language 4 (2): 251–287.
Kreuz, Roger J. and Sam Glucksberg

1989 How to be sarcastic: The echoic reminder theory of verbal irony.
Journal of Experimental Psychology 118 (4): 374–386.
Lakoff, Robin
1973 The logic of politeness; or, minding your P’s and Q’s. In Papers
from the Ninth Regional Meeting, Chicago Linguistic Society, Clau-
dia Corum, T. Cedric Smith-Stark and Ann Weiser (eds.), 292–305.
Chicago: Chicago Linguistic Society.
1989 The limits of politeness: Therapeutic and courtroom discourse. Mul-
tilingua 8 (2–3): 101–129.
Leech, Geoffrey
1983 Principles of Pragmatics. London: Longman.
Matsumoto, Yoshiko
1988 Reexamination of the universality of face: Politeness phenomena in
Japanese. Journal of Pragmatics 12 (4): 403–426.
Okamoto, Shinichiro
2002 Politeness and the perception of irony: Honorifics in Japanese. Meta-
phor and symbol 17 (2): 119–139.
Preston, Dennis R.
1999 A language attitude approach to the perception of regional variety. In
Handbook of Perceptual Dialectology: Volume One, Dennis R. Pre-
ston (ed.), 358–373. Amsterdam/Philadelphia: John Benjamins.
Toplak, Maggie and Albert N. Katz.
2000 On the use of sarcastic irony. Journal of Pragmatics 32 (10): 1467–
1488.
Watts, Richard J.
1992 Linguistic politeness and politic verbal behaviour: Reconsidering
Ehlich (eds.), 43–69. Berlin: Mouton de Gruyter.
Contributors
D. J. ALLERTON is emeritus professor of English linguistics at the Univer-
sity of Basle (Switzerland), where he was professor from 1980 till 2003.
He had previously been (senior) lecturer in general linguistics at the Uni-
versity of Manchester (U.K.). His publications include Essentials of Gram-
matical Theory 1979, Valency and the English Verb 1982, Stretched Verb
Constructions in English 2002, as well as over 50 articles. His work has
centred on grammar and phraseology, but has ranged widely across seman-
tics, pragmatics, text linguistics, sociolinguistics and phonetics. Another of
his current interests is graphemics.
FRANZ ANDRES MORRISSEY was born in Altdorf in 1957, educated in

St.Gallen and Zurich and started working in TEFL during his studies and
managing a linguisitic consultancy until 1992. He specialised in applied
linguistics and sociology of language before developing a teaching interest
in phonetics and phonology. Besides teaching academic and creative writ-
ing at Berne University, he is a happy family man, has performed poetry in
places from Baltimore to Yerevan (not quite A to Z) and played in various
rock and folk groups invariably distinguished by almost total obscurity.
ANITA AUER is a Junior Lecturer in the English department at the Univer-

sity of Leiden. There she is involved in a research project called “The
Codifiers and the English Language”, funded by the Dutch national re-
search organisation NWO. She received her Ph.D. in English and German
historical linguistics at the University of Manchester. Her research interests
include (socio-)historical linguistics, corpus linguistics, language change,
stylistics, prescriptivism, comparative standardology and the interface be-
tween manuscript and print.
TONY BEX retired as a Senior Lecturer in English Language and Linguis-

tics at the University of Kent in 2001. Prior to that, he worked in Papua
New Guinea, Saudi Arabia, Sudan and Algeria. He has published widely
on aspects of the English language and maintains his interest in both the
language and teaching by examining for Trinity College and working in a
local language school.
URS DÜRMÜLLER, Prof. Dr. phil. habil., is the head of the English Division
at the PH Bern Teacher Training College, Bern University of Applied Sci-
394 Contributors
ences, and a lecturer in English Linguistics in the Department of English of

the University of Bern, Switzerland. He has had repeated affiliations with
the Department of Linguistics, University of California at Berkeley (1978–
80, 1982, 1998), and was Visiting Professor in English Linguistics at Al-
bert-Ludwigs-Universität in Freiburg i.Br., Germany in 1988–89.
(www.cx.unibe.ch/~durmull/).
After finishing her PhD thesis at the Free University of Berlin in 1985,
which was assigned the “Ernst Reuter” award, ELKE HENTSCHEL worked at
the University of Belgrade (Yugoslavia). From 1991 to 1997 she was em-
ployed as a scientific assistant in Berlin, and acted as guest professor in
Belgrade and in Aarhus (Denmark). After her postdoctoral lecturer qualifi-
cation at the European University "Viadrina" in Frankfurt/Oder (Germany),
she accepted a professorship at the University of Osnabrück (Germany) in
1998. Since 2000, she has been a professor at the University of Berne
(Switzerland). Elke Hentschel is the founder and has been the editor of
Linguistik Online since 1998. As a consulting editor of Multinlinga, she
has cooperated with Richard Watts since 1989.
JULIANE HOUSE was born in Berlin, Germany. She received her first de-
gree in translation and international law from Heidelberg University and
her PhD in linguistics and applied linguistics from the University of To-
ronto. She is currently Professor of Applied Linguistics at Hamburg Uni-
versity and co-director of the German Science Foundation’s Research Cen-
tre on Multilingualism, where she also co-ordinates the multilingual
communication group. Her research interests include contrastive pragmat-
ics, discourse analysis, politeness theory, English as a lingua franca and
intercultural communication. She has written and edited numerous articles
and books, among them A Model for Translation Quality Assessment,
Translation Quality Assessment: A Model Revisited, Interlingual and In-
tercultural Communication, Cross-Cultural Pragmatics, Misunderstanding
in Social Life, Multilingual Communication.
ADAM JAWORSKI is Professor at the Centre for Language and Communica-

tion Research, Cardiff University. His current research interests include
discursive approaches to media language, tourism, and mobility, as well as
visual and nonverbal communication. He co-edits the book series Oxford
Studies in Sociolinguistics for Oxford University Press (with Nikolas
Coupland).
Contributors 395
MIRIAM LOCHER was inspired by Richard Watts’ work on politeness in the

1990s and has worked on the topic of linguistic politeness and the exercise
of power in oral communication, which resulted in the monograph Power
and Politeness in Action: Disagreements in Oral Communication (Mouton
de Gruyter, 2004) and the edited collection Impoliteness in Language
(Mouton de Gruyter, 2007, together with Derek Bousfield). She has also
worked in the field of health discourse (Advice Online: Advice-giving in an
American Internet Advice Column, 2006, John Benjamins). She holds a
Lizentiat (‘MA’) from the University of Zurich and a PhD (supervised by
Richard) and post-doctoral thesis (Habilitation) from the University of
Berne. Currently, she is working in the English Department of the Univer-
sity of Berne, Switzerland, and considers herself fortunate in having had
the privilege of working together with Richard Watts for several years.
ADRIAN PABLÉ is currently Maître Assistant of English linguistics at the

University of Lausanne. Between 2001–2005 he worked as an assistant for
Richard J. Watts at the University of Berne, where he taught historical
linguistics. He also taught at the University of Neuchâtel. Adrian Pablé
obtained both his MA (in English, Italian and American History) and his
doctorate (in English linguistics) from Zurich University. His main re-
search interests are in historical dialectology, translation studies, linguistic
theory and literary linguistics. So far he has published two books and sev-
eral articles in international journals.
INGRID PILLER (PhD, Dresden 1995) is Professor of Applied Linguistics at

Macquarie University, Sydney, Australia, where she also heads the Adult
Migrant English Programme Research Centre. Before joining Macquarie
University, she taught at the University of Basel, where she co-directed a
research project on languages and identities in the Swiss tourism sector.
Her research interests are in intercultural communication, language learn-
ing, multilingualism, and other language aspects of globalisation. She is
currently writing a textbook on Intercultural Communication for Edin-
burgh University Press, which aims to provide a critical introduction to the
field from a sociolinguistic and discourse-analytic perspective.
DENNIS R. PRESTON is University Distinguished Professor of English at

Michigan State University. He is former President of the American Dialect
Society and Director of the Linguistic Institute of the Linguistic Society of
America. His work focuses on sociolinguistics, dialectology, and variation
in second language acquisition; recently he has been particularly concerned
396 Contributors
with folk beliefs about and attitudes towards language, the subject of a
2000 book Folk Linguistics with Nancy Niedzielski (Mouton de Gruyter).
He recently shared an excellent meal with Richard Watts in Costa Mesa,
CA at the annual meeting of the American Association for Applied Lin-
guistics.
CAROL G. PRESTON is a former teacher, teacher trainer, and administrator

in English as a second language and bilingual education (Spanish-English)
in public schools and at community college and university levels. She is
now engaged in editorial work in linguistics. She recently shared the worst
Italian meal she ever ate with Richard Watts as they listened to the great
bell of Rovereto during the International Pragmatics Association meeting
in Riva del Garda, where, in contrast, excellent food was served.
DANIEL SCHREIER taught and lectured in Switzerland, New Zealand, Ger-

many and in the USA, and he currently holds a Chair of English Linguis-
tics at the University of Zurich, Switzerland. His research interests include
sociolinguistics, dialectology, language variation and change and contact
linguistics. He is author of Isolation and Language Change (2003) and
Consonant Change in English Worldwide: Synchrony Meets Diachrony
(2005, both published with Palgrave Macmillan), St Helenian English:
Origins, Evolution and Variation (2007, Benjamins), and co-author (with
Karen Lavarello-Schreier) of Tristan da Cunha: History People Language
(2003, Battlebridge).
JÜRG RAINER SCHWYTER is Professor of English Linguistics at the Univer-

sity of Lausanne, where he teaches all aspects of the linguistics curriculum.
His research interests include historical linguistics, legal English past and
present, general stylistics, and questions of standardisation. He is currently
researching a book on the language policy of the early BBC.
DANIEL STOTZ is a professor of language education at the Zurich Univer-

sity of Teacher Education. Besides teacher training, he is involved in socio-
linguistic and applied linguistic research on foreign and second language
learning and materials development. In the 1980s, he was Richard Watts’
first assistant and doctoral student at the University of Berne; he later
taught English at a business college and developed language courses for
scientists at the Swiss Federal Institute of Technology. Between 1998 and
2004, he built up a research and development centre at the Zurich Univer-
sity of Applied Sciences Winterthur.
Contributors 397
JÜRG STRÄSSLER studied at the University of Zürich and took his first de-
gree in 1977 in English Language and Literature, German Language and
Physics. He then took up his studies in Theoretical Linguistics at the Uni-
versity of Cambridge (Gonville and Caius College) and finished this post-
graduate course with an M.Phil. dissertation on Truth-Conditional Ap-
proaches to Semantic Theory. In 1982 he finished his dissertation on
Idioms in English, A Pragmatic Analysis at Zürich University, being Rich-
ard Watts’ first PhD candidate. Since 1981 he has been teaching English at
a Swiss grammar school. In 1994 he started lecturing in phonetics and
phonology at the University of Zürich and from 1994 till 2002 he held the
position of a senior assistant to Richard J. Watts and lecturer in linguistics
at the English Department of the University of Berne and worked as the
assistant to the editor of Multilingua. Since 2006 he has again been lectur-
ing in linguistics at the University of Berne. During the last few years he
has concentrated on lexically-driven parsing within a generative frame-
work. Apart from syntax and semantics, he has also always been interested
in phonetics and phonology, especially forensic and clinical phonetics.
PETER TRUDGILL was born, at roughly the same time as Dick Watts, in
Norwich, England, not much more than 100kms from Dick's childhood
home. Peter Trudgill taught at the Universities of Reading and Essex in
England, before working in Switzerland at the Universities of Lausanne
and Fribourg between 1992 and 2005. He is currently Professor Emeritus
of English Linguistics at Fribourg; Honorary Professor of Sociolinguistics
at the University of East Anglia, Norwich; Adjunct Professor of Sociolin-
guistics at the University of Agder, Kristiansand, Norway; and Adjunct
Professor in the Research Centre for Linguistic Typology at La Trobe Uni-
versity, Melbourne, Australia. He is known for his work in sociolinguistics,
but was recently described as a “theoretical dialectogist”, a description he
finds appropriate. His publications include Sociolinguistics: an Introduc-
tion to Language and Society; Dialects in Contact; and New Dialect For-
mation: the Inevitability of Colonial Englishes.
MERCEDES VIEJOBUENO is a faculty member in the Department of English

of the Faculty of Philosophy and Letters at the National University of Tu-
cumán, Argentina. She teaches English grammar and is especially inter-
ested in the pragmatics of interpersonal communications, including aspects
of politeness and impoliteness, particularly as they contrast between Eng-
lish and Spanish.
398 Contributors
KATIE WALES is a Research Professor of English, University of Sheffield.

She has formerly held Chairs in English Language at Royal Holloway Uni-
versity of London and the University of Sheffield. Her many publications
and research interests range widely over English language and stylistics
from Old English to the present day. Like Dick Watts, she has a passionate
interest in British folk-songs and ballads, especially from the North-East of
England.
Contact information
D. J. Allerton
Englisches Seminar der Universität, CH-4051 Basel, Switzerland, D-
[email protected]
Franz Andres Morrissey

Department of English Languages and Literatures, University of Berne,
Länggass Str. 49, CH-3000 Bern 9, Switzerland, [email protected]
Anita Auer
Universiteit Leiden, Opleiding Engels/LUCL, P.N. van Eyckhof 4, Postbus
9515, 2300 RA Leiden, The Netherlands, [email protected]
Tony Bex
84, Grange Road, Ramsgate, KENT CT11 9PX, UK, [email protected]
Urs Dürmüller
Länggass Str. 49, CH-3000 Bern 9, Switzerland,
[email protected]
Elke Hentschel
Institut für Germanistik, Universität Bern, Länggass-Strasse 49, CH-3000
Bern 9, Switzerland, [email protected]
Juliane House
Universität Hamburg, Institut für Allgemeine und Angewandte Sprachwis-
senschaft, Von-Melle-Park 6, D-20146 Hamburg, Germany, jhouse@uni-
hamburg.de
Adam Jaworski
Centre for Language and Communication Research, ENCAP, Cardiff Uni-
versity, Humanities Building, Colum Drive, Cardiff CF10 3EU, Wales,
UK, [email protected]
400 Contact information
Miriam Locher
Länggass Str. 49, CH-3000 Bern 9, Switzerland, [email protected]
Adrian Pablé
Section d’anglais, Bâtiment Anthropole, Université de Lausanne, CH-1015
Lausanne, Switzerland, [email protected]
Ingrid Piller
Macquarie University, AMEP Research Centre, NCELTR, W6C, NSW
2109 Australia, [email protected]
Dennis R. Preston
4409 Copperhill Drive, Okemos, MI 48864, USA, [email protected]
Carol G. Preston
4409 Copperhill Drive, Okemos, MI 48864, USA, [email protected]
Daniel Schreier
English Department, University of Zürich, Plattenstr. 47, CH-8032 Zürich,
Switzerland, [email protected]
Jürg Schwyter
Section d'anglais, Bâtiment Anthropole, Université de Lausanne, CH-1015
Lausanne, Switzerland, [email protected]
Daniel Stotz
Zurich University of Teacher Education, Lagerstrasse 5, P.O. Box, CH-
8090 Zürich, Switzerland, [email protected]
Jürg Strässler
English Department, University of Zurich, Plattenstrasse 47, CH-8032
Zurich, Switzerland, [email protected]
Peter Trudgill
The School of Language, Linguistics and Translation Studies, University
of East Anglia, Norwich NR4 7TJ, UK, [email protected]
Contact information 401
Mercedes Viejobueno
Facultad de Filosofía y Letras, Universidad Nacional de Tucumán, Avda.
Benjamín Aráoz 800, San Miguel de Tucumán - Tucumán (CP 4000), Ar-
gentina, [email protected]
Katie Wales
School of English, University of Sheffield, Sheffield S10 2 TN, UK,
[email protected]
Subject index
AAVE, 110, 111 change–of–state verbs, 34
academic writing, 13, 282–296 cleft sentence, 7, 39, 40, 120
Académie Française, 6, 134, 142 Cockney, 54, 60, 70, 176, 196, 209,
accent, 4, 50, 52–60, 74, 176, 177, 210
180, 186, 187, 195–215, 231– codification, 3, 5, 6, 134, 144, 222,
233, 240, 244, 246–247 230, 232, 233, 234
accommodation, 177, 197 coherence marker, 362
agreement communicative act, 58, 232, 367
contact, 39 communicative practice, 301, 305
tense, 39 competence, 12, 226, 239, 248, 249,
unorthodox, 7, 37–39 251, 255–274, 282, 296, 360,
agresividad, 373, 374, 377, 381, 384 363
American English, 7, 9, 23–42, 47, contact (dialect/language), 69–80,
70, 85, 105–123, 151, 152, 155, 86–89, 97–98, 100, 107, 177,
166, 242, 244, 245, 247, 248, 234, 243, 246, 257
249, 252, 284, 370, 371 content and language integrated
Americism, 24, 33 learning (CLIL), 270
Anglicism, 323–344 Continental European English, 244
Anglicization, 315 convergence, 48, 61–62, 177
angloversal, 118 correctness, 1, 2, 5, 7, 10, 109, 131,
Appalachian English, 119 134, 136, 137, 139, 149, 151,
audience design, 197 152, 160, 175, 176, 177, 180,
authority, 1–7, 11, 121, 132, 141, 186–188, 225, 228–231, 265,
143, 164, 176, 187, 199, 229, 268
242, 260, 301, 302 cosmopolitanism, 304, 314, 315
creole, 71, 78, 86, 87, 98, 100
BBC, 7, 10–11, 58, 59, 175–188, creolisation, 97, 98, 100
207 crítica, 372, 373, 378, 381, 384
bilingualism, 60, 258, 259 criticism, 372, 373, 375, 380, 383,
Briticism, 24, 42 384
British English, 5, 7–9, 23–42, 85, cross–cultural, 16, 281
86, 88, 95, 99, 108, 150, 151, crossing, 260, 303, 304, 311, 314
152, 155, 156, 166, 199, 200, cusum, 284–296
209, 243–252
Broadcast English, 11, 182, 184, decreolisation, 86
185, 187 Denglish, 14, 323
burla, 372, 373 descortesía, 372, 373, 378, 381, 384
descriptivism, 4, 10, 16, 34, 85, 141,
Canadian English, 76–80 142, 144, 166, 223, 368
Canadian raising, 77–80 dialect, xx, 3–9, 12, 16, 35, 47–62,
CANCODE, 12, 221–235 69–80, 86, 87, 99, 100, 105, 108,
404 Subject index
110, 122, 135, 177, 196, 205, General American (GenAm), 200
206, 215, 222, 227, 228, 229, genre, 11, 13, 106, 107, 131, 158,
231, 256–269, 310, 316, 326 196, 199, 202, 203, 210, 215,
diffusion, 52, 53, 54, 59, 177 232, 284, 286, 290, 293, 295–
diglossia, 12, 256, 258, 259, 271, 296, 304, 306, 317
306 Geordie, 50, 55, 57, 60, 201–202
disaffiliative action, 354 German English, 7, 23, 24, 42
disagreement, 354 German–Americanism, 8, 24, 25, 26,
discourse community, 140 41
discourse marker, 61, 358, 362 global citizenship, 304
dishonesty, 373 Global English, 42, 47, 239
divergence, 47, 48, 57, 61–62, 108, globalisation, 8, 13, 56, 62, 239,
199, 373 241, 249, 269, 270, 271, 272,
diversification, 56, 57, 87, 100, 246 296, 310, 314, 315, 317, 351
dormancy, 54 glocalisation, 271
grammar
Early Modern English, 10, 122, 150, discourse, 11, 221–235
152, 153, 155, 160, 161, 165 emergent, 224–225, 232
East India Company, 89, 94 systemic, 222, 223, 261–262
empiricism, xx, 143 grammaticalisation, 222, 226, 363
English Native Norm Irrelevance
Hypothesis, 355, 356, 360, 363 High German, see Standard German
English for special purposes (ESP), historical sociolinguistics, 88, 128
243, 245, 252 hypercorrection, 74, 97, 177, 209,
English as foreign language (EFL), 302
xix, 4, 11–14, 85, 221–235, 239–
252, 255–274, 281–296 idealisation, 223–226
English in Switzerland, 13, 23, 240 identification, 240, 245–246, 303–
English as lingua franca (ELF), see 304
lingua franca identity, xix, 48, 56–61, 199, 211,
ergative, 34, 35, 37, 38 227, 233, 246–247, 256, 272–
Estuary English, 50, 52, 54, 55, 57, 274, 302–304, 311, 355
59, 210 ideology, xx, 13, 127, 134–135, 143,
evincive, 362 239, 247, 249, 256, 273, 301–
expanding circle varieties, 85, 100, 316
227, 228, 232, 234, 235, 239– immersion, 12, 255–274
252 impoliteness, 3, 14–16, 137, 349–
364, 367–387
face attack acts, 367–387 infinitivitis, 7, 40
falsedad, 373 inflection, 10, 113–114, 149, 325,
folk singing, 11, 60, 195–215 327–328, 344
folk speech, 49, 52, 59, 106, 107, inner circle varieties, 85, 227–228,
110, 118, 122, 370 234, 239–252
founder effect, 92, 93 interlanguage, 244, 291
Subject index 405
International English, 11, 12, 47, New England folk speech, see folk
227, 233, 235, 239–252 speech
IPA, 185–187 non–standard, see standardisation
Ireland, 8, 70, 74, 76–77 norm, 1–17, 40, 55, 56, 57, 61, 62,
75, 110, 127, 144, 195, 227, 233,
jocularity, 373 284, 302, 325, 351–364, 368,
379
koiné, 72, 79, 86–87 normalization, 302, 316, 317
language academy, 6 Old English, 8, 73–74

language determination, 3, 5, 135 orthography, see spelling
language imperialism, 61, 227, 233, outer circle varieties, 85, 100, 227–
252 228, 230, 232, 234, 239–252
language policy, 175, 180, 185, 269–
271 paradigmatic variation, 156
langue, 222 parole, 222
Late Modern English, 152, 154–155, perfect, 26–32
165 performance, 11, 60, 196–198, 211,
latency, 54 215, 356
lest, 149–166 phonetic script, see also IPA, 182
levelling, 8, 48, 50–51, 53, 55, 57, phonological constraint, 119
95, 106, 108, 116 pidgin, 86–87, 97–98, 100
lingua franca, 11, 13, 16, 99, 239– place names, 73, 181–182, 184, 302
252, 270, 281, 296, 314, 317, politeness, xxi, 3, 9, 14–16, 54, 131,
364 133–139, 141, 247, 349–364,
linguascape, 301–317 367–387
politic behaviour, 15–16, 351, 356,
Macdonaldisation, 227 364
maldad, 372, 373, 377 pop music, 195–215
Mancunian, 57, 208, 210 possibility to/for, 40
mandative subjunctive, 150, 166 pragmatic constraint, 111, 122
Mid–Atlantic English, 248, 252 pragmaticalisation, 363
Middle English, 8, 74–76, 151 prediction, 48, 50, 52, 223
minority language, 258, 302 prescriptivism, 4–7, 9–10, 11, 14–
mockery, 372, 373 15, 61, 123, 130–144, 155, 166,
mood, 10, 149, 150, 160, 161, 164, 179, 186, 229, 325
165, 166 prestige, 6, 55, 69, 86, 199, 239,
morpheme boundary, 329–331, 342 242, 246–247, 249, 271, 301
multilingualism, xx–xxi, 12, 85, 87, printing, 4, 6–7, 128–140, 187
98–99, 244, 250, 255–274, 311 pronunciation, 3–4, 6–7, 10–11, 48–
49, 51, 53–55, 59, 175–188,
new dialect formation, 8, 70, 71, 73, 195–215, 231, 233, 243, 250,
74, 76, 77, 80, 108 252, 324, 333–334, 337
406 Subject index
Public school pronunciation, see RP, Standard German, 72, 199, 255–274,
7, 177 306, 324, 327, 332
standard, oral, see pronunciation
QSUM, see also cusum, 13, 284 standardisation, xx–xxi, 1–17, 48,
51, 85, 127–128, 133–135, 139,
reallocation, 74, 76–80 176, 187, 228, 232, 234, 247,
Received Pronunciation (RP), 4, 7, 284, 294, 316–317, 327–328,
49, 54–55, 176–177, 180, 187– 342, 344
188, 195–215, 246, 249 non–standardisation, 4, 7–9, 61,
referee design, 199–200, 205 85–100, 108–110, 121–122, 208,
reference style, see style 225, 256
regiolect, 55–56 standardness, 61, 301
regional variation, 3, 8, 47–62, 77 stigma, 6, 8, 54, 61, 105, 121–122,
register, 3–4, 61, 151–152, 259, 271, 135, 176,
302 style, 3–4, 11–13, 32, 60–61, 76,
relational work, 14–15, 356, 360, 105, 109, 137–138, 140, 151–
364 152, 154, 177, 180, 187, 195–
rock, 195–215 215, 263, 281–296, 324
rudeness, 367–387 initiative, 199
rule, 1, 4–5, 14, 131, 160, 164, 178, reference, 11, 199–202, 204,
211, 221–222, 229, 295, 323– 206–207, 209–210, 213–215
343 responsive, 199
rule, 367, 379 stylometrics, 285
subjunctive, 10, 110, 132, 149–166
Salem Witchcraft Papers, 9, 105–123 substratum, 41, 85, 97, 100
sarcasm, 204, 368–387 Swiss English, 7, 23–24, 42, 85
Scouse, 50, 57, 59, 207–208, 210
second language acquisition, 240, Teenage language, 324
243–244, 249, 257, 270 text type, 154, 250, 283, 295
selection, 135, 226, 274 time adverbial, 28–30, 32
Self–Centered Hypothesis, 356, 360, tinge hypothesis, 369, 385, 387
363 tourism writing, 13, 301–316
social alignment, 354 travelogues, 13, 301–316
sociolect, 76, 108
sonority, 11, 212–215 varieties of English, 1–17, 24, 42,
spelling, 6–7, 109–110, 139–140, 47–62, 69–80, 85–100, 107–108,
152, 182, 185, 243, 252, 283, 111, 121–122, 135, 137, 139,
325–327, 242–343 150, 166, 206, 215, 221–235,
St Helena (StHe), 8–9, 78, 85–100 239–252, 255–274, 303
standard, 1–17 vernacular, 48, 54, 56, 95, 109, 111,
Standard English, 1–17, 49, 51, 61, 118, 123, 243, 250
85, 100, 116, 128, 144, 175, 221,
222, 228–232, 324 World English, 24
Author index
Abrahamsen, Adele, 222 Bourdieu, Pierre, 274, 351
Addy, Sidney, 49 Bousfield, Derek, 15, 367
Aitchison, Jean, 176 Boyer, Paul, 106–107, 123
Alexander, Henry, 105, 107–108 Branford, William, 69, 70
Algeo, John, 26, 77, 167 Bregy, Anne–Lore, 270
Allerton, D. J., 26, 40, 42–43 Breiteneder, Angelika, 243
Alptekin, Cem, 234 Briggs, Asa, 175–177
Anderson, Isaac, 43 Bright, John A., 230
Andres, Franz, 255–256, 270 Brinton, L., 69
Ash, John, 162, 164 Britain, David, 59, 79, 100
Atkinson, Pete, 207 Brogan, Hugh, 42
Atwood, E. Bagby, 110, 116, 124 Brohy, Claudine, 270
Auer, Anita, 150, 154, 162, 168 Brooke, Thomas H., 89, 92
Augst, Gerhard, 344 Brown, Penelope, 14–15, 367–369
Austin, Paddy, 367–368 Brutt–Griffler, Janine, 61
Büeler, Xaver, 270
Bachmann, Thomas, 260 Burchfield, Robert, 47–48, 180, 191
Bailey, Guy, 110, 122 Burquest, Donald A., 212
Bailey, Richard, 48, 248 Bussmann, Hadumod, 284
Bailey, Stephen, 284 Butler, Charles, 128
Barbe, Katharina, 369
Barber, Charles Laurence, 150, 154 Cameron, Deborah, 58, 302
Bassett, Marvin, 111 Carmosino, Giovanni, 210–211
Bauer, Laurie, 70–71 Carter, Ronald, 4–5, 221, 223–226,
Baugh, L. Sue, 284 229–230, 233
Baumgarten, Nicole, 363 Cassidy, Frederick, 75
Bayly, Anselm, 162, 164 Caxton, William, 6–7
Beal, Joan, 56–58 Chambers, Jack, 54, 61, 77, 95
Bechtel, William, 222 Chang, W. R., 285
Becker, Tabea, 328 Charleston, Britta Marian, 150, 153
Beebe, Leslie M., 367 Cheshire, Jenny, 42, 85, 229
Belanger, Terry, 134, 139 Chomsky, 222
Belcher, Diana D., 295 Clarke, David, 210
Bell, Allan, 197–200, 205, 211 Clayson, Alan, 195
Beniak, Édouard, 72, 75 Cobbett, William, 164–65
Bentahila, Abdel–ali, 234 Cohen, Anthony Paul, 62
Berthele, Raphael, 262 Collier, John of Urmston, 48, 49
Bex, Anthony R., 85, 228–230, 232 Colston, Herbert L., 368
Bloomfield, Leonard, 4, 142 Combrink, Johan, 72
Bonfiglio, Thomas P., 85 Cook, Guy, 226
Boucher, Jonathan, 47 Cook, Vivian J., 235
408 Author index
Cooley, Marianne, 105 Farringdon, Jill M., 13, 286, 288–

Coupland, Nikolas, 61, 211, 301, 291, 293, 295, 297
310 Feak, Christine B., 284
Cran, William, 48 Fee, M., 69
Crowley, Terry, 85 Fell, John, 10, 128, 130, 140–144
Crowley, Tony, 229 Fenning, Daniel, 130
Crystal, David, 47–48, 60–62, 87, Ferguson, Charles A., 199, 306
241–242 Fillbrandt, Eva–Liisa, 150
Culpeper, Jonathan, 367–368 Firth, Alan, 353
Cumming, Susanna, 222–223 Fisher, Ann, 130–131, 140
Fitzmaurice, Susan, 134–135
da Silva Neto, Serafim, 72 Flender, Reinhard, 196–197
Dalby, Andrew, 62 Forsyth, Richard S., 287
Daniel, Samuel, 47 Foster, Alexander, 165
Darot, Mireille, 72 Foster, Margaret E., 165
Davis, Eirlys, 234 Foucault, Michel, 304
de Morgan, Augustus, 285, 287 Fowler, Henry Watson, 39, 149
de Saussure, 222 Frings, Theodor, 71
Defoe, Daniel, 134 Frith, Simon, 199
Demont–Heinrich, Christof, 311
Dent, Susie, 62 Gal, Susan, 304–305, 315
Devitt, Amy J., 13, 295, 296 Gardiner, Jane, 164
Dews, Shelly, 368–369, 375, 383– Gargen, Henry, 92
385, 389 Gibaldi, Joseph, 283, 296
Dons, Ute, 160 Gibbs, Raymond W. Jr., 383
Doyle, Brian, 54 Giles Howard, 211
Dröschel, Yvonne, 23 Gimson, A. C., 176, 180, 187, 189
Duncan, J., 60 Glucksberg, Sam, 383
Duncan, N., 60 Gnutzmann, Claus, 11, 239
Dürmüller, Urs, 240–241, 245, 311 Goffman, Erving, 267, 303, 354
Dylewski, Radoslaw, 110, 123 González–Díaz, Victorina, 150
Gordon, Alexander, 140
Earle, John, 166 Gorji, Mina (ed.), 15
Eastwood, John, 4–5 Görlach, Manfred, 42, 149–153
Eckert, Penelope, 356 Gosse, Philip, 89–90, 92
Edmondson, Willis, 352, 354 Graddol, David, 11, 47, 85, 87, 239,
Eelen, Gino, 17 242, 247
Eisenberg, Peter, 342 Grant, Benjamin, 99
Ellis, Alexander, 49, 53–54, 79, 176, Graves, Barry, 203
189 Greaves, Paul, 128, 160
Ellis, Stanley, 58 Grice, H. Paul, 235, 360
Elmes, Simon, 50, 58–59, 61, 206 Griffiths, Bill, 50
Elsness, Johan, 26 Grin, François, 258
Ensslin, Astrid, 301 Grose, Francis, 48–49, 53
Author index 409
Grund, Peter, 107, 123, 150 Irvine, Judith T., 304–305, 315
Gu, Yueguo, 387
Jack, Gavin, 303
Häcki Buhofer, Annelies, 260 Jaffe, Alexandra, 301–302
Haegeman, Liliane, 150 Janisch, Hudson R., 92–94
Hägi, Sara, 358–359, 275 Jaworski, Adam, 301–302
Halliday, Michael A. K., 224, 361– Jenkins, Jennifer, 11, 42, 243–244,
363 246–247, 252
Hammarström, Göran, 70 Jenkins, Paul O., 203
Hamp–Lyons, Liz, 284 Johnson, Sally, 301
Hancock, Ian, 98 Johnson, Samuel, 6–7, 134, 140
Hannah, Jean, 26, 42 Johnston, R.J., 62
Hannerz, Ulf, 304 Jones, Daniel, 6–7, 178, 182, 185–
Harris, Roy, 177, 224 186, 189, 324
Hartley, Sue, 54 Jones, Hugh, 135, 137
Hartman, James, 7 Jonson, Ben, 128, 160
Hasan, Ruqaiya, 224, 361 Jordan, Robert R., 284
Haugen, Einar, 134 Jorgensen, Julia, 369, 383
Hawkins, Peter, 212, 214
Heasley, Ben, 284 Kachru Braj B., 85, 87, 227–228,
Heller, Monica, 257–258, 268 230, 234, 240, 242, 247
Hentschel, Elke, 327 Kaplan, Joan, 368, 375, 385
Heslop, Rev. Richard, 49 Katz, Albert N., 368
Hewings, Martin, 4, 17 Keller, A. Tabouret, 58, 73
Hoban, Russell, 62 Kerswill, Paul, 54, 57, 59
Hoggart, Richard, 49, 58, 62 Kienpointner, Manfred, 367
Holloway, William, 49 Kirkham, Samuel, 106
Hollyman, K. Jim, 72 Kirkman, Rick, 43
Holmes, David I., 287 Kleinberger Günther, Ulla, 270
Honey, John, 86, 144, 229 Kolde, Gottfried, 258
Hopper, Paul J., 222 Kortmann, Bernd, 167
Hornby, Albert Sidney, 288, 290 Krapp, George P., 105, 190
House, Juliane, 247, 352, 354–356, Krashen, Stephen, 257
360, 363 Kreuz, Roger J., 383
Huang, Chun, 368 Kroskrity, Paul V., 301
Huddleston, Rodney, 4–5, 39, 151, Kurath, Hans, 77–78, 106–107, 117
154, 157 Kytö, Merja, 107, 123
Hughes, Arthur, 180, 191
Hughes, Rebecca, 221, 223 Labov, William, 177
Humphrys, John, 229 Lakoff, Robin, 367
Hundt, Marianne, 150 Langé, Gisella, 270
Langner, Heidemarie, 325
Ihalainen, Ossi, 53–55, 111 Lanham, Len, 71
Intemann, Frauke, 11, 239 Lass, Roger, 70
410 Author index
Laver, John, 78 Mellinger, Wayne M., 304

Le Page, Robert, 58, 73, 75, 211 Melliss, John C., 90
Le Pape Racine, Christine, 270 Mencken, Henry Louis, 123
Lee, David YW, 158–159 Meuter, Tessa, 269
Lee, Jinhee, 325 Michael, Ian, 129, 143
Leech, Geoffrey N., 150, 167, 367 Michaelson, Sidney, 286
Leigh, Percival, 165 Middleton, Thomas, 154
Leith, Dick, 85, 87 Miège, Guy, 161
Lesznyak, Agnes, 352 Miller, Jim, 361
Leuthold, Heinrich, 261 Milroy, James, 2, 4–6, 54, 58, 86,
Levinas, Emmanuel, 356, 360 127, 134–135, 142, 229
Levinson, Stephen C., 14–15, 367– Milroy, Lesley, 2, 4–6, 51, 54, 86,
369 127, 134, 142, 187, 229
Lippi–Green, Rosina, 85 Mishoe, Margaret, 111
Lipski, John, 70, 72 Mitchell, Tony, 216
Llamas, Carmen, 55, 57 Moag, Rodney, 72
Locher, Miriam A., 14–15, 17, 136, Modiano, Marko, 248, 252
364 Moessner, Lilo, 150, 167
López–Couso, María José, 167 Mollin, Sandra, 42
Lowth, Robert, 130–140, 163–164, Montgomery, Michael B., 88, 109,
168 111
Lüdi, Georges, 259 Morin, Yves–Charles, 72
Lye, Thomas, 161 Morton, Andrew Q., 286, 288, 293,
297
MacCannell, Dean, 310 Mougeon, Raymond, 72, 75
Macneil, Robert, 48 Mufwene, Salikoko S., 69, 86, 92
Mair, Christian, 25, 42 Mugglestone, Lynda, 49, 176, 190
Mandelbrote, Scott, 139 Mühlhäusler, Peter, 86
Mangold, Max, 324 Müller, Max, 142
Markova, Ivana, 356 Murray, Heather, 7, 11, 306
Marsh, David, 270 Murray, Lindley, 130–131, 137,
Marti, Werner, 345 163–164, 168
Martin, Jim, 361 Musgrove, Frank, 53
Matsumoto, Yoshiko, 387
Mattoso Camara, Joaquim, 72 Nicholson, J., 49
McArthur, Tom, 177, 186, 190 Nielsen, Hans Frede, 73
McCarthy, Michael, 4–5, 221, 223– Nikula, Tarja, 267, 270
226, 230, 233 Nissenbaum, Stephen, 106–107, 123
McConnell–Ginet, Sally, 356 Nixon, Neil, 208
McCrum, Robert, 48 Norton, Bonny, 274
McDavid, Raven, 77–78
McGregor, Gordon P., 230–231 Okamoto, Shinichiro, 368, 387
McLean, I. P., 285 Ono, Tsuyoshi, 222–223
McWhorter, John, 71 Orton, Harold, 49, 58, 185, 189
Author index 411
Ospelt, Barbara, 260 Rohdenburg, Günther, 168

Övergaard, Gerd, 150, 166 Röhrenbeck, Ulrike, 324
Romaine, Suzanne, 69
Pablé, Adrian, 110 Rosenberger, Lukas Michael, 85
Påhlsson, Christer, 55–56 Rosewarne, David, 54
Pauleau, Christine, 72 Ross, Garry, 122
Payne, David L., 212 Russell, Dave, 56, 59
Penny, Ralph, 56
Pennycook, Alastair, 227, 233 Sager, Juan C., 281
Peschel, Corinna, 328 Samuels, Michael, 75
Peters, Pam, 150, 166, 168 Sankoff, Gillian, 87
Phillipson, Robert, 12, 227, 233–234 Savage, W. H., 151–152
Phipps, Alison, 303 Scharloth, Joachim, 258–259, 275
Piller, Ingrid, 306, 317 Schieffelin, Bambi B., 301
Pirie, David B., 284 Schlüter, Julia, 168
Pittner, Karin, 324–325 Schmid–Joos, Siefrid, 203
Pitzl, Marie–Louise, 243 Schmitt, Peter, 324
Poirier, Claude, 72 Schneider, Edgar W., 85, 88, 109
Poole, Joshua, 160–161 Schourup, Lawrence, 362
Potter, Jonathan, 224–225 Schreier, Daniel, 71, 78, 86, 90, 100
Powell, Anthony, 43 Schütte, Dagmar, 324
Preisler, Bent, 230 Scott, Jerry, 43
Preston, Dennis R., 370 Seidlhofer, Barbara, 51, 62, 243–
Priestley, Joseph, 130, 140, 142 244, 246–247, 252
Prodromou, Luke, 226–227, 234 Serpollet, Noëlle, 150, 166
Pullum, Geoffrey K., 4–5, 40, 151, Setter, Jane, 7
154, 157 Shapiro, Peter, 196
Shaw, John, 163–164
Quirk, Randolph, 6, 39, 43, 47, 87, Siebenhaar, Beat, 345
151–152, 156–157, 166 Sieber, Peter, 260
Sievens, Mary Beth, 106, 110
Rampton, Ben, 61, 260, 303 Silverstein, Michael, 302
Rauhe, Hermann, 196–197 Simmelbauer, Andrea, 50
Ray, John, 49, 55 Simpson, Paul, 196, 199–200, 208,
Redfern, Richard K., 149 213
Reith, J. C. W., 175–176, 186, 188, Sinclair, John, 4
190 Sitta, Horst, 260
Rhodes, Barrie, 50 Smith, Jeremy, 187
Ris, Roland, 259 Sperber, Dan, 224, 360
Rissanen, Matti, 107, 123 Steadman–Jones, Richard, 131
Ritzer, George, 317 Stebler, Rita, 270
Rivard, Adjutor, 70, 75 Stein, Dieter, 85, 135–136
Roach Peter, 7 Stotz, Daniel, 256–257, 269–270,
Robertson, John, 207 275
412 Author index
Strang, Barbara, 156 Walker, Terry, 62, 150

Strässler, Jürg, 189 Wall, Arnold, 70
Sudbury, Andrea, 78 Wallis, John, 128–129, 160
Swales, John M., 13, 281–282, 284, Wanner, Anja, 295
295–296 Ward, William, 163
Swann, Joan, 85, 87, 197 Watt, Dominic, 180, 191
Sweet, Henry, 48, 164, 176, 205 Watts, Richard J., 7–9, 11, 14–15,
Swift, Jonathan, 6, 132–134 17, 85, 134–135, 138, 140–141,
228, 230, 255–256, 258, 260,
Tajfel, Henri, 355 269–270, 274, 281, 301–302,
Taylor, Talbot J., 223 306, 311, 351, 355–356, 363–
Taylor, C. C., 57, 60 364, 367
Taylor, Peter, 53 Webster, Noah, 47, 123
Terrell, Tracy, 257 Weinert, Regina, 361
Terry, Richard, 131 Weinreich, Uriel, 26
Tieken–Boon van Ostade, Ingrid, Wells, John C., 54, 78, 97, 180, 190–
128, 131–134, 144, 168 191, 200–201, 206, 208, 210,
Tissot, Laurent, 306 Wenger, Etienne, 356
Toolan, Michael, 222 Werlen, Iwar, 259
Toplak, Maggie, 368 Weydt, Harald, 327
Torr, Andrew, 283 Wharton, John, 128
Townsend, A.R., 57, 60 White, James, 162, 164
Traugott, Elizabeth Closs, 149, 222 Wichmann, Anne, 367
Tribble, Christopher, 284 Widdowson, Henry G., 51, 62, 222–
Trudgill, Peter, 3–5, 8, 23, 26, 42, 223
48, 52–53, 55–57, 62, 69–71, Widdowson, John, 50, 56
73–79, 86, 108, 177, 180, 191, Wieser, Judith, 26
196–197, 200, 205, 209, 211, Williams, Ann, 54, 59
215–216 Wilson, Deirdre, 224, 360
Turabian, Kate L., 283, 296 Winner, Ellen, 368–369, 375, 383,
Turner, George W, 70 385, 389
Wolff, Dieter, 270
Upton, Clive, 48, 54, 56 Wolfram, Walt, 119, 124
Urry, John, 303, 306 Woolard, Kathryn A., 301
Wright, Joseph, 49
Veblen, Torsten, 313 Wright, Laura, 85, 98, 124
Vorlat, Emma, 168
Yule, G. Udny, 285
Wagner, Max, 71
Wakelin, Martyn, 108 Zifonun, Gisela, 324
Wales, Katie, 53–54, 58

Standards and Norms in The English Language Contributions To The Sociology of Language

Uploaded by

Copyright:

Available Formats

Standards and Norms in The English Language Contributions To The Sociology of Language

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

Standards and Norms in The English Language Contributions To The Sociology of Language

Uploaded by

Copyright:

Available Formats

Standards and Norms in the English Language

Library of Congress Cataloging-in-Publication Data

Standards and norms in the English language / edited by Miriam A.

Bibliographic information published by the Deutsche Nationalbibliothek

Miriam Locher 1 March 2008 Jürg Strässler

David J. Allerton Jenny Cook-Gumperz

Monica Heller Eva Maurer

Marcia Schwartz Beatrice Türler

Watts, Richard J. and Urs Weidmann (eds.)

3.1. Single authored

1988a Language as a marker of socio-cultural identity in Switzerland. In

1995a Justifying grammars: A socio-pragmatic foray into the discourse

Watts, Richard J. and Beat Homberger

linguistic features are bound up in the construction of social categories,

Part I. English over time and space

Part II. English usage in non-native contexts

Part III. Issues on politeness and impoliteness

Contributors ........................................................................................... 393

Miriam A. Locher and Jürg Strässler

1. Norms and standards: The theme of the collection

This collection of papers is inspired by the work of Richard Watts, whose

standard, a metre is always defined as the length of the path travelled by

2. English over time and space

the media, especially in printing, and employed in the educational systems

pronunciation and for users of English as a foreign or second language”

2.1. Variation in English varieties and dialects

The chapters contained in this section deal with phonological, morphologi-

cases at least, grammatical Immigrant Americanisms, and in some cases

2.2. Research on English standardisation processes

2.3. Issues of standards in oral production

through three phases, from a “strict and dogmatic prescriptivism” (186) to

3. English usage in non-native contexts

language. In addition, he argues that corpora work, such as the CANCODE

Stotz exemplifies these issues with an analysis of classroom interaction and

tourism sector, such as Croatian, Portuguese, Serbian, or Spanish. The

4. Issues on politeness and impoliteness

Politeness research is another important field of linguistic enquiry for

ticular discursive practice. He introduced the concept of ‘politic behaviour’

Mercedes Viejobueno, Carol G. Preston and Dennis R. Preston. In her con-

The theme of this collection is a discussion of the notions of ‘norms’ and

Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik

Watts, Richard J., Sachiko Ide and Konrad Ehlich (eds.)

Like Dick Watts, I was a professor of English linguistics at a Swiss univer-

Americanization affects English-speaking Britain at least as much as it

LOB 3 occurrences (all meaning ‘unable to speak’)

Although the origin of the second meaning (‘stupid’) in American English

2. Loss of the perfect

(1) Gordon has waited patiently for promotion.

(2) Gordon has often waited for the children.

(3) Gordon has occasionally blundered.

(4) Gordon has blundered.

Non-punctual processes, as illustrated by (1), have duration, as do repeti-

tuality in question (including the series of repetitions, where there is one)

(11) Why do I feel nauseous lately?

(12) I’m not using the Ferrari’s much lately.

(13) What’s up with the Pope lately?

It is also occasionally found used with the simple past.

“already done”: about 1,020,000 hits

The temporal adverb yet is the negative and interrogative equivalent of