Language Anxiety

in memory of my brother
Michael J. Machan
1946–2007

˝Hœ
b S KØ I
Ø
ø 

ıªæ EØ
l
Ł ; Kªg
b ¼ıŁ Kç Æ¥
Æ
Ø çªÆ  Yåø;
Y øº   
æøŁ 
Ææ ı ºº Iªæı.
Homer, The Odyssey, 11.81–3
Language Anxiety
ConXict and Change in the History of English

T I M W I L L I A M M AC H A N

1
 3
Great Clarendon Street, Oxford ox2 6dp
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With oYces in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
# Tim William Machan 2009
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
First published by Oxford University Press 2009
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Data available
Typeset by SPI Publisher Services, Pondicherry, India
Printed in Great Britain
on acid-free paper by
Biddles Ltd., King’s Lynn, Norfolk

ISBN 978–0–19–923212–3

1 3 5 7 9 10 8 6 4 2
 Contents

Acknowledgements vii
On the Use of Phonetic Symbols ix

1 Language, Change, and Response 1

2 A Moveable Speech 27
3 Narratives of Change 81
4 Policy and Politics 130
5 Say the Right Thing 186
6 Fixing English 238

Works Cited 267

Index 289
This page intentionally left blank
 Acknowledgements

The chronological and geographical breadth of this book—conXict over

change in the history of English, of all times and all places—has been both
exhilarating and challenging. It has taken me, at least intellectually, to exotic
locales and compelled me to work up material and disciplines with which
I had had only passing familiarity. At the same time, it has introduced me
to new kinds of complexity for the transmission of English and those who
speak it, including the fact that certain kinds of concerns, institutions,
and arguments evince a stubborn persistence across time. And this persistence
has opened up not only new characteristics of English but also important,
if also sometimes contested, shared experiences of those who speak it. If
language since Babel has divided us, it can also provide the means to help
us understand one another better.
For providing me with the opportunity to consider all these ideas and put
them into a book, I am grateful to quite a few people. At OUP, John Davey has
been friend as well as editor, seeing promise in the book’s idea, oVering
encouragement and insight, and eYciently bringing it into production.
I thank the press’s anonymous reviewers for their valuable suggestions and
comments, David Matthews for reading the typescript and improving its
substance and style in many ways, and R. F. Yeager for his continued encour-
agement of my work. Shaun Hughes shared with me his knowledge of Maori
and Maori English, and A. N. Doane oVered several valuable opportunities
to consider and critique the book’s arguments with a learned audience.
I thank Michael A. McKinney, former Dean of Arts and Sciences at Marquette
University, for crafting an administrative schedule that allowed me to draft
the book even as I was still Chair of English, and Krista RatcliVe, for Wlling
in as Chair and allowing that schedule to happen. The Wnal version of the
book was written with the generous aid of a full-year sabbatical from
the Marquette Sabbatical Review Committee and a Summer Stipend from
the National Endowment for the Humanities; any views, Wndings, conclu-
sions, or recommendations expressed in this book do not necessarily reXect
those of the Endowment. Closer to home, I thank Christine, Charlie, and
Tim for indulging my passion for language and keeping me connected to
those who speak it.
Throughout this book I make use of the International Phonetic Alphabet
(IPA), a standardized system for representing sounds irrespective of the
viii Acknowledgements

languages that use them. For readers who are unfamiliar with the IPA, I have
tried to provide contextual information that will clarify what’s being
discussed, and I also include a brief note on phonetic transcription. Some
passages of Middle English use two graphs no longer extant in English:
Z (yogh) for gh and þ (thorn) for th. Citations of the New York Times are to
the Chicago edition, and unless otherwise noted or taken from a published
translation, translations of foreign-language materials are my own.
TWM
March, 2008
 On the Use of Phonetic Symbols

Nearly any English word shows that sounds are very diVerent things
from letters or graphs. A word like this, for example, has four graphs: t, h, i,
and s. The connection between these letters and the sound of the word they
produce is in many ways conventional, since the graph i can sound quite
diVerently in right, as can the graph s in dogs, where it is ‘voiced’ (articulated
with vibrating vocal cords) and sounds more like a z. Conversely, the sound
of the i in this can also be represented by o (women). And th of course uses
two letters to represent one sound, which itself has two versions: a ‘voiced’
one in this and a breathy, ‘voiceless’ one in thin.
To talk about sounds as distinct from letters, linguists have developed their
own abstract system known as the International Phonetic Alphabet. In some
cases, as in p or b, the IPA symbol points to essentially the same sound as that
predicted by English spelling conventions. In others, as in i or e, it can
represent something entirely diVerent. The following few correspondences
should provide enough information for non-linguists to follow any of this
book’s phonological arguments.

IPA symbol comparable sound

i ee in beet
i i in bit
u oo in boot
U u in put
^ u in putt
e ai in bait
e e in bet
o oa in boat
O ou in bought
æ a in bad
a o in cod
A a in father
@ a in about
‚ th in thin
ð th in this
˛ ng in batting
x On the Use of Phonetic Symbols

One Wnal important distinction in phonetic transcriptions is between

phones and phonemes and, concomitantly, between phonetic and phonemic
transcriptions. A phone is simply a sound in a language and is placed between
square brackets, []. A phoneme is an abstraction—it is one of a group of the
smallest distinct, meaningful sounds in a language and one that might
actually be realized by several diVerent phones. It is represented between
slanted lines, //. So [e] refers simply to the vowel sound in bait, but /e/
signiWes that that sound has phonemic status in English. In either case, a
colon after the vowel indicates that its duration is sustained (‘long’); [ma] is
an old-fashioned diminutive for one’s mother, but to [ma:] something, in a
dialect that drops post-vocalic r, is to damage it.
 1

Language, Change, and Response

Accents and attitudes

On 4 July 2000, I received a telephone call from a local radio station that had
gotten my name from my American university’s Experts Directory, where I
was listed as a resource for the history and structure of the English language.
Was it true, the woman inquired, that I was a specialist in English linguistics,
and would I consent to a radio interview about the English used in the
blockbuster movie The Patriot, starring Mel Gibson and then just-released?
What about English in The Patriot, I asked, to which she replied: ‘The
Americans don’t have British accents, and we want to know whether that’s
accurate or not.’
I was to be live on the air in Wfteen minutes, and despite this, and despite
the fact that I hadn’t seen this Wlm about the American War of Independence
(though I since have) and had never really considered the issue, I quickly
agreed to participate. What makes an accent an accent, I thought—a collec-
tion of features that a speaker uses or the perception of a listener? How
determinative of a dialect is an accent? How and why do we make abstractions
like a British accent from the demonstrable varieties of speech at any one time,
thereby suggesting that all people from Britain (or America) speak alike? For
that matter, to what extent would speakers on either side of the Atlantic at the
time of the Wlm’s setting want to sound alike or unalike? How can we reconcile
the various conXicting eighteenth-century comments on the consistent, re-
gionally non-speciWc quality of American speech, on its inferiority to British
speech, on its similarity to the language as spoken in the south and east of
England, and on its status as the future of the language? There were simply too
many intriguing issues to let this interview pass by, and specialists in English
linguistics, after all, get precious few opportunities for radio exposure.
I did some quick fact checking and was ready when the interviewer called
back to repeat her question on the air. I began with typical academic dis-
avowals of the sort: dialect is a combination of various syntactic and lexical
2 Language Anxiety

practices as well as phonological ones; speakers speak for all sorts of reasons of
self-identiWcation and social positioning as well as for the communication of
ideas; the sound of earlier, pre-electronic forms of language is always a
reconstruction; any language embraces social and regional variation, so that
a British or American accent is an abstraction at a fairly high level; any
individual may or may not approximate a regional norm; and so forth.
Then I pointed out that to an American ear, two of the most prominent
features of a loosely deWned British accent were a low central vowel [A] in
words like half and calf and the absence of post-vocalic [r] in words like fourth
and Xoor; in a loosely deWned American accent the vowel would be realized as
the front [æ] and the [r] generally articulated. These phonological diVerences,
I noted, reXect changes that occurred in Britain but not America in roughly
the late seventeenth and early eighteenth centuries, with the dropping of post-
vocalic [r] characteristic of New England due to the later inXuence of British
speech patterns on the American east coast.1 If Mel Gibson’s character in 1776
is presumed to have grown up in the American colonies, I went on, then he
would have been born around 1730. And this means that, in all likelihood, he
would not have had low vowels in half and calf and would not have dropped
the [r] in fourth and Xoor, because these are features that developed in Britain
but not the American colonies. The interviewer’s response surprised me: ‘You
mean that the British are the ones who changed English and the Americans
speak a more traditional, correct form?’ Sensing that the points I had tried to
make were losing their subtlety and precision, I said, ‘Well, for these two
features, at that time, much British but not American speech had changed,
so that in these two features, at that time, yes, much American speech was
more conservative’. ‘So the British version of English is a development of
the American version?’ And my hunch that history, nuance, and theory were
being utterly erased was conWrmed when I hung up the phone and heard the
interviewer and her co-host talk with satisfaction about how fascinating it was
that Americans preserved a traditional accent and that the British were the
ones who had changed the language.
In the end, this episode has become far more interesting to me than has the
question of whether Mel Gibson should have spoken with some version of a
British accent in The Patriot. Bringing together a number of popular and
1 Roger Lass, ‘Phonology and Morphology’, in The Cambridge History of the English Language,
iii, Lass (ed.), 1476–1776 (Cambridge: Cambridge University Press, 1999), 114–16, and John H. Fisher,
‘British and American, Continuity and Divergence’, in The Cambridge History of the English Language,
vi, John Algeo (ed.), English in North America (Cambridge: Cambridge University Press, 2001), 75–7.
Both changes took place over several centuries, with the dropping of post-vocalic [r] still a cause for
comment in the early nineteenth century, and so by ‘occurred’ I mean that they became so extensive as
to be decisive.
 Language, Change, and Response 3

powerful notions of language, the question presupposed much about history,

language, and identity. It is striking, for example, that on American Inde-
pendence Day in the year 2000—a date of iconic if not historic signiWcance for
the United States—a popular radio talk show host should wonder at all about
how the colonists sounded during the Revolutionary War. For the topic to be
newsworthy, or at least chat-worthy, this long-ago accent must have been
thought to have some consequence for Americans as they faced a new
millennium, though it would never have occurred to me, even if I had seen
the Wlm, to raise the question on my own. Another striking presumption is
that languages have a well-deWned constancy and consistency, that a British
accent has an absolute integrity and identiWablity tantamount to the qualities
of a material object, regardless of who views it and when they do so. And this
presumption is held, in part, by avoiding the messy complications of empir-
ical data—involving, for example, the relation between two variable phono-
logical features and an accent—and clinging instead to neat, simple truisms.
Even more striking, by extension, is the implication that an accent serves as a
marker of identity: we are how we speak. And perhaps most striking of all is
the satisfaction inherent in the apparent recognition that Americans, some-
how, speak an older, more original English. Not only is there a real English,
then, but it resides in the United States and not England, where it was
changed.
My radio interview was amiable enough (if also, alas, largely unnoticed), but
this is not always the case when people talk about language variation and change
and their relations to personal and social identity. In fact, for languages in
general and so for English in particular—whether it’s the English of the United
Kingdom, the United States, New Zealand, or any other part of the Anglophone
world—language change throughout the language’s history has consistently
Wgured as both tool and symbol in acrimonious conXicts over education,
immigration, morality, and so forth. When English speakers have lamented
the condition and goals of schools, they have cited the state of language usage
as both cause and consequence of this condition. When they have worried over
the integrity or diversity of society, they have argued for and against the
persistence of non-English languages. And when they have identiWed signs of a
literal or Wgurative apocalypse, they have pointed to language and how it, in
turn, points to social and moral issues beyond itself. While English as such
may never have caused a war, it has served as a cudgel to speakers of Celtic
languages, Native Americans, and immigrants alike, whose own languages have
been seen as communicatively inadequate, socially inhibiting, morally debili-
tating, and even unpatriotic. Bilingual education, dictionaries, and language
planning—all of them responses to language variation and change—have been
4 Language Anxiety

both burdensome and contentious. For those involved in their production and
management, such activities have been lucrative as well, and the very fact that
language change can provide economic opportunity itself reXects just how
seriously Anglophones have historically taken the issue.
Language change produces, in a word, anxiety. And perhaps a better word
than produces here is focuses, for part of the anxiety is in fact over the nature of
the relationship between language change and transformation in the extra-
linguistic worlds of politics, religion, and social interaction. Does language
change portend changes in these worlds? Does it cause them? Is it the direct
result of other kinds of change? Or does it merely accompany them? All this
anxiety is already present in the earliest Western account of language change,
that of the Tower of Babel, in which change and variation are Wgured as the
divine punishment visited by God on humans for their arrogance and pre-
sumption. It is likewise already present in England in Alfred the Great’s
preface to his ninth-century translation of St Gregory’s Pastoral Care, in
which he cites the decline of Latin and English literacy as both sign and result
of the general moral failure into which the Viking raids had plunged England.
And this anxiety over change and variation has remained present through
modern English’s encounters with non-western languages and their speakers
and through the recognition, driven home by standardized written English,
that even among native Anglophones language continues to diversify.

Theories of change and variation

Underlying my discussion so far are two of the best-established and least
disputable facts in social history. One is the fact of synchronic language
variation—the existence of grammatical diVerences (in lexicon, syntax, phon-
ology, and morphology) among contemporaneous varieties of a language,
such as modern southern British English and modern southern United States
English. Where the former has lift, the latter has elevator; where the former
uses the plural are with mass nouns like government, the latter uses the
singular is. The second fact is that of diachronic change, or change across
time within the same language or dialect.2 Thus, while medieval Londoners
used -eth ([E‚] or [Eð]) as the third person singular verbal inXection, modern
ones use -s ([Es] or [Ez]), making for a peace that passes rather than passeth
understanding. And while the Anglo-Saxons used déor to mean ‘four-footed

2 Conventionally, dialect is used to refer to mutually intelligible social or regional forms of the same
language. I will use these terms in this same way throughout this book, but most of the time I prefer
variety as a neutral term that encompasses both dialect and language and thereby avoids the pragmatic
and grammatical diYculties of distinguishing one from the other.
 Language, Change, and Response 5

beast’, contemporary Anglophones typically use deer, its modern reXex, to

refer to speciWc kinds of quadrupeds—ruminants from a family (cervidæ)
with deciduous antlers, such as reindeer and whitetail deer.
Scholars may argue about the origins of the universe or the demise of
dinosaurs, politicians may dispute environmental causes and eVects, and
educators may contest the implications of standardized test scores. But
these facts of change and variation, as I say, are indisputable, for nearly as
early as the development of writing there is evidence both of them and also of
speakers’ awareness of them. It is variation among languages to which Homer
nods in the Odyssey, when a disguised Odysseus, on his return to Ithaca,
fabricates for his unwitting wife Penelope an account of his origins on Crete,
an island that has ‘many j peoples in it, innumerable; there are ninety cities. j
Language with language mix there together’.3 In the Book of Judges, variation
among the dialects of Hebrew enables the men of Gilead to identify the Xeeing
Ephraimites, with catastrophic consequences:
The Gileadites captured the fords of the Jordan leading to Ephraim, and whenever a
survivor of Ephraim said, ‘Let me cross over’, the men of Gilead asked him, ‘Are you an
Ephraimite?’ If he replied, ‘No’, they said, ‘All right, say ‘‘Shibboleth’’ ’. If he said,
‘Sibboleth’, because he could not pronounce the word correctly, they seized him and
killed him at the fords of the Jordan. Forty-two thousand Ephraimites were killed at
that time.4

Looking back to the Antique over a millennium later in his Troilus and
Criseyde, GeoVrey Chaucer imagines more beneWcial consequences of these
facts of language change and variation. In response to an audience that might
express skepticism over how his lovers conducted themselves in word and
deed, he observes with particular eloquence:
Ye knowe ek that in forme of speche is chaunge
Withinne a thousand yeer, and wordes tho
That hadden pris, now wonder nyce and straunge
Us thinketh hem, and yet thei spake hem so,
And spedde as wel in love as men now do;
Ek for to wynnen love in sondry ages,
In sondry londes, sondry ben usages.5

3 Homer, The Odyssey, trans. Richmond Lattimore (New York: HarperCollins, 1963), 19.173–5.
4 Judges 12: 5–6.
5 Troilus and Criseyde, 2.22–8, The Riverside Chaucer, ed. Larry D. Benson, 3rd edn. (Boston, MA:
Houghton MiZin, 1987).
6 Language Anxiety

As well established as the facts of language change and variation may be, the
reasons behind them are a good deal more contested, even murky. All theories
must account for the same aspects of variation and change, including: gram-
mar (phonology, morphology, syntax, lexicon, and pragmatics), styles, strat-
egies, and discursive conventions; alterations in an individual’s or society’s
repertoire, such as the multilingualism of Homer’s Crete; inconsistent regu-
larity in the creation and transmission of forms and varieties; the relations
between the speech of individuals and the aggregate language they use; and
the impact of context, including language planning, on usage. Perhaps not
surprisingly, this abundance of linguistic features in need of explanation has
produced an abundance of explanations. According to Jean Aitchison, ‘For
centuries, people have speculated about the causes of language change. The
problem is not one of thinking up possible causes, but of deciding which to
take seriously.’6 Whether it is the biblical account of the Tower of Babel,
attributing diachronic change to divinely righteous indignation, or Dr John-
son’s representation of synchronic variation, explaining neologisms as words
that authors ‘have introduced by their knowledge of foreign languages, or
ignorance of their own, by vanity or wantonness, by compliance with fashion,
or lust of innovation’,7 theories of linguistic change have indeed been as
imaginative as they have been proliWc. Very generally, contemporary ap-
proaches emphasize one of two ends of an explanatory spectrum, ascribing
them primarily either to systemic pressures or to social ones.
Systemic explanations stress the structural characteristics of language, the
ways in which various grammatical parts, by means of their contrastive
distinctions with one another, mutually constitute a coherent and functional
whole. What gives this whole meaning, and what allows speakers to make
meaningful utterances with it, is the system of the language, and in such
mechanistic theories speakers are not merely separate from but secondary to
the language they speak. In Roger Lass’s memorable phrase, linguistic change
‘occurs over ‘‘geological’’ time, beyond the capacity of humans to act, since no
actor can see the consequences of his actions. A speaker engaged in a change is
not an agent, but a victim.’8 As Lass’s reference to ‘geological time’ suggests,
systemic explanations situate a good deal of the explanation in metaphors

6 Aitchison, Language Change: Progress or Decay?, 3rd edn. (Cambridge: Cambridge University
Press, 2001), 133.
7 Johnson, A Dictionary of the English Language (1755; rpt. New York: AMS, 1967), i, bir .
8 Lass, Historical Linguistics and Language Change (Cambridge: Cambridge University Press, 1997),
367. Lass is synthesizing points initially made in Edward Sapir, Language: An Introduction to the Study
of Speech (New York: Harcourt-Brace, 1921), 165–6.
 Language, Change, and Response 7

from the hard sciences, if not in the hard sciences themselves. While most
scholars would point to the anatomical developments of the head, mouth,
and chest as prerequisites for the development of human speech, for example,
mechanistic theories of change describe much closer, more intrinsic connec-
tions between language and biology. In this way, mechanistic theories reveal
the mutually reinforcing connections between Darwinian evolution and
comparative linguistics that have been present since the simultaneous nine-
teenth-century developments of both disciplines.9 Derek Bickerton thus sees
language change as intimately situated within evolutionary and social change.
He posits a proto-language among early hominids that, with the development
of syntax, sequentially enabled fully human language, meta-thinking (or
double consciousness), social sophistication, and the increased record of
artifacts in the period 100,000 to 50,000 bc.10 One similar evolutionary theory
of change imports the concept of punctuated equilibrium, by which long
periods of structural stasis of organisms are interrupted by sudden periods of
transformation, to describe the history of recorded (and unrecorded) lan-
guages.11 Another, likening languages to species and individuals’ grammars to
the members of those species, conceives language change as the collective
change of individual grammars; in this framework, it is language change that
is gradual and grammatical change sudden.12 Even more generally, Nikolaus
Ritt, understanding systemic replication as a kind of linguistic survival im-
perative, suggests that languages don’t so much change as evolve, using
speakers as hosts and building new, coherent systems from the incompletely
learned or transmitted earlier systems. For Ritt, ‘the apparent similarities
between languages and biological life forms are no coincidence . . . Instead,
they reXect deeper and more general design principles, which may character-
ise the organisation of many systems within the universe.’13
Social explanations of language change, exempliWed by the work of William
Labov or Peter Trudgill, may grant that the process of a change—the way a

9 David Lightfoot, The Development of Language: Acquisition, Change and Evolution (Malden, MA:
Blackwell, 1999), 21–47.
10 Bickerton, Language and Human Behavior (Seattle, WA: University of Washington Press, 1995). A
comparable approach is that of Robbins Burling in The Talking Ape: How Language Evolved (Oxford:
Oxford University Press, 2005). Rather than stress the development of human speech, however,
Burling stresses the development of human understanding, for it is his argument that when the
utterances of early humans were invested with meaning, this investment encouraged their further use
and role in social development.
11 R. M. W. Dixon, The Rise and Fall of Languages (Cambridge: Cambridge University Press, 1997).
12 Lightfoot, The Development of Language.
13 Ritt, SelWsh Sounds and Linguistic Evolution: A Darwinian Approach to Language Change (Cam-
bridge: Cambridge University Press, 2004), 110.
8 Language Anxiety

phonological alteration takes place, for example—is describable in mechan-

istic terms that reXect Ritt’s ‘general design principles’. The collective changes
known as the Great Vowel Shift, thus, are often described as representing a
push chain, whereby once the mid-high front and back long vowels /e/ and /o/
moved up in the mouth’s phonological space, they pushed the falling and
diphthongization of the highest front and back long vowels (/i/ and /u/) and
dragged along the next lowest long vowels (/E/ and /O/) into the space they
had left; these newly made mid-vowels, in turn, dragged the next lower long
vowels into their own, previous space. In simpler terms, it’s as if the vowel
sound of fate rose in the mouth until it was the same as feet, which pushed the
original feet to sound like Wght. Since a push-chain mechanism like this has
occurred in other languages at other times, it can at least be understood to
reXect a general, systemic principle of organization for language. Where social
explanations of change diVer from systemic ones is that they go beyond such
strictly mechanical principles to highlight the uniquely human features of
speech, its use, and transmission. In the case of the Great Vowel Shift, Jeremy
Smith and others have argued that the real explanation of why vowel raising
and diphthongization occurred lies in increased immigration to London,
particularly from East Anglia, in the late fourteenth and Wfteenth centuries.
This immigration, they say, brought diVering phonologies into contact with
one another, and with the matching of phonology to issues of solidarity and
power—whether a speaker wants to be associated with a group or distin-
guished from it—the phonologies themselves came to have social valuations,
thereby causing speakers to alter their pronunciations towards or away from
particular speech patterns. In this way, it is the relative social prestige of
speakers and their varieties that drives variation and change.14
More generally, sociolinguists have identiWed a number of social factors
and contexts that demonstrate, to the extent that one can control the vari-
ables, intimate associations with various kinds of grammatical change. Like
phonological mechanisms, these associations recur with suYcient frequency
cross-linguistically as to merit consideration as sociolinguistic universals. The
prestige of a given variety, as has been alleged in the Great Vowel Shift, can
indeed Wgure signiWcantly in change, as can the social desire to stigmatize or
ameliorate a particular word. The noun Miss, for instance, has become
marked in many varieties of modern English as a designation for an adult,

14 See Jeremy J. Smith, An Historical Study of English: Function, Form and Change (London:
Routledge, 1996), 79–111; and Sound Change and the History of English (Oxford: Oxford University
Press, 2007), 127–53. More generally, see Lass, ‘Phonology and Morphology’, 72–85.
 Language, Change, and Response 9

unmarried woman and has consequently been restricted to young girls, while
gay has in most domains lost its earlier pejorative connotations in reference to
homosexuals and become widespread as a neutral term. The history of gay
provides a particularly rich illustration of how the connotations of words can
shift in response to social pressure. The word began in the fourteenth century
as an adjective describing light-heartedness, then by the seventeenth century
designated a dissipated lifestyle (particularly that of prostitutes), and then,
before its present usage as a noun, referred derisively to homosexuals.15
Another social source of change, especially the introduction of new vocabu-
lary, is contact with non-native speakers and the materials of their world; this
was the case when English speakers encountered Native Americans and
borrowed such words as caribou, hickory, and moccasin.
Change—or better, perhaps, resistance to change—can also arise from the
explicit desire of members of a group to strengthen their solidarity by devel-
oping words and syntax that are intentionally obscure to outsiders. This is what
M. A. K. Halliday calls ‘the sociolinguistic play potential of one’s own variety
of the language’16 and has been documented in Philadelphia, New York City,
Belfast, and other urban environments, where the covert prestige of a non-
standard variety may inXuence speakers to preserve it at the expense of
an overtly prestigious standard and whatever social opportunities that stand-
ard might seem to provide. Conversely, when nineteenth-century speakers of
Irish overwhelmingly switched to English as the language of both workplace
and home, they revealed the extent to which economic opportunity (if not
social prestige) and perceived connections between it and language can drive
change. The relative social status of women and men has also been identiWed
as a primary motivation for linguistic maintenance or change, with many
studies suggesting that working-class women, perhaps better attuned to the
possibility of social advancement, are more likely to speak a variety closer to a
standard language than are working-class men. Still other studies have iden-
tiWed socially marginal groups, including adolescents, as the ones whose weak
ties to a community bring them into contact with other speech communities
and who risk no negative social consequences that would discourage them
from transferring the words or practices of one community to another. For
this reason, adolescents and other weakly tied speakers are sometimes iden-
tiWed as the instigators of change and variation. In short, any of the social
characteristics and factors that one can imagine—age, sex, class, ethnicity,

15 OED, s. v. gay, adj., adv., and n.

16 Halliday, Language as Social Semiotic: The Social Interpretation of Language and Meaning
(London: Edward Arnold, 1978), 160.
10 Language Anxiety

geographic origin, education, and so forth—might induce speakers to change

their language in one way or another, and other speakers, in turn, to follow or
resist them.17
In view of such diverse explanations for linguistic variation and change, the
desire to throw up one’s hands and explain language and its changes as neither
mechanistic nor social, neither natural nor artiWcial, but as the proverbial
third kind of thing altogether is not surprising.18 In one way or another, all of
the approaches I’ve outlined must presume an answer to the same funda-
mental question: What is language? Perhaps even more challenging than ‘Why
does language change?’ this is ultimately a philosophical question, and un-
answerable in absolute terms, too, for any response is inevitably a hypothesis.
And for this reason, explanations of the ontology of language have been
as diverse as have been those of language change. To philosophers from
Descartes to Wittgenstein to Derrida, for example, language exists in a
dynamic with extra-linguistic reality, which it variously processes or
constructs, while for linguists it tends to be more of an abstract code that
maps representations (sounds and words) onto objects and ideas.
Perhaps the most challenging ontological feature of language involves the
relation between what an individual might say and what the collective of
individuals speaking a given language might say. Is English what I speak, or
what the majority of people in the United Kingdom speak, or an abstraction
of what all the Anglophones in the world speak? And if the latter, how do we
account for the diachronic and synchronic diVerences among all of us so as to
be able to maintain that we all speak the same language? Similarly, how can
English change for me as well as for hundreds of millions of other Anglo-
phones whom I’ll never meet and to whom I’ll never speak? Oddly enough, an

17 Some representative and particularly inXuential illustrations of the social approach to lin-
guistic change are: John J. Gumperz and Dell Hymes (eds.), Directions in Sociolinguistics: The
Ethnography of Communication (New York: Holt, Rinehart and Winston, Inc., 1972); Halliday,
Language as Social Semiotic; Hymes’ Foundations in Sociolinguistics: An Ethnographic Approach
(Philadelphia, PA: University of Pennsylvania Press, 1974); William Labov, Sociolinguistic Patterns
(Philadelphia, PA: University of Pennsylvania Press, 1972); Labov, Principles of Linguistic Change, i,
Internal Factors, ii, Social Factors (Oxford: Blackwell, 1994, 2001); James Milroy, Linguistic Variation
and Change: On the Historical Sociolinguistics of English (Oxford: Blackwell, 1992); Lesley Milroy,
Language and Social Networks, 2nd edn. (Oxford: Blackwell, 1987); Salikoko S. Mufwene, The
Ecology of Language Evolution (Cambridge: Cambridge University Press, 2001); Suzanne Romaine,
Socio-historical Linguistics: Its Status and Methodology (Cambridge: Cambridge University Press,
1982); Peter Trudgill, The Social DiVerentiation of English in Norwich (Cambridge: Cambridge
University Press, 1974); Trudgill, Sociolinguistic Variation and Change (Edinburgh: Edinburgh
University Press, 2002).
18 Rudi Keller, On Language Change: The Invisible Hand in Language, trans. Brigitte Nerlich
(London: Routledge, 1994).
 Language, Change, and Response 11

old textual-critical chestnut illuminates this ontological question with par-

ticular clarity. Addressing the ephemerality of a work of literature, the text of
which may appear in multiple documents, F. W. Bateson once asked: ‘If the
Mona Lisa is in the Louvre, where are Hamlet and Lycidas?’ Like Shakespeare’s
play or Milton’s poem, which can be located in no one place but instead found
in multiple books in the libraries of multiple readers, the English language is
clearly limited to no one speaker or locale. But language is perhaps even more
complex in this regard than are literary works. Unlike Hamlet or Lycidas,
which are uniform in at least every copy of a particular edition, iterations of
English diVer from one time period to another, from one locale to another,
from one speaker to another, and even, for that speaker, from one occasion to
another.
Further, the facts that help to construct the ontologically problematic fact
of English are themselves sometimes intractable, and this intractability ac-
counts for a fundamental Wssure in popular as well as professional deWnitions
of language. On one side are those who see the language in the utterances that
real people produce in real circumstances, replete with irregularities in
pronunciation, word order, and phrasing, and veriWable in the spoken or
printed record of English. On the other are those who conceive the language
as an abstract system that is itself regular, even if it is irregularly realized by
speakers; this abstraction would be Ferdinand de Saussure’s langue or Noam
Chomsky’s ideal speaker-listener, and its data come from the knowledge and
intuitions of its analysts. The nature of individual linguistic facts (such as in
phonology) complicates even more this paradigmatic dichotomy and its
relevance to the ontology of English. For example, a sound like [p], found
in sip, is one kind of fact in English; this sound and the several variants of it
are scientiWcally veriWable in the language through tools such as voice
spectography, even if a speaker might be unaware of this veriWability or
even the sound itself. The phoneme /p/ is an entirely diVerent kind of fact.
This exists only through an analysis that groups together a variety of sounds
(such as the [p] of sip and the aspirated [ph ] of pull) that are related in
articulation (both are bilabial, voiceless stops) but cannot be used individu-
ally in a given language to diVerentiate one word from another, as the voicing
of [p] as [b] can diVerentiate bull from pull in English. Without aspiration of
/p/, pull might sound funny, but it would still be pull; with voicing, it would
be another word. And yet a third kind of fact is something like the push-
chain mechanism that I described above. As Lass has noted, in a case like the
Great Vowel Shift, several linguistic facts (each already slippery) are under-
stood to be part of a coherent process: ‘The impact of the ‘‘establishment’’ of
the GVS is this: once such an object as what we could now call a ‘‘covarying
12 Language Anxiety

shift’’ is invented, a whole set of new questions becomes askable for the
Wrst time.’19
Philosophical speculation like this has relevance, I think, to the nature of
language change and also, therefore, to responses to it, which are the concerns
of this book. Given any distinction between English and the speakers of
English, we need to account for how change can move among speakers at
any one time—how English can change for both New Zealanders as well as
South Africans, if in fact it does—since speakers develop their language
primarily from individual contacts and their own native language capacity
and not from consultation of books or some kind of a central linguistic
clearing house. By extension, we need to determine how to draw distinctions
between historical stages of a given language (Old English versus Middle
English, for example) or between structurally similar languages (such as
Danish and Faeroese). And precisely the same complex issues appear when
we attempt to diVerentiate the variations peculiar to any individual (i.e.,
idiosyncratic forms) or to any group of individuals (i.e., dialect forms)
from changes in the aggregate language. This might be done in several ways.
We might, for example, distinguish dialect forms from forms in the early
stages of diVusion through a language by the number of speakers using a
form, by their social prestige, by their attitude, or by the duration of the
form’s usage. Each of these criteria would produce diVerent, contestable
results. In the contemporary United States, for instance, dude is widespread
among younger speakers as (sometimes) a slightly ironic term of endearment.
Measured solely by the percentage of speakers who use the form, given the
overall youth of the American population, dude might well qualify as an
ongoing lexical change for United States English, but this would not be the
case if the prestige of these speakers, or the percentage they represent of all
living Anglophones, were taken into account. The same is true for guys, which
the same cohort (and this cohort alone) widely uses in reference to groups of
men and women of all ages.
Given the character of linguistics as a discipline, philosophical resolutions
of such linguistic problems are inevitable. First, there is not, nor ever can there
be, an empirically demonstrable or even simply accepted quantity of struc-
tural features that by itself marks the diVerentiation of one language or dialect
from another. There is no formula that dictates x number of unique phono-
logical features, plus y number of syntactic features, plus z number of

19 Lass, Historical Linguistics and Language Change, 36. More generally see his On Explaining
Language Change (Cambridge: Cambridge University Press, 1980).
 Language, Change, and Response 13

morphological ones create a language; or that a decrease of x can be oVset by a

speciWc increase of y; or that if x, y, and z are diminished by some value, the
result is not a language but a dialect. This is so because in a very real sense
outside the realm of structural linguistics, diVerentiation of dialects and
languages draws on speakers’ attitudes as well as their speech forms. Faeroese
is thus a language and not a dialect of Danish in part because of the role it
played in nineteenth- and twentieth-century nativizing movements in the
Faeroes and in part because the Faeroese themselves regard it as a distinct
language and have cultivated it accordingly.
The second factor that compels philosophical resolutions of some linguistic
problems is that linguistics, unlike a hard science, cannot isolate its subjects in
a laboratory, control for all variables but one, and replicate situations in order
to verify results. In the twenty-Wrst century it cannot take a medieval popu-
lation from London, introduce a late-medieval population from East Anglia,
and watch to see whether social diVerentiations and something like the Great
Vowel Shift develop. This means that all of its explanations lack the predictive
power of those from biology, chemistry, and physics. And this leaves the
linguist needing to Wnd English even more desperately than the literary critic
needs to Wnd Hamlet.
From another fundamentally philosophical perspective, whether one under-
stands the English language as an abstract system or as the conXicting and
inconsistent usages of its speakers, any kind of change seems counter-intuitive.
Optimal communication, an often-cited motivation for change, makes this
very point. When the Old English long vowels /e:/ and /e:o/ merged in the tenth
and eleventh centuries as /e:/, for instance, homophony would have arisen
between the personal pronoun used for males (hé) and the one used for females
(héo). On the presumption that such homophony would create communica-
tive confusion, scholars have cited a drive for optimal communication—for
the need to maintain crucial semantic distinctions—as the factor that led to the
new form she, presumably through palatalization of the initial [h] in héo.
Theoretically and practically, this makes a good deal of sense. But while
optimal communication might have strong explanatory power in the very
narrow context of vowel mergers, its general explanatory power is limited. If
even a subconscious desire for optimal communication is pre-eminent, then
once this communication has been reached, additional change would seem to
be both superXuous and counter-productive. And since by most theories
humans have been speaking for at least 100,000 years, there would seem to
have been plenty of time to achieve optimality.
Yet variation and change continue, and the results of change occasionally
render its persistence even more counter-intuitive. While biological evolution
14 Language Anxiety

is the process by which random variation produces contextually successful

features, which are then ‘selected’ in the sense that they are the ones that
survive, linguistic change sometimes has strikingly unsuccessful results. In-
deed, as Labov has noted, language actually points to conclusions that oppose
natural selection: ‘the major agent of linguistic change—sound change—is
actually maladaptive, in that it leads to the loss of the information that the
original forms were designed to carry’.20 More generally, change and variation
are responsible for a great many socially debilitating situations. They produce
mutually unintelligible languages and their attendant barriers to communi-
cation, the communicative obstacles that even regional variation can present,
and the sociolinguistic drive to instruct generation after generation of stu-
dents in the details of spelling, punctuation, and usage, which are never
internalized and transmitted to subsequent generations in some Lamarckian
fashion. In view of the tumult of history and the blame placed on inadequate
communication, I would venture that if there truly is a general drive to
optimal communication, it has failed miserably.
Linguists sometimes neutralize the counter-intuition of change and variation
by drawing a distinction between teleology and direction. They say that when
one sound becomes another, the transformation may be directional—[e] may
move in the direction of [i], for example—but it is not teleological, because it
does not go towards any speciWc end for any speciWc purpose. As with a found
fact like a push-chain shift, however, whatever issues such a distinction resolves,
it also enables new questions about why change occurs, where, and when. And in
so doing, it also leaves unanswered questions that emerge from other found
facts, such as how or why such a non-directional change could be irreversible, a
quality commonly attributed to all linguistic change. If change just happens, that
is, it would seem just as likely as not that a sound would happen to become once
again what it once was. Nor does this distinction between teleology and direction
answer questions that challenge its own validity as a fact. In the most common
push-chain explanation of the Great Vowel Shift, to which I’ve already alluded,
the restoration of balance in phonological space—the distribution of vowels by
height and frontness—would very much seem to be a goal for the changes (once
begun) and not simply their direction. And this quality, like all linguistic
versions of Ritt’s ‘general design principles’, would seem once more to reXect
at least a kind of teleology.
From a pragmatic as well as a philosophical view, indeed, perhaps the most
challenging ontological issue is that linguistic change and variation don’t

20 Labov, Principles of Linguistic Change, ii, Social Factors, 10.

 Language, Change, and Response 15

simply persist. They thrive. It’s worth recalling here that while animal life has
been on earth for well over 100 million years, humans have spoken for perhaps
only 100,000 of these years. Broadly speaking, then, if the language faculty is at
least partially biological, as (again) most linguists would argue, the extent of
the structural diVerences among the roughly 6,000 to 6,500 languages that are
current today would seem to exceed natural diVerentiation of a shared
feature, particularly (but not only) if they all developed from the speech of
a common population.21 It could even be argued that change and variation
have exceeded any rational explanation of what causes them or, consequently,
how they might be curtailed. More narrowly, what the Oxford English Dic-
tionary shows diachronically, a journey across any sizeable Anglophonic
region shows synchronically: however counter-intuitively, change and vari-
ation have continually outweighed any systemic and social advantages of
linguistic uniformity. Such change is in fact so extensive and intrinsic to
languages that for Lass and his mechanistic approach to change, English or
French or Latin are not so much stable communicative codes as loosely
aYliated sets of changes: ‘A language is a population of variants moving
through time, and subject to selection.’22 In this same vein, the recent school
of ‘emergent linguistics’ emphasizes that whatever syntactic and phonological
rules govern human speech do not precede language but emerge variably in
the act of speech.23 And inherent variation and change are the implications,
signiWcantly, not only of evolutionary models that see languages as replicating
systems subject to the same kinds of mutation of all such systems, but also of
the static Chomskyan model. One of the original insights of generative
grammar, thus, is that any theory of language must account for the fact that
speakers have the recursive ability to reuse syntactic structures and thereby to
produce grammatically correct sentences that nonetheless have never been
uttered before. As rule-governed as transformations may be, their primary
explanatory focus is variation.
Nearly a century ago, Robert Bridges, then England’s Poet Laureate, argued
that the advent of the wireless radio would control the evolution of speech
and be conducive to the standardization of English. It didn’t happen, of

21 See Daniel Nettle, Linguistic Diversity (Oxford: Oxford University Press, 1999), 2–3.
22 Lass, Historical Linguistics and Language Change, 377.
23 See Brian MacWhinney, ‘Emergent Approaches to Language’, in Joan Bybee and Paul Hopper
(eds.), Frequency and the Emergence of Linguistic Structure (Amsterdam: John Benjamins, 2001),
449–70. See also Charles-James N. Bailey’s argument for ‘developmentalism’, or the notion that change
and variation are not simply fundamental to language but rather are language (Essays on Time-Based
Linguistic Analysis [Oxford: Clarendon Press, 1996]). I return brieXy to this issue in the second chapter.
16 Language Anxiety

course. Indeed, as Labov noted towards the end of the twentieth century,
‘change is continuing at a rapid rate in every city of North America that has
been studied with any care. This result clashes sharply with the common-
sense expectation that constant exposure to the network standard on radio
and television would lead to convergence and the gradual elimination of
dialects.’24 Whatever their origins, powerful diversifying mechanisms must
indeed be at work.

From sentence structure to social upheaval

‘Tongues, like governments, have a natural tendency to degeneration,’ Dr
Johnson once observed; ‘we have long preserved our constitution, let us
make some struggles for our language’.25 This linking of language, govern-
ment, and degeneration demonstrates Johnson’s characteristic wit, but it also
highlights the anxiety that can be produced by language change. In Johnson’s
formulation, governments are like the lost boys in William Golding’s Lord of
the Flies: if left to their own devices, they will evidently by nature regress to
their unprincipled, undisciplined condition. And in view of this inherent drift
towards decline, the struggle against regression is arduous and constant,
requiring the unvarying attention of the governing and governed alike. In
comparing language to government, Johnson implies that the two share a
constructed nature, that they are designed and produced by human beings for
the beneWt of other human beings. For language to be like government,
further, it must share not only government’s natural qualities but also its
natural tendency to degeneration, which needs to be resisted as much as
governmental decline is resisted. Tantamount to one another, language and
government (for Johnson) are members of the same natural category.
To be sure, not every scholar or popular commentator shares Johnson’s
anxiety. During the nineteenth-century development of comparative linguistics,
linguists like Max Müller, the Wrst Professor of Comparative Philology at
Oxford, saw the reduction of inXectional morphology among Indo-European
languages, English in particular, as unambiguous improvement. Otto Jespersen,
one of the last of these critics, once tied this improvement to improved
interpersonal communication in particular: ‘In the evolution of languages the

24 Labov, Principles of Linguistic Change, ii, Social Factors, 3; Bridges, ‘The Society’s Work’, Society
for Pure English, Tract XXI, excerpted in W. F. Bolton and David Crystal (eds.), The English Language,
ii, Essays by Linguists and Men of Letters, 1858–1964 (Cambridge: Cambridge University Press,
1969), 93–4.
25 Johnson, A Dictionary of the English Language, i, ciiv.
 Language, Change, and Response 17

discarding of old Xexions goes hand in hand with the development of simpler
and more regular expedients that are rather less liable than the old ones to
produce misunderstanding.’26 And some recent critics have enthusiastically
embraced both the modern development of world Englishes and the apparent
resurgence of non-English languages (or non-standard varieties of English) in
largely Anglophone populations.27 Without going so far as to connect change to
improvement, Lass and still other linguists maintain that at least in the abstract,
change is neither positive nor negative: ‘Languages are imperfectly replicating
systems and therefore throw up variants during replication; the fact of variation
itself is neutral.’28 Nevertheless, the anxiety expressed by Johnson and the
suspicion implied by my radio interviewer have insistently animated responses
to changes in English, whether the changes involve the structure of the language
or its relations with other languages.
One of the improbable best sellers of 2004, for instance, was Lynne Truss’s
Eats, Shoots & Leaves, a diatribe, somehow both witty and irascible, that
linked together an eclectic group of putative changes—speciWcally de-
clines—in punctuation, education, communication, and standards in general.
Rife with emotive images like a ‘zero tolerance for bad punctuation’, a
‘dismally illiterate world’, and a ‘swamp from which [English] so bravely
crawled less than two thousand years ago’, Truss’s book captured popular
imagination (and hysteria) by reviving Johnson’s view of language change—
in this case, alleged changes in punctuation practices—as a degeneration
homologous with the degeneration of the world at large: ‘The reason it’s
worth standing up for punctuation is not that it’s an arbitrary system of
notation known only to an over-sensitive elite who have attacks of the vapors
when they see it misapplied. The reason to stand up for punctuation is that
without it there is no reliable way of communicating meaning.’29 More
ambitious in scope and execution, another widely read response to language
change and variation (John Honey’s) concentrates on the values to be gained
from fostering linguistic uniformity. He sees a reduction of dialects as an
ethically good thing, since ‘accent diVerences are one of the greatest obstacles
to genuine social equality’ in England and ‘causing children to learn Standard
English is an act of empowerment which will give them access to a whole

26 Jespersen, Language: Its Nature, Development and Origin (London: G. Allen & Unwin, 1922), 263.
Also see Lass, Historical Linguistics and Language Change, 292.
27 On the cultivation of the plural Englishes as an academic discipline, see Tom McArthur, The
English Languages (Cambridge: Cambridge University Press, 1998), 56–77.
28 Lass, Historical Linguistics and Language Change, 354.
29 Truss, Eats, Shoots & Leaves: The Zero Tolerance Approach to Punctuation (New York: Gotham,
2004), 20.
18 Language Anxiety

world of knowledge and to an assurance of greater authority in their dealings

with the world outside their own homes, in a way which is genuinely
liberating’. The ‘function of prescription’, then, ‘is not the prevention of
change but rather the management of change—a process of control which
allows change to be seen as an orderly process’.30
Truss and Honey further a popular tradition of linguistic anxiety that
Deborah Cameron has aptly labeled ‘verbal hygiene’ and to which I’ll return
in Chapter 5.31 This tradition of popular anxiety over language change
remains as healthy today as it did when Cameron coined the term; for that
matter, since Johnson was not an academic but a popular critic, the tradition
was healthy as far back as the eighteenth century. Every year, perhaps im-
probably, Lake Superior State University in Sault Ste. Marie, Michigan, issues
its ‘List of Words Banished from the Queen’s English for Mis-Use, Over-Use
and General Uselessness’.32 While the list is compiled in fun—and fun it most
certainly is—its appeal for over thirty years now rests on readers’ suspicion of
new words, like blog and carbs, much as my radio interviewer’s response drew
on a suspicion of new pronunciations and dialect variation. Similarly, when in
a 1999 Washington Post column Bob HirschWeld whimsically announced the
existence of a new computer virus—the Strunkenwhite virus, which would
prevent the transmission of any emails containing grammatical errors—his
satire had eVect and humor precisely because computer users could be
presumed to be aware of the divergence between traditional written Standard
English (as advocated in Strunk and White’s landmark Elements of Style) and
the casual variety that has developed for Internet communication.
Such suspicion cannot be dismissed as merely a popular fad, for it is shared
by linguists, too. What Müller saw as morphological improvement, Franz
Bopp, his rough contemporary and like him one of the founders of compara-
tive linguistics, saw as decay. More recently, John McWhorter, in his own best-
selling book whose subtitle contains the phrase ‘The Degradation of Language’,
has described modern America as having changed its attitudes towards lan-
guage in such a way that it has lost the ability to produce, understand, and
enjoy any rhetoric but the colloquial. By tracing what he sees as the poverty of
political discourse to this rhetorical change, McWhorter agrees with Truss and
Johnson about the linking of language and government: ‘Modern America,

30 Honey, Does Accent Matter? The Pygmalion Factor (London: Faber and Faber, 1989), 174, and
Language is Power: The Story of Standard English and its Enemies (London: Faber and Faber, 1997),
42, 147.
31 Cameron, Verbal Hygiene (London: Routledge, 1995).
32 [Link]
 Language, Change, and Response 19

then, is a country where rigorously polished language, of a sort only possible

when channeled through the deliberate activity of writing, is considered
insincere. And this is not a Yankee keystone, but a trait only a few decades
old. And it leaves us culturally and even intellectually deprived.’33 David
Crystal invests the same degree of signiWcance in change and variation, though
from a diametrically opposed perspective. Writing speciWcally in response to
Truss and other verbal hygienists, he champions the diversity they attempt to
control. His book, Crystal says, tells ‘the story of the Wght for English usage—
the story of a group of people who tried to shape the language in their own
image but, generation after generation, failed. They looked at language around
them, and didn’t like what they saw. ‘‘Fight’’ is not my metaphor, but theirs.’34
Linguists’ anxiety about language change, however, characteristically fo-
cuses on issues of social and even global upheaval rather than on grammar
and usage. The spread of English as a world language has been a particular
concern for analysts who see the issue as not simply linguistic but cultural and
economic. Provocatively titled books like Language in Danger: The Loss of
Linguistic Diversity and the Threat to Our Future and Language Death—the
cover of which displays the heartbeat line from an EKG, presumably on the
point of cessation—oVer statistics on the worldwide decrease of languages
since the expansion of English and American interests in the past several
centuries. By one estimate 75–90% of the languages spoken today will disap-
pear by the next century; according to another, by 2070 ‘fewer than one-tenth
of the languages spoken [in North America] before European contact will still
survive’; and according to still another, of the 250 Aboriginal languages
spoken when Britain annexed Australia in the late eighteenth century, 90
remain in use, and of these 70 are near extinction.35

33 McWhorter, Doing Our Own Things: The Degradation of Language and Music and Why We
Should, Like, Care (New York: Gotham, 2003), 67. McWhorter’s work, like Deborah Tannen’s studies of
language and gender, reXects how anxiety about language change bridges the academic/popular
divide; the appearance of Doing Our Own Things from a popular press was accompanied by Emily
Eakin’s lengthy article entitled ‘Going at the Changes in, Ya Know, English’ in the New York Times (15
November 2003, A15, 17). A similar bridge appears in Tony Crowley’s critique of John Honey’s views of
standards and dialects; while taking a distinctively academic position opposite Honey’s on nearly every
issue, Crowley shares his popular inXammatory style and tone. See ‘Curiouser and Curiouser: Falling
Standards in the Standard English Debate’, in Tony Bex and Richard J. Watts (eds.), Standard English:
The Widening Debate (London: Routledge, 1999), 271–82.
34 Crystal, The Fight for English: How Language Pundits Ate, Shot, and Left (Oxford: Oxford
University Press, 2006), ix.
35 Dixon, The Rise and Fall of Languages, 116–17; Andrew Dalby, Language in Danger: The Loss of
Linguistic Diversity and the Threat to Our Future (New York: Columbia University Press, 2003), 148;
and Stephen May, Language and Minority Rights: Ethnicity, Nationalism and the Politics of Language
(New York: Longman, 2001), 145.
20 Language Anxiety

The reasons such change occasions interest and concern vary from the
intellectual to the cultural to the economic. Robert Dixon, thus, describes as
urgent the ‘task to document . . . before they disappear’ languages whose
speakers are shifting to another language, while Crystal sees language death
as rapidly increasing and the changes involved in language shift as qualita-
tively diVerent from those involved in other kinds of linguistic change.
He speaks of the need to develop ‘in people a sense of the value of a language,
and of what is lost when a language dies . . . there is an urgent need for
memorable ways of talking, to capture what is involved: we need to develop
ear-catching metaphors—language as a ‘‘national treasure’’, perhaps, or as a
‘‘cause for celebration’’, or a ‘‘natural resource’’ ’.36 More subtly, Daniel Nettle
and Suzanne Romaine have connected the retraction of languages not simply
to the emergence of political units but to the economic history of human
society. First agriculture and then the industrial revolution, they suggest,
produced social reorganizations that favored shared language as opposed to
the multilingualism fostered among isolated, undeveloped communities.37
Language shift and death may thus stand as the inevitable byproducts of social
sophistication.
Since language instruction, much less use, is not value-free, the global
expansion of English at the expense of indigenous languages has alarmed
other critics not simply because it will facilitate the demise of these languages,
but because it necessarily brings with the language Anglo-American cultural,
economic, and political values. Merely to speak English, in this analysis, is to
further imperialism (to ‘inhabit’ its discourse), and the subtitle of one
relevant critical essay is the catchy ‘The Threat from Killer Languages’.38 The

36 Dixon, The Rise and Fall of Languages, 117; and Crystal, Language Death (Cambridge: Cambridge
University Press, 2000), 98. For similarly emotive arguments, see Peter Mühlhäusler, Linguistic Ecology:
Language Change and Linguistic Imperialism in the PaciWc Region (London: Routledge, 1996), 269–81;
and Michael E. Krauss, ‘Keynote—Mass Language Extinction and Documentation: The Race Against
Time’, in Osahito Miyaoka et al. (eds.), The Vanishing Languages of the PaciWc Rim (Oxford: Oxford
University Press, 2007), 3–24.
37 Nettle and Romaine, Vanishing Voices: The Extinction of the World’s Languages (Oxford: Oxford
University Press, 2000). An eVective extension of this argument is K. David Harrison’s When
Languages Die: The Extinction of the World’s Languages and the Erosion of Human Knowledge (Oxford:
Oxford University Press, 2007). Demonstrating how issues like animal taxonomy, calendars, and
conceptions of time manifest language-speciWc ways of organizing experience, Harrison argues that
as language diversity atrophies, so, too, does that of world views.
38 See, for example, Robert Phillipson, Linguistic Imperialism (Oxford: Oxford University Press,
1992); Tove Skutnabb-Kangas and Phillipson (eds.), Linguistic Human Rights: Overcoming Linguistic
Discrimination (Berlin: Mouton de Gruyter, 1994); Alastair Pennycook, The Cultural Politics of English
as an International Language (London: Longman, 1994); and Skutnabb-Kangas,‘Linguistic Diversity
and Biodiversity: The Threat from Killer Languages’, in Christian Mair (ed.), The Politics of English as a
World Language: New Horizons in Postcolonial Cultural Studies (Amsterdam: Rodopi, 2003), 31–52.
 Language, Change, and Response 21

fact that non-Anglophones shift to English thereby raises as much alarm—

even hysteria—as the fact of the internal changes that Truss and Honey
describe. And in each case, enormous signiWcance and power are understood
to reside in change and variation, the very phenomena that Lass presents as
neutral.
Changes in the status of English have also raised concerns for English itself,
beginning with the fact that speakers of English as a second language vastly
outnumber native speakers. Indeed, of the one billion speakers currently
estimated to have some command of English—and some estimates are
twice that—perhaps only 400 million use English as their primary, mother
tongue. Native speakers have thereby been transformed into minority stake-
holders in a language that has become increasingly pre-eminent in inter-
national business and communication. Even if other languages are unable to
challenge the status of English as a global language, an increase in the
numbers of their speakers, coupled with the changing character of the popu-
lation of Anglophones and the varieties of language they speak, could have
adverse economic as well as social consequences for the United Kingdom, the
United States, and other traditional Anglophone powers. ‘We may Wnd the
hegemony of English replaced by an oligarchy of languages,’ notes David
Graddol, ‘including Spanish and Chinese. To put it in economic terms, the
size of the global market for the English language may increase in absolute
terms, but its market share will probably fall.’ Here, it is English that is
threatened by the changes of other languages, and the metaphor from eco-
nomics would not seem to be coincidental: the future of English, so this
argument goes, needs to be monitored as a way of monitoring the future of
English markets.39
To an extent, anxiety about language change could be understood as a
manifestation of a more general kind of nostalgia for golden ages and halcyon
days. While the proliferation of grammatically casual electronic communica-
tion on the Internet may have catalyzed this kind of non-historical memory,
then, it certainly did not create what may partly be a feature of the human
condition.40 Like Dylan Thomas looking back on his childhood Christmases

39 Graddol, The Future of English?: A Guide to Forecasting the Popularity of the English Language in
the 21st Century (London: British Council, 1997), 3. As a publication of the British Council, which
describes itself as ‘the United Kingdom’s international organisation for educational opportunities and
cultural relations’, this volume, replete with tables, charts, and statistics, lays claim to a peculiar kind of
hybrid authority, part linguistic, part social, and part economic.
40 Cf. Aitchison, Language Change, 13.
22 Language Anxiety

in Wales, or middle-aged individuals remembering the fashion, music, and

culture of their youth, speakers of all ages can imagine language as once
having attained a standard never reached again. This kind of anxiety is always
powerful, precisely because it rests on memory, which can be measured in
neither quantity nor quality and therefore provides no internal check on the
concerns it might foster.
But the fact that nostalgic complaints are generally unmeasurable and
therefore unveriWable goes only part way towards understanding why changes
in grammar, vocabulary, or usage should occasion the concerns I have
sketched so far. Is the language of Beowulf truly a swamp, back to which our
‘dismally illiterate world’ is slouching? Does the appearance of a word like blog
truly diminish language and thought? Is modern America truly ‘culturally and
even intellectually deprived’ because the general adult population, now nearly
entirely literate, does not cultivate the stylized rhetoric of an age that
restricted literacy by sex, race, and class? And is the shift of language any
more unusual—and therefore any more alarming—than the transformation
of Old English into Middle, or Middle into Modern? When even the persist-
ence of multilingualism causes concern—not because it resists the economic
power of a majority language group or works against national unity, but
because languages mediate power and multilingualism therefore simply has
the potential to be coopted in various kinds of power relations41—one begins
to suspect that anxiety about language change and variation is really about
something else. That something is the subject of this book.

The scope of this book

Language Anxiety explores how anxiety over language change and variation
has transhistorically motivated and underwritten sociopolitical behavior,
ideological formation, and mythological construction—how it has been
largely a constant in the Anglophone world. It suggests that as a constant,
further, this anxiety has served to displace and channel other kinds of social
concerns, whether of economics, race, ethnicity, sex, or class. Put much more
directly, my thesis is that anxiety over language change has euphemistically
displaced anxiety about other issues and that so long as the anxiety remains
centered on language, the other issues can never be fully addressed.

41 Louis-Jean Calvet, Language Wars and Linguistic Politics, trans. Michel Petheram (Oxford:
Oxford University Press, 1998).
 Language, Change, and Response 23

In order to depict and understand this anxiety, my approach to language

change is quite broad. As my preliminary examples have already suggested,
I approach change at a level encompassing the grammatical (such as phono-
logical variation or lexical addition), the pragmatic (such as discourse con-
ventions and standardization), and the social (such as language contact and
bilingualism). And I do this because a word like hopefully as a sentential
adverb, pronunciations like wanna ([wAn@]) for want to [wAnt tu], and the
mere presence of non-native speakers can in fact all elicit moral outrage, calls
for educational reform, and even, in some instances, legislative ministrations.
Much of what I examine has parallels among at least other Western European
languages, in which the myth of Babel has been as foundational as it has been
for English. While a more broadly focused book is thus certainly possible, and
while I will oVer comparative evidence from other language traditions, prac-
tical and theoretical considerations limit my concerns to English: it is the
tradition I know best and the one I want most to understand.42 In my linking
of change with language and the speakers who use it, I also hope both to
conceptualize a range of change and variation within one critical category
and to suggest the mutual eVects speakers and their language have upon
each other.
And though I speak of an English tradition, neither my outlook nor my
method is chronological. In discussing speciWc illustrative examples (whether
government policies, literary expressions, or published volumes), I will indeed
concentrate on the historical context that gives a particular example its
particular force. Alfred’s anxiety about England’s change from multilingual-
ism, for example, is rooted in both the immediate Viking raids and also, more
generally, in discursive traditions about conquest as a visitation by God that
go back, in England, to Gildas’s sixth-century De excidio Britanniae liber
querulus. And in some cases, as in the eighteenth-century codiWcation of
English, particular arguments and publications are predicated on and devel-
oped from earlier published positions, all of which therefore can be connected
to one another in ways useful to the understanding of sociolinguistic think-
ing. But in the main, this book’s orientation is not chronological but topical,
not exhaustive but thematic. The reason for this orientation is simply put:
although it may take various forms at particular moments in history and

42 The diYculty of a comparative approach is that it almost inevitably produces generalities so

broad as to work against focused, detailed conclusions. For an example of just such an approach, see
Peter Burke, Languages and Communities in Early Modern Europe (Cambridge: Cambridge University
Press, 2004).
24 Language Anxiety

although it may be particularly insistent in the current moment, the kind of

anxiety I address does seem to be a historical and social constant, it does seem
to have run through a lot of English sociolinguistic activity irrespective of
historical speciWcs. Language change and variation are by no means the only
social phenomena to mediate such extra-linguistic anxieties, nor is such
mediation the only social work that variation and change accomplish. It
remains the case, however, that connections between change and anxiety
have been particularly prominent in the history of the English language. In
fact, this anxiety has demonstrated greater durability across time and space
than have many grammatical features; it has been a constant for English, even
as words, inXections, and syntactic structures have come and gone. In this
sense, I would argue that anxiety about change might even be regarded as one
of the language’s distinctive characteristics, as one of the features that justify
the single label English for Wfteen hundred years and thousands of miles of
linguistic variation.
The next chapter expands on the theoretical propositions of this one by
surveying the kinds and contexts of synchronic variation and diachronic
change that have aVected English, including grammatical changes and
changes in linguistic repertoire through contact with other languages. There
I want to Xesh out two simple but enormously consequential claims: that
English always changes—typically in regular fashions—and that the social
meanings assigned to changes are distinct from the changes themselves.
Together, the Wrst and second chapters provide the theoretical and practical
premisses for the following three chapters, each of which identiWes a diVerent
nexus of social institutions, principles, and actions and surveys the ways in
which these invest sociolinguistic meanings in variation and change. The
third chapter considers how language change has been represented in myth,
beginning with the Tower of Babel and the myths and miracles associated with
language shift. From there I proceed to the ideological and cultural impact of
languages like Esperanto and Volapük, which have been invented to counter-
act language change, and then to representations of language change and
variation in literary works, where dialect writing and related rhetorical devices
have conceived such linguistic transformation as the object of humor and
social critique and also as a reXection of social and individual deviation.
Chapter 4 examines how Anglophone governments have responded to lin-
guistic change by using language planning to manage contact between English
and other languages. As was the case when English came into contact with
Welsh, American Indian languages, and Maori, such contact has sometimes
arisen through expansion by Anglophones, but it has also resulted from
 Language, Change, and Response 25

immigration by non-Anglophones into predominantly Anglophone coun-

tries, as has been the case with the spread of Asian languages in both Australia
and the United States today. In these circumstances, whether through laws,
military action, or court decisions, Anglophone governments have often used
English to further the construction of national identity. In the Wfth chapter,
I shift my focus to the linguistic reasoning that has underwritten eVorts to
retard or eliminate language change. Here I consider how approaches to both
language and the instruments of its codiWcation (such as dictionaries) have
themselves produced and channeled extra-linguistic concerns. As a conclu-
sion I draw together the evidence of the book to argue for the importance of
distinguishing the fact of language change from the social meanings assigned
to it. Redirection of anxiety about language change and variation to the social
concerns it mediates, I maintain, is the most eVective way to approach (not to
mention resolve) the latter.
All this sounds dispassionate and academic, and these are precisely the
notes I hope to sound in what follows: to dissociate the personal as well as the
political from the processes of language change and variation. At the same
time, as I will emphasize on a number of occasions, I take the recognition that
English is not simply a structured code but a lived experience to be founda-
tional to understanding all these matters. I may here strive to be an objective
analyst, but I am also an academic and a native Anglophone from the
Midwestern region of the United States. Whatever I say about language,
including whatever social meanings I see invested in it, will necessarily be
shaped in some way by my own linguistic background.
As an English teacher, I’ve had plenty of opportunity to witness and
experience the channeling of social issues into language. I’ve heard students
attribute poor grades in science classes to the alleged English deWciencies of
their Asian American, but native Anglophone, teaching assistants, and I’ve
had a non-American acquaintance summon me to a conversation at a con-
ference wine hour in order to have me demonstrate how Midwesterners
talked. I’ve been told that the appearance of Spanish signs in local department
stores is an indication of just what’s wrong with the United States today and,
at a dinner party, that I was an ‘idealist’—and this was meant as an insult—if
I didn’t think that the active suppression of non-standard English was in the
best interests of everyone, especially minorities.
What puzzles, oVends, amuses, or outrages me may not always aVect other
speakers in the same way. And this is to be expected, for it testiWes yet again
to the variability inherent in natural language and speakers’ responses to it.
More importantly, the variable judgements of Anglophones about the social
26 Language Anxiety

meanings of any particular linguistic form or act underscore the experiential

quality of language. And perhaps most importantly, such variability manifests
the fact that what we agree or disagree about is often not linguistic but social
and that if our conversations on such matters are to progress, they sometimes
need to leave language behind.
 2

A Moveable Speech

Where sounds go when they go away

Sometimes, the oddest questions, the ones that would seem to be the most
easily dismissed, end up requiring unexpectedly complex answers. Here’s one:
since the globe spins, could one travel around the earth by Xoating over the
ground in a helicopter and waiting for the world to turn? Here’s another: if
lime and orange are colors as well as fruit, why isn’t grapefruit? An intuitive
response to both questions may well be that they’re simply naive and not
worth answering. But the truth of the matter is that eVective answers end up
being anything but simple. Responses to the Wrst question must draw on
issues like the physics of time, friction, gravity, and relative motion, while
cogent responses to the second depend on semantics, lexical history, and
usage. Without recourse to these admittedly sophisticated heuristics, replies
can take on the peremptory, even mystical quality of an answer like ‘because
that’s the way it is’. And that quality only serves to make the questions appear
even more complex.
Another of these unexpectedly complex questions is, where does pain go
when it goes away? There’s a strong philosophical streak in it, since it
presumes something about the ontology of pain. A physician might say that
pain isn’t so much a ‘thing’ as a response to something, speciWcally to
stimulation of the nerves or muscles or bones, but to anyone suVering it,
particularly children, pain seems real enough. If anything, its reality seems
conWrmed by its disappearance. If a thorn in one’s hand causes pain, and
when the thorn is removed from one’s hand, the pain disappears, too, the pain
would seem to have been extracted along with the thorn. And if pain can be
removed like a thorn, which can be discarded in the trash, what happens to
the pain afterwards? What does one do with it?
The answer to this odd question requires the same kind of sophistication
that I directed at my earlier questions. A physician would simply insist that
one doesn’t really do anything with pain, because as a biochemical response of
28 Language Anxiety

the nervous system, pain ceases when what’s causing it ceases. If I’ve cut my
Wnger, once the cut closes and heals over, the factors eliciting a painful
response from my nervous system cease, too. What does not cease, however,
at least so long as I am alive, is some biochemical response from my nervous
system. I will continue to feel with some of the same nerves that once
registered pain, but with my cut healed what I feel will be, variously, some-
thing rough, soft, wet, or—in the case of minimal tactile stimulation—
eVectively nothing at all. Medically, when pain goes away, much of it in
eVect becomes a new response, no longer painful but perhaps indiVerent
rather than pleasurable.
The implicit comparison I am drawing, of course, is with language change
and variation. When a sound, word, or syntactic structure changes, what
happens to the original linguistic form? More generally, when change involves
an aggregate of features from a language—when Old English becomes Middle
English, say—what becomes of the original form of the language? Do sounds
and varieties ever truly go away, and if so, where do they go? Or, like painful
nervous stimulation that becomes non-painful, do they eVectively become
other sounds and other varieties? In looking at the dominant language of a
particular region like modern Ireland, how does one decide whether that
language is the same as what was spoken three centuries before, or an altered
version of that language that is still somehow the same language, or a diVerent
language entirely? What sociolinguistic consequences follow from the various
possible answers to such questions? Presuming we can diVerentiate an ori-
ginal form or variety from its developments, does the former take any
conceptual, social, or linguistic precedence over the latter? On what grounds?
What I mean to suggest is that questions about what happens to forms and
varieties that change have the uncomfortable impact of an odd question
about pain. On one hand, they might seem simplistic, predicated on miscon-
ceptions about natural language as a category. Granting that Irish and English
are distinct languages that have been used by real people, we can’t (or perhaps
shouldn’t) ask whether one simply became the other or treat the languages as
if they existed independently of their speakers. The English spoken in most
of Ireland today is clearly not the same language as the Irish that predomin-
ated in the eighteenth century: even leaving aside the genealogy of the
languages, their structural diVerences are so great that no mutual communica-
tion between their speakers would be possible. On the other, like a question
about pain, these kinds of questions are rooted in experience, speciWcally
speakers’ experiences with language. In that regard, they are predicated
on a view of language as a social phenomenon, and it is after all in a
social context—whether of speakers addressing one another or confronting
 A Moveable Speech 29

historical documents—that linguistic change and its consequences are particu-

larly salient. Indeed, it is in this kind of context that anxiety about change and
variation arises and has its greatest signiWcance, for (I contend) speakers ulti-
mately worry not about forms or languages as such but about the presumed
social implications of those forms and languages. When language changes, its
forms—phonology, morphology, lexicon, and syntax—reconWgure the lan-
guage as a system, but it is the meaning speakers assign to these forms, their
changes, and the reconWgured system that deWne speakers’ interest and response.
Just as the structural boundaries between dialects and languages can be malle-
able, depending partly on speakers’ attitudes towards their environment, cul-
ture, and fellow speakers, so the outcome of change—in eVect, its teleology—
can rest on speakers’ perceptions of how language mediates their social world.
In asking where sounds and languages go when they go away, I am really
asking something about speakers’ views of themselves and their language.
Change itself, Lass has maintained, is neutral. It is also inevitable, multiform,
and structured for both grammatical alteration and language shift, which is a
community’s (or speaker’s) replacement of its own language with another.
While not absolutely predictable, such change for any one language proceeds
in patterned fashions that replicate similar, non-random patterns in other
languages and other time periods. The social consequences of variation and
change are likewise not random but structured, deriving their meanings from
the way they challenge or support ideas, practices, and institutions. But if
structured change is neutral, the signiWcance speakers attach to it always
reXects individual and collective preference and partiality. Even though
linguistic changes and their meaning share the qualities of regularity and
inevitability, they thus diverge in this crucial way.
This is the issue I want to explore in this chapter, then: this sometimes
shifting distinction between social evaluations of change and change itself.
Changes to the structure of a language, I will argue along with Lass, are simply
neutral linguistic facts: a vowel disappears or acquires a new quality, a
syntactic structure gains complexity, vocabulary expands, and so forth. What-
ever social impulses might contribute to change and whatever institutions or
ideas might assign them signiWcance are entirely diVerent matters, for the
social meanings that facilitate and accrue to change and variation reXect the
partiality of speakers and speech communities. If change is inevitable and
constant, these meanings necessarily vary. They do so, moreover, in ways
that both respond to and further extra-linguistic considerations. I will also
argue that the same kinds of inevitability and regularity that describe struc-
tural change apply to social and regional variation as well as to language
contact. Perhaps the most important concomitant point here is that empirical
30 Language Anxiety

judgements about change and the social recognition of them do not neces-
sarily coincide with one another. At a structural level, for example, it may be
possible to demonstrate that prevailing vowel articulations reXect alterations
in a word’s pronunciation or, more broadly, phonemic change, but such
demonstration has not prevented speakers from judging the new pronunci-
ations as ‘good’, ‘bad’, or ‘wrong’.
It is the disjunction between the fact of linguistic change and the social
meanings that accrue to it that produces much of the anxiety with which I am
concerned in this book. Further and perhaps most importantly, if in theory a
clear distinction exists between change and its social signiWcance, in practice—
in the historical development of English in various speech communities—this
distinction has been both situational and strategic. SpeciWcally, immediate
social concerns have led to the judgement of some linguistic phenomena as
natural—and therefore neutral—change and other phenomena as socially
meaningful in their manifestation or violation of sociolinguistic norms.
Clear, empirical distinctions between synchronic variation, diachronic change,
shift, faulty grammar, and discursive impropriety cannot always be easily
drawn.

The regularities of grammar

In the opening chapter I described change and variation as indisputable facts
of social history. To their inevitability I here add their regularity and, in view
of this regularity, the qualiWcation that in a fundamental way language is
change. Whether at the level of sound, word, syntax, or linguistic repertoire,
change and variation are not random processes or the simple byproducts of
laziness and ineptitude but rather develop in structured ways that allow for
empirical analysis and explanation and that Wgure in various languages of
various epochs. They are, in fact, linguistic universals. If this simple claim
might seem unremarkable to most linguists, the anxiety associated with
linguistic change and variation suggests that it is not more generally under-
stood or accepted.
The regularity of which I speak operates at several levels of linguistic
structure. Nouns, for example, are everywhere the most heavily borrowed
category, and function words, like determiners and prepositions, the least
borrowed. When languages come into contact with one another, similarly, the
survival of any particular language generally depends more on the kinds of
domains in which it is used than on merely the number of its speakers. And
phonological variation is so consistent and inevitable as to play an essential
role in the diVerentiation of meaning. So-called strong verbs like sing and
 A Moveable Speech 31

drink, for instance, form their preterites by altering the root vowel to produce
sang and drank. And they do so because in many Indo-European languages,
including the Germanic family, the universal process of context-conditioned
vowel gradation (Ablaut) came to carry the functional load of verb tense. In
these verbs, thus, the distinction between present and preterite—a distinction
that probably all Anglophones would accept as not only normal but crucial to
the production of meaning—rests, ultimately, on the fact that early speakers
were inconsistent in their articulations of the same vowel.
The collection of changes aVecting tense vowels in English and collectively
designated the Great Vowel Shift oVers an elaborate example of just this point:
how in its non-random regularity, change serves as a vital deWning feature of
English (and, for that matter, all natural languages). Occurring at various
times in various locales—and echoed by parallel changes in German and the
Scandinavian languages—these changes generally originated in the early
Wfteenth century and were completed within two centuries. As a reconWgura-
tion of English phonology, the Great Vowel Shift represents a transformation
of the language’s grammar—of what it means to speak English. What had
been /a/ became /e/, for example, and what had been /i/ diphthongized
(eventually) as /ai/, so that in an oral recitation of Chaucer’s ‘Wife of Bath’s
Tale’ the Knight would address his lady and his wife as [ladi] and [wif], while
in one of John Dryden’s translation she’d be his [ledi] and his [waif].1 As
I noted in Chapter 1, it’s as if the Wrst syllable of fodder were to be pronounced
like fate, whose vowel in turn came to be like the one in feet, whose own vowel
became like the one in Wght. Far from being haphazard, changes like the Great
Vowel Shift manifest such regularity, with the same sound under the same
circumstances tending to alter in the same way, that the nineteenth-century
neogrammarians regarded sound change as admitting no exceptions, though
recognition of synchronic variation and the messiness of the historical record
would relax this claim for most linguists since then. ‘All grammars leak’,
observed Edward Sapir,2 and the Great Vowel Shift had at least one leak: in
northern England and Scotland, the high back vowel /o/ seems to have moved
forward as well as up to (eventually) /ü/ rather than directly up to /u/, with
the result that in those regions the original /u/ was not pushed to diphthong-
ize as /au/, as it did in other early modern dialects of English. Modern English

1 For the sake of clarity, my transcriptions here are intentionally broad. Dryden, for example, could
well have realized the stressed vowel of lady as [e:] or even as a diphthong like [ei], while its Wnal vowel
would likely have had an oV-glide for Chaucer if not Dryden. Since the Great Vowel Shift aVected only
stressed long vowels, it would have had no impact on the Wnal syllable of lady, thus leaving it the same
for Chaucer, Dryden, and, indeed, most modern speakers.
2 Sapir, Language: An Introduction to the Study of Speech (New York: Harcourt, Brace, 1921), 3.
32 Language Anxiety

house, pronounced [hus] (rhymes with loose) in Old and Middle English, thus
became [haus] via the Great Vowel Shift but retained the former pronunci-
ation in most Scots varieties. This northern leak was in fact so steady that
Jeremy Smith has argued it represents a distinct co-varying shift.3 This may
well be the case, but for my purposes, the same conclusion emerges: both
shifts (and any leaks) were so consistent as to support a claim for the
regularity of sound change.
Further support emerges from the strikingly parallel set of Primitive Ger-
manic phonological phenomena collectively known as the First Consonant
Shift. Like the Great Vowel Shift, the First Consonant Shift describes a series
of changes (these involving voiced and voiceless stops) that occurred over
several centuries and that, in the way they relate to one another, constitute
evidence for the found fact of a co-varying shift. To oVer just one example, the
voiced Indo-European aspirant /bh / (a breathier version of the b in bull)
became the continuant /ß/ and eventually the unaspirated stop /b/ (the b of
stab); the original voiced stops, in turn, devoiced, so that what had been /b/
became /p/; and an original voiceless stop like /p/ became a fricative like /f/.
Also like the Great Vowel Shift, the First Consonant Shift leaked a bit. The new
voiceless fricatives became voiced in certain circumstances, for example, while
in much of what is now Germany this co-varying shift was extended in a series
of changes known as the Second Consonant Shift. One measure of their
regularity is the fact that the shifts I have described are among the most
established and crucial in conceptualizing the Germanic languages as a group.
It is the First Consonant Shift, ultimately, that accounts for some of the most
pervasive etymological doublets in English: dentists who work on teeth,
podiatrists who work on feet, brothers of a fraternal order, paternalistic fathers,
wheeling cycles, and all kinds of gender. The leaking of the voiceless fricatives
explains why the singular preterite of the verb be is was but the plural were,
and the leaking of the extended co-varying shift explains why Germans drink
Wasser in place of the water drunk by Anglophones.4 But the Shift also
signiWcantly helped deWne the methods and objectives of comparative lin-
guistics in general. It was the early nineteenth-century linguist and folklorist
Jacob Grimm, indeed, whose ‘law’ provided the earliest comprehensive ac-
count of these changes and served as a model for subsequent attempts to
deWne putative co-varying shifts.

3 Smith, Sound Change and the History of English (Oxford: Oxford University Press, 2007), 127–53.
4 Accounts of the First Consonant Shift and the leaks described by Verner’s Law (was) and the
Second Consonant Shift (water) can be found in most comparative or historical Germanic grammars.
 A Moveable Speech 33

In calling the Great Vowel Shift and the First Consonant Shift regular, I do
not mean to suggest that such change is necessarily, or perhaps even charac-
teristically, rational. For one thing, as the school of ‘emergent linguistics’ has
increasingly suggested, such changes depend not simply on pre-existing
grammatical structures but on the individual articulations of speakers.
Indeed, the basic idea of emergent structure, according to two of its adherents,
is that ‘what may appear to be a coherent structure created according to some
underlying design may in fact be the result of multiple applications or
interactions of simple mechanisms that operate according to local principles
and create a seemingly well-planned structure as a consequence’.5 And an-
other factor that works against strict rationality for linguistic change is that in
both examples of co-varying shift, however the changes began and proceeded,
they eventually produced a phonology strikingly similar to the one they had
replaced. Although two new diphthongs emerged in the process of the Great
Vowel Shift, excepting lengthened /E:/ English began and ended the changes
with essentially the same inventory of long tense vowels: /i/, /e/, /a/, /O/, /o/,
and /u/.6 And as for the First Consonant Shift, Primitive Germanic had the
same inventory of voiced and voiceless stops as had Indo-European; where
it did diVer from Indo-European was in having phonemic voiceless fricatives
(/f/, /‚/, /x/) rather than phonemic voiced aspirated stops (/bh /, /dh /, /gh /),
but even in this case, there was phonetic overlap in aspiration, articulation,
and, in words described by Verner’s Law (e.g., was / were), voicing.7 In very
casual terms it’s as if in each case, after centuries of phonological reconWgura-
tion extensively aVecting the language’s lexicon, sounds mostly traded places,
both going away and remaining at the same time—which would not seem to
reXect any preconceived, rational structure.

5 Joan Bybee and Paul Hopper, ‘Introduction’, in Bybee and Hopper (eds.), Frequency and the
Emergence of Linguistic Structure (Amsterdam: John Benjamins, 2001), 10. See also Bybee’s Phonology
and Language Use (Cambridge: Cambridge University Press, 2001); and Frequency of Use and the
Organization of Language (Oxford: Oxford University Press, 2007).
6 For the sake of argument, I simplify here by omitting the varying glides described by many
phonologists for the realization of the vowels in many varieties of Modern English.
7 Phonological balance is sometimes claimed as the motivation of and improvement provided by
the First Consonant Shift. That is, after the changes Germanic, unlike Indo-European, had pairings of
not only voiced and voiceless stops but also voiced and voiceless fricatives. As a mechanistic
explanation, however, phonological balance is limited in the same way as functional optimality,
which I discussed in Chapter 1. Beyond that, even granting this explanation this fundamental fact
remains: after the Shift, Germanic had virtually all of the sounds it had before it. A perhaps more
striking example of this kind of non-rational change involves the vowel in modern English that. In
Primitive Germanic, this vowel was /a/; in Old English it became, variously, /æ/ or /E/; in Middle
English it became /a/ or /E/; and in Early Modern English it generally became /æ/ once more; then, in
certain phonological environments in varieties of British English, it returned yet again to [a] or [A].
34 Language Anxiety

For the Great Vowel Shift, accordingly, an answer to an apparently simple

question like ‘Where did /i/ go?’ is anything but simple. Did it, ontologically,
become /ai/, so that this very same vowel now somehow appears with a
diVerent sound in words like wife? Is a vowel, in other words, much like
individuals who might be understood to retain the same identity, even when
they put on new hats and coats? Or did the sound remain the same, irre-
spective of the words in which it occurred, so that it in eVect became what had
been /e/ in words like feet, previously pronounced [fet] but now pronounced
[fit]? In this case, is a vowel like a furnished room that stays unchanged,
whoever occupies it? Or did it in some fundamental way simply disappear,
replaced by an entirely new /i/ and generating the new diphthong /ai/?
These are not idle questions. Judgements that Middle English /i/ becomes
/ai/ in words like wife accept the fact that language is a system incorporating
change and transcending its speakers and forms at any one time. In this
system, change and variation, at least in certain phonological circumstances,
occur as natural processes that do not disturb the essential character of a
language. By such analyses, English is still English, even if its vowels trade
places and we now designate it Early Modern rather than Middle. And this
kind of analysis identiWes certain kinds of change as systematic not only in the
way they unfold but also in the fact that their occurrence is as grammatical as
speciWc rules of syntax or morphology. It is a syntactic rule, for example, that
articles placed before their nouns do not violate but rather adhere to the
grammar of English, just as it is an apparent systematic rule that certain kinds
of change and variation, as in co-varying shifts like the Great Vowel Shift, can
occur without violating the integrity of the language.
The interesting thing about rules like these is that while their positive forms
parallel one another, their negations do not. A violation of an English
syntactic rule would take the form of a structure like dragon the, in which
the article postposes its noun. Such violations are empirically veriWable,
practically inWnite (there are many kinds of incorrect structures, like the
dragon the, the the dragon, and so forth), in many cases relatively stable over
time (dragon the has always been ungrammatical in English), and recogniz-
able as unacceptable by all speakers. I know of neither a grammar handbook
nor a variety of English that would recognize dragon the as well formed.8
Violations of rules about change and variation are not so simply illustrated,
however, because these kinds of violations reXect not empirical facts but

8 Within the Germanic language group, however, Old Norse did allow for enclitic articles on nouns
not preceded by an adjective: drekinn, literally ‘dragon the’, but inn grænn dreki, ‘the green dragon’.
Modern Scandinavian languages show reXexes of this structure.
 A Moveable Speech 35

judgement calls over which linguistic phenomena exceed permissible change

and variation for a language and thus transgress the language’s integrity. It is
by this kind of situational evaluation that the pronunciation of lady can
change from [ladi] to [ledi] and still qualify as grammatical English, while
want to becoming realized not as [want tu] but [wan@] (wanna) can illustrate
non-standard or incorrect speech.
One might well ask on what grounds a distinction like this has been drawn.
And one way to approach this important and complex issue is through the
common distinction between prescriptive and descriptive grammar, which
underwrites all manner of government and educational as well as purely
linguistic activity. Prescriptive grammar, generally, refers to what the word
grammar invokes in the minds of most educated people: rote drills about i
before e in spelling or never ending a sentence with a preposition that often
seem to run contrary to what people naturally wish to do with English and
that were learned partially and sometimes painfully in schoolrooms and
through exercises. Locatable in dictionaries as well as grammar books, this
kind of grammar has standards of absolute right and wrong that provide, in
the form of explicit rules, a way to measure grammatical competence.
While prescriptive grammar might be thought akin to the driving rules of
the road, whose violation results in accidents and citations, descriptive gram-
mar is like the rules that fashion a car from metal, rubber, plastic, and so forth;
driving a car the wrong way down a street, while proscribed as illegal, is
nonetheless physically possible, but an object is not a car if it lacks wheels.
Descriptive grammar, then, provides an account of how well-formed—i.e.,
natural and acceptable—utterances are made in a language. Its grammatical
rules describe what it means to speak English: what sounds are used, how
words are made, and the ways in which words combine in sentences. In this
sense, learning descriptive rules means learning the language, whether pas-
sively as an infant or actively in an adult second-language course. And rather
than something like ‘don’t split an inWnitive’, a descriptive rule would be of
the sort I noted above: ‘adjectives are positioned between determiners and
nouns’, so that the great green dragon is a well-formed English noun phrase,
but great green the dragon is not. Given the variability of how people speak, of
course, descriptive grammar allows a good deal of latitude in identifying
what’s well formed and resists passing evaluative judgements on the variation
it identiWes. Where prescriptive grammar would enjoin speakers not to use
‘double negatives’, descriptive grammar would note that in some modern
varieties, such as African American Vernacular English, multiple negation has
the same intensifying eVect that it did for Chaucer and Shakespeare.
36 Language Anxiety

Or so it goes in theory, at least. As several critics have noted, as crucial as

the categories of prescriptive and descriptive grammar may be, an imperme-
able distinction between the two is not easy to maintain.9 The latter, which
characteristically draws its data from the critics’ experiences, ultimately makes
many of the same kinds of value judgements about acceptable and unaccept-
able language as the former. Descriptive grammar may label correct structures
well formed instead of right, and it may refrain from attaching any moral or
intellectual signiWcance to them, but like prescriptive grammar it remains in
the business of distinguishing English from non-English. In the case of
something like the voiceless bilabial stop [p] as opposed to a click from a
language like Xhosa, it is essentially accurate to say that a simple description
of English would include the one but not the other. But many grammatical
rules are less clear. There doesn’t seem to be a strict rule that precludes the
phrase the green great dragon, for instance, but to my ear, there’s nonetheless
something odd about its ordering of adjectives. And if the reason it sounds
odd is that in English we prefer adjectives of size to precede those of color, this
remains a feature of style and not one of the structure of the language: the
phrase’s ordering of determiner, adjectives, and noun is well formed.10 Even a
descriptive rule as apparently innocuous as ‘in most cases form English
preterites by adding the suYx -ed’ (variously articulated as [d], [t], [
d], or
[
t]) works against an easy distinction between descriptive and prescriptive
rules. Above all, it fails to account both for varieties that include the deletion
of Wnal stops and for ones regarded as creoles or pidgins, in which preterite
formation typically follows an entirely diVerent, periphrastic paradigm by
forming verbs with the root and done or been (e.g., he done walk rather than he
walked).
What grammatical descriptions can easily overlook is that simply by virtue
of the identiWcation of a base form from which other forms develop—or
against which they are measured—the description of certain rules necessarily
becomes the prescription of a particular variety of English. Many analyses of
preterite verbs, for example, describe the deletion of Wnal stops as a rule that
sequentially follows the one adding the suYx -ed, and this kind of rule has the
eVect of rendering varieties that employ deletion as secondary and derivative.
It’s certainly true that grammatical precedence might well be based on history
and thereby in this example justify a claim for the voiced stop preterites as
being the base form from which other preterites develop. Yet this same claim

9 e.g., Roy Harris, The Language Machine (Ithaca, NY: Cornell University Press, 1987).
10 If there is a stylistic ‘rule’ here, it’s not easy to state. Green happy dragon, with an adjective of color
preceding one of disposition, seems unexceptionable to me; and even green small dragon doesn’t sound
as odd to me as green great dragon.
 A Moveable Speech 37

simpliWes the diversity of the historical record, for competition between these
two preterite forms dates at least to the eighteenth century and is therefore
scarcely ephemeral. More generally, similar kinds of competition characterize
the history of English; throughout the fourteenth and Wfteenth centuries, for
example, alternate forms of the singular third person indicative verbal inXec-
tion ([E‚] and [Es], spelled eth and es) were in use, and in certain isolated
rural districts that competition continued until fairly recently. The most
important point for my purposes, however, may be the fact that arguments
that draw on historical precedence tacitly bypass the role social evaluations
play in identifying variation as acceptable or erroneous. The existence of both
dental preterites and periphrastic ones is a simple linguistic fact. The de-
cisions to derive the second from the Wrst and then associate it with a
particular group of speakers and even pronounce it incorrect—these are all,
to varying degrees, value judgements.
While Wctional, Mark Twain’s Adventures of Huckleberry Finn makes these
points about both the malleability but also the social implications of grammar
with particular eVect. Even before the book begins Twain voices the descrip-
tive dispassion of a linguist in claiming that he has utilized seven dialects in
the book. He further dispassionately maintains that he has done so not
haphazardly but ‘pains-takingly, and with the trustworthy guidance and
support of personal familiarity with these several forms of speech’.11 He
thereby grants these kinds of variation status as markers of legitimate social
and regional varieties of English; his characters are not, as he observes, all
trying to speak the same way, for the simple reason that not all Anglophones
speak the same way. Yet the novel’s moral and rhetorical force relies precisely
on the disjunction between Huck’s ignorance of prescriptive grammar, due to
his lack of formal education, and the rectitude of his conscience. His language
marks him as an ignorant social outcast, while his heart declares his con-
science, and language and heart together thereby undermine neat truisms
about prescriptive grammar, class, and virtue.
As tenuous as this distinction between descriptive and prescriptive gram-
mar may sometimes be, it does more for English than simply underwrite
many educational and government actions—though this is signiWcant in and
of itself. The distinction also supports a broader categorical distinction
between change and variation that are judged internal to a language, and
change and variation that violate its essential character and to which social
meanings can accrue. Here, too, the distinction can seem permeable, even

11 Twain, Adventures of Huckleberry Finn, ed. Henry Nash Smith (Boston, MA: Houghton
MiZin, 1958), 2.
38 Language Anxiety

impressionistic. Trivial phonological variation such as the pronunciations of

[in] as the termination of words ending in -ing (often represented as in’) can
elicit indignation as corruptions of English, even as extensive phonological
changes such as the First Consonant Shift have been readily accommodated as
internal changes that leave the integrity of the language and its speakers intact.
By way of an explanation for the Shift, for example, Grimm himself dispas-
sionately regarded the consonant changes, somehow, as reXections of Ger-
manic pride and impetuosity: the same life force that gave rise to Germanic
culture gave rise to the Germanic languages. In a similar collapse of language
and culture, Grimm’s contemporary Heinrich Meyer-Benfey thought that the
changes may have arisen from the eVects of mountain air on Germanic
speakers. In each case, it’s as if Germanic predates the co-varying shift that
deWnes it, for change was presumed to be inherent in the language as the
expression of circumstances and social identity; beyond speakers’ control, it
was therefore also beyond the purview of value judgements. Recently, Smith
has oVered a far more empirically sophisticated explanation for the Shift by
tracing it to social and linguistic contact between early Germanic speakers and
the Raeti to their south.12 While this account transcends Grimm’s and Meyer-
Benfey’s in every way, it shares with them the sense that some changes, even
ones that aVect many of a language’s consonantal phonemes, can occur
without undermining the integrity of that language. All three analyses seem
fundamentally essentialist: German is still German, whatever changes it
undergoes, so long as those who speak German speak it.
Comments on the Great Vowel Shift reXect the impressionism of much
linguistic categorization with just as great a force, precisely because, unlike
any modern discussions of the First Consonant Shift, they sometimes do
involve not only Grimm’s dispassionate acceptance of change as part of a
language’s grammar but also the indignant response to unnatural change that
Honey, McWhorter, and other modern critics express. On one hand, for
example, Lass describes the essence of the Great Vowel Shift as a systemic
change much like that of the First Consonant Shift: ‘each non-high long vowel
raises one height, and the high vowels diphthongise, dropping their Wrst
element by one height’.13 John Hart, who witnessed (or perhaps heard) the
Shift as it occurred and who in his Orthographie provided one of the richest
sources for phonological discussion, can share Lass’s dispassion. On occasion
he, too, makes simple and scientiWc observations about, for instance, the fact

12 Eduard Prokosch, A Comparative Germanic Grammar (Philadelphia, PA: Linguistics Society of

America, 1939), 55; Smith, Sound Change and the History of English, 84–7.
13 Lass, ‘Phonology and Morphology’, in Lass (ed.), The Cambridge History of the English Language,
iii, 1476–1776 (Cambridge: Cambridge University Press, 1999), 73.
 A Moveable Speech 39

that the English and the Scots ‘doe at any time sounde i, in the foresayde
sounde of ei’. Yet on the other hand, Hart also evokes a present-day value
judgement on the changes when he sees them as detrimental to communica-
tion. While he resists connecting the changes to speciWc social groups or
assigning them speciWc social valuations, he does consider them illustrative of
the kinds of deviations (from putatively correct pronunciations of graphs)
that characterize what he sees as the period’s general abuse of language.14
As impressionistic as distinctions between natural, grammatical change
and its converse may be, they nonetheless Wgure signiWcantly in the kinds of
meanings that can accrue to a particular change. When changes are judged
natural, discussion typically focuses on their mechanics—on how they change
the grammar of a language—so that their meanings are purely linguistic and
not social. For Western European languages, for instance, the First Consonant
Shift Wgures not as a corruption of Indo-European but as a catalyst in the
creation of the Germanic languages, which themselves then represent a
separate group of languages with its own linguistic integrity. And, Hart
notwithstanding, the Great Vowel Shift likewise typically represents not
deterioration of medieval English pronunciation but an aspect of a categorical
grammatical transformation that produced Modern from Middle English;
despite the extensiveness of the change, we use the same noun, ‘English’, to
refer to the language spoken before and after its occurrence.
More narrowly, incidental grammatical change can be accepted without
comment, when categorized as an ordinary internal development, or as a
linguistic and social mistake, when otherwise categorized. Thus, a Primitive
Germanic form like *horso remained hors in Old English but through the
process of metathesis became hross in Old Norse and hros in Old High
German and Old Saxon; Old English brid became Modern English bird in
the same way. In all cases metathesis occurred without popular or scholarly
disapproval.15 Yet in the realization of professor as perfessor ([p@rfEs@r]), a
common form in popular representations of colloquial speech, r-metathesis
very clearly correlates with ignorance. It is by this same phonetic process that
Modern English ask is sometimes realized as [æks], with the two Wnal sounds
reversed, and here again the metathesis has frequently been identiWed as a
problem—and served as the subject of jokes—even though, as has often been
noted, the metathesized pronunciation has a long history. Though the Old
English form of the verb was in fact ascian ([AskiAn]), the spelling ks occurs

14 Hart, An Orthographie (1569; rpt. Menston: Scolar Press, 1969), 32.

15 A similar example may be Old English yrnan and Modern English run, though the derivation of
the former is complicated; see OED, s. v., run, v. The asterisk in front of horso means that the form is
unattested—necessarily so, since Primitive Germanic was not a written language.
40 Language Anxiety

already in the Anglo-Saxon period and persists afterwards in the United States
as well as in Britain. In his 1789 Dissertation on the English Language, Noah
Webster in fact remarks not only that ‘the word ax for ask was used in England’
but also that ‘ax is still frequent in New England’.16 The signiWcant point here
is that it is not simply a phonetic change or a sound that is inherently marked
for Anglophones. As Labov comments, indeed:
The force of social evaluation, positive or negative, is generally brought to bear only
on superWcial aspects of language: lexicon and phonetics. However, social aVect is not
in fact assigned to the very surface level: it is not the sounds of language which receive
stigma or prestige, but rather the use of a particular allophone for a given phoneme.
Thus the sound [i:@] is not stigmatized in general, since it is the prestige norm in idea,
but it is stigmatized as an allophone of /æ/ in man.17

A process like metathesis or a diphthong like [i:@] becomes stigmatized as

external corruption rather than internal change only when speakers, for
whatever non-linguistic reasons, assign negative value to these linguistic
phenomena in certain circumstances.
One way that changes accepted as internal and natural diVer from those
that are not is that descriptions of the former often take the form of plain
tables of cognates like this:
name: Name (G), nomen (L), Z
Æ (Gr), ainm (OIr), nafn (ON)
two: zwei (G), duo (L), ı (Gr), tveir (ON), dwa (Pol)

What is characteristically absent in discussion of changes perceived as gram-

matical and represented in this fashion is any consideration of the demon-
strable and necessary social processes like migration and colonization that
would have produced, in this case, the cognates or their consequences. It is as
if, as Lass observes of linguistic facts like co-varying shifts, the establishment
of a category of natural grammatical change enables certain kinds of questions
but precludes others, in this case ones about how cognates come into exist-
ence and what they imply sociolinguistically. Perhaps the most signiWcant of
these questions is about the legitimacy of a variety as a language.
A table of cognates simply accepts as given the fact that social processes like
immigration and colonialization have resulted in new languages. No one, to
the best of my knowledge, has lamented—has attached any social meaning
to—the shifting of stops and fricatives that helped deWne the Germanic
language group, or the creation of new diphthongs by means of the Great

16 Webster, Dissertations on the English Language (1789; rpt. Menston: Scolar Press, 1967), 386. See
further OED, s. v. ask, v., and DARE, s. v. ask, v.
17 Labov, Principles of Linguistic Change, ii, Social Factors (Oxford: Blackwell, 2001), 28.
 A Moveable Speech 41

Vowel Shift. By comparison, when immigration and colonization have Wgured

in the development of creoles or the diVusion of global English, discussion
has indeed focused on issues like the competence of speakers or whether
particular varieties can even be labeled English. And so even though similar
processes of sociopolitical expansion and regular grammatical change under-
wrote both the emergence of the Germanic languages via the First Consonant
Shift and the development of new, transplanted varieties of English in previ-
ously non-Anglophone countries, the former has been categorized as natural
change while the latter, lacking this categorization, has produced anxiety and
even outrage. Lass’s scheme of ‘each non-high long vowel’ raising one height,
similarly, depicts the Great Vowel Shift as a purely internal linguistic phe-
nomenon and as such one that precludes value judgements of the sort
modern critics can make of the variant pronunciations they hear. Absent
from Lass’s scheme and from the category of natural grammatical change
more generally is any sense that Middle English phonology is a swamp from
which Modern English phonology climbed. Conversely, neither does his
scheme imply that Middle English phonology was the dry land from which
modern pronunciation retreated—that vowel raising was a diminishment of
language and thought or diphthongization a source of the modern world’s
intellectual and cultural deprivation.
The primary distinction I have been drawing so far is between structural
change and social meaning. While the former is regular, constant, and,
perhaps, irrational, the latter is often impressionistic and even opportunistic,
depending a good deal less on the nature of the linguistic phenomena and a
good deal more on the context of the evaluation. The entire tense vowel
system of English might change—has changed—without violating the integ-
rity of the language, while the deletion of Wnal stops or the realization of Wnal
[˛] as [n] (ing becoming in’) contributes to utterances that are judged no
longer English. Or, as signiWcantly, such utterances might be classiWed as bad
English in one sense or another and thereby justify an evaluation of an
individual’s worth and ability. As Henry Higgins says of Liza Doolittle’s
pronunciation in Shaw’s Pygmalion, ‘A woman who utters such depressing
and disgusting sounds has no right to be anywhere—no right to live. Re-
member that you are a human being with a soul and the divine gift of
articulate speech: that your native language is the language of Shakespear
and Milton and The Bible; and don’t sit there crooning like a bilious pigeon.’
What gives Higgins’s observation its linguistic force is its recognition that
Liza’s phonology varies so considerably from his own. What gives it its
social force, of course, is not simply the categorization of this variation as
outside the kinds of variation permitted within English grammar but also the
42 Language Anxiety

attribution of social meaning to it within a nexus of socially described

varieties of English: ‘You see this creature with her kerbstone English: the
English that will keep her in the gutter to the end of her days. Well, sir, in three
months I could pass that girl oV as a duchess at an ambassador’s garden party.
I could even get her a place as lady’s maid or shop assistant, which requires
better English.’18
Prescriptive and descriptive grammar, I have suggested, overlap in their
reliance on categories of acceptable and unacceptable forms. I have joined to
that claim an even more far-reaching one. Built on this shifting and perme-
able foundation, judgements of a change as natural and inevitable or aberrant
and regrettable can themselves largely be functions of extra-linguistic atti-
tudes. In eVect, both sets of distinctions are situational and adaptive, with the
same linguistic phenomena potentially pointing to diVerent categorizations,
interpretations, and responses in ways and for reasons that have nothing to do
speciWcally or necessarily with language. Truss’s advocacy of zero tolerance for
errors in punctuation, for example, rests on a sense of punctuation as intrinsic
to the grammatical expression of meaning—as something constitutive of the
descriptive grammar of English. But she also represents punctuation as
subject to socially constructed standards of absolute correctness. Her stylized
outrage evokes the conXict and controversy of putatively unnatural gram-
matical dissolution, even though its instigation simply illustrates the variation
and change that have historically characterized English. Punctuation is in fact
a particularly fraught topic in grammar. In English, it is sporadic in the Old
and Middle English periods, typically directed towards oral recitation and
rarely transmitted in the copying of a work. And yet as medieval manuscripts
amply illustrate, the absence of punctuation was in no way a hindrance to the
production of meaning. While punctuation practices expanded with the
introduction of print, into the nineteenth century they remained primarily
rhetorical (rather than syntactic) and largely the results of publishing houses’
styles. It is only in the modern period that writers like Hemingway began
consistently to exercise authorial concern over punctuation in ways that
justify Truss’s sense of its centrality to a work’s meaning, a sense that other
modern writers, like Thomas Wolfe, who relied on the extensive editorial
ministrations of the editor Maxwell Perkins, completely undermine.
This kind of interpretation has an opportunistic quality, though the op-
portunities are all the interpreters’ and not the language’s. Depending on
contextual factors, various critics may variously judge punctuation to be
largely an irrelevancy, a quality of descriptive grammar, or a prescriptive

18 Shaw, Complete Plays with Prefaces (New York: Dodd, Mead, 1962), i, 206.
 A Moveable Speech 43

feature that can be left to others as a means to clarify, but not create, meaning.
I don’t see anything nefarious in such variable categorization. It is an example,
rather, of how change and variation acquire social meaning. But for this
very reason, I also don’t see the impact of grammatical classiWcation, however
shifting it may be, to be of any less signiWcance. And for this reason,
beyond any social consequences that accrue to variations judged violations
of a language’s prescriptive grammar, the decision of whether or not to treat
particular linguistic phenomena as internal variation and change goes a
long way towards deWning the contours of a language along with the groups
of individuals who do—and do not—speak it.
The history of gender in English grammar illustrates this point especially
well. As an Indo-European and Germanic language, Old English inherited a
system of anaphoric pronoun usage and inXectional concord that correlated
words for he (hé), she (héo), and it (hit) with particular declensions of nouns
and, by extension, particular declensions of adjectives as well. In the strictest
terms, the presence of such grammatical gender means that a noun’s morph-
ology rather than its semantic content determines which pronouns and
adjectival forms concur with it. An Old English bóc (book), thus, is a héo, a
stán (stone) is a hé, and a lið (‘Xeet’) is a hit. While initially distinctions like
these were largely the matter of systemic grammar—speciWcally morph-
ology—already in the Anglo-Saxon period modiWcations took place that
suggest speakers’ intentional interventions and reclassiWcations of gender as
a grammatical feature. This was so, evidently, because for certain animate
nouns, grammatical and natural gender conXicted:
Word Sense Grammatical gender
wı́f adult female; married female neuter
wı́fmann adult female masculine
mægden young unmarried female neuter

For words like these, the pronoun héo, matching natural rather than gram-
matical gender, became typical, although in non-animate nouns grammatical
gender continued to predominate into the twelfth century. Patterns of usage
complicate grammatical analysis in still other ways suggestive of Anglo-Saxon
speakers’ inXuence not simply on pronoun usage but on grammatical clas-
siWcation and the language itself. As the distance in an Old English sentence
between a pronoun and its antecedent noun increases, for example, so, too,
does the likelihood that the pronoun will follow natural and not grammatical
gender. Grammatical categorization, in other words, was determined by the
social impulses of discursive conventions. The general shift to natural gender
is likewise social in the geography of its history, for like later Middle English
44 Language Anxiety

changes in the third person plural pronouns and inXection of third person
singular indicative verbs, it seems to have begun in the north and east of
England and gradually spread southwards and westwards.
All these changes commenced in a period of extensive contact with Old
Norse, which, in Anne Curzan’s view, may not have instigated the shift but did
help it along: ‘Whether or not confusion over inXectional endings or gender
between Old English and Old Norse speakers was a cause of the grammatical
change in English is a matter of speculation; but the evidence supports
theories that language contact can and will speed changes incipient in one
language.’19 In these ways, speakers’ attitudes, conversational strategies, and
social interactions—all of them strategies for prescribing usage—inXuenced
and even reshaped English’s descriptive grammar. This inXuence appears
in yet another way. When the decisive shift away from grammatical gender
took place (roughly 1150 to 1250), it initially aVected masculine nouns more
than feminine ones, with the result that feminine pronouns co-occurred with
inanimate nouns longer into the history of English than did masculine ones.20
To this day a version of grammatical gender persists with she, in references to
the moon, days, cars, or ships, even if such usage strikes many modern
Anglophones (including me) as odd or obsolete.
The historical variation of grammatical gender in Old English suggests,
then, not absolute, impermeable categories of grammar and usage but Xuid
ones subject to redeWnition by (in this case) Anglo-Saxon speakers. For largely
social reasons, inherited grammatical gender remained strong with inanimate
nouns, but among animate ones referring to humans, natural gender came to
have an established and increasing signiWcance. And it did so apparently with
neither social conXict nor confusion, despite the extensive and non-trivial
reconWguration of English grammar that it involved: its concomitant was a
complete reconWguration of the morpho-syntax of the English noun phrase.
And it did so, too, despite the fact that it was implicated in, if not driven by,
social processes, which have in many cases focused all manner of anxiety on
change. ‘The diVusion of natural gender’, Curzan observes, ‘seems to follow

19 Curzan, Gender Shifts in the History of English (Cambridge: Cambridge University Press, 2003),
123. It is worth noting that sign language has been susceptible to the same kinds of management that
other natural languages have experienced. There have been increasingly successful attempts, for
example, to modify signs that might have racist or sexist overtones, as for Chinese or homosexual.
Such management has not been without controversy, raising the same charges of ‘political correctness’
that Curzan addresses. See Jennifer Senior, ‘Sign Language ReXects Changing Sensibilities’, New York
Times, 3 January 1994, A1, 8.
20 Curzan, Gender Shifts in the History of English, 110. On the statistical probabilities of grammatical
gender with animate Old English noun, see 62 V.; and on the inXuence of a pronoun’s distance from
an antecedent nouns, see 99.
 A Moveable Speech 45

patterns of historical syntactic change such as reanalysis and extension with-

out a cataclysmic disruption in this diVusion.’21
And yet by the sixteenth and seventeenth centuries, grammatical and
natural gender had again been reanalyzed in ways that pointed to just such
cataclysms. Handbooks like Thomas Wilson’s 1553 Arte of Rhetorique had then
begun to see pronominal usage as not only a grammatical issue involving
what was and wasn’t correct or even real English but also a reXection of a
larger natural social order: ‘Some will set the carte before the horse, as thus.
My mother and my father are both at home, euen as thoughe the good man of
the house ware no breaches, or that the graye Mare were the better Horse. And
what thoughe it often so happeneth (God wotte the more pitye) yet in
speakinge at the leaste, let vs kepe a natural order, and set the man before
the woman for maners sake.’22 Reiterated in the eighteenth-century prolifer-
ation of grammar books and strengthened by laws that described wives as
legally subsumed by their husbands, such pronominal prescriptions straddle a
line between the prescriptive and descriptive, proscribing particular usages as
both socially unacceptable and wrong but also claiming to situate such
proscriptions within the describable grammar of English. As with Truss’s
complaints about punctuation some four and one-half centuries later, Wilson
eVectively elides distinctions between grammar and usage in ways that en-
hance the claims of the latter by putatively situating them not in stylistic
choice but in the objective, descriptive facts of English.
Such elision likewise has underwritten consideration of the noun man,
whether alone or in compounds like chairman, when it was argued to retain a
sense of ‘human being’ over that of ‘male human being’ on the grounds that
this would have been an original, etymological meaning. And it underwrites
advocacy for generic he (rather than they) as necessary for references to
singular non-gendered antecedent nouns like author or president, on the
grounds that concord in number transcends other grammatical issues and
that the masculine pronoun subsumes the feminine. Both kinds of arguments
defy history as well as logic in order to make putatively grammatical claims
about what are ultimately social discourses and practices. The etymological
fallacy (as Curzan labels it) may account for some of Milton’s verbal excess in
Paradise Lost, but it explains little else about how the meanings of words
evolve. While the use of generic they with singular antecedents has been

21 Curzan, Gender Shifts in the History of English, 123.

22 Wilson, The Arte of Rhetorique (1553; rpt. Gainesville, FL: Scholars’ Facsimiles and Reprints,
1962), f. 89r . See also Ann Bodine, ‘Androcentrism in Prescriptive Grammar: Singular ‘‘They’’, Sex-
indeWnite ‘‘He’’, and ‘‘He or She’’ ’, in Deborah Cameron (ed.), The Feminist Critique of Language:
A Reader (London: Routledge, 1990), 166–86.
46 Language Anxiety

recorded since the Middle Ages, early modern advocacy of generic he rather
than she presumed and appealed to the typicality (if not naturalness) of men
in certain professions, such as the law and medicine, and thereby blurred
still further any distinction between internal grammar and external usage.
And this kind of reasoning to language from social expectations, in turn, helps
sustain other gendered patterns in English language usage: the fact that insults
speciWcally directed at females tend to be more sexually charged than
those directed at males, and the fact that in pairs of semantically parallel
words (such as bachelor and spinster, or stud and bitch), the female word has
been the more likely to take on a negative connotation.
Changes involving the gender of English pronouns have been as consistent
and structured as phonological change, and their signiWcance has been just as
situational. Indeed, usage continues to evolve in response to social pressures,
with the quasi-natural generic he, itself the replacement of grammatical
gender, still sometimes advocated in formal discussions of the pronominal
system alongside (more typically) they, he or she, he/she, or s/he. All these
pronominal changes have likewise not been absolutely rational, since there is
no morphological or semantic analysis by which it would be true to claim that
a language with grammatical gender is somehow clearer or superior to one
with natural gender; today, many languages have one, many languages the
other. And though the movement from primarily grammatical to primarily
natural gender may have been structured and may have proceeded along clear
chronological and geographic axes, there is no evidence in its earliest stages
for the kind of social motivations (of class, age, or provenance) that have been
identiWed in modern changes and claimed for the Great Vowel Shift, nor is
there any evidence of conXict or anxiety occasioned by the movement.
Anxiety did arise, of course, when the pronominal system became implicated
in social attitudes and institutions. Just as the history of gender and pronouns
in English thus complicates any easy distinction between descriptive and
prescriptive grammar, so it undermines an interpretation of changes sub-
sumed by the former as more natural and internal than those subsumed by
the latter. An initial movement towards natural gender was less a monolithic
grammatical transformation and more a shift in emphasis among competing
Old English systems, just as a shift in analytical emphasis reconceived patterns
of usage as not simply grammatical but connected to social verities. Gram-
matical gender in eVect became natural gender, gaining signiWcance and
controversy in the process. The classiWcation of linguistic change, it seems,
can be as opportunistic as the change itself, and as this classiWcation changes,
so, too, does the language.
 A Moveable Speech 47

Phenomena like the Great Vowel Shift, the First Consonant Shift, and the
history of gender in English illustrate several crucial features of change and
the social meaning that sometimes accrues to it. The immediate question of
where sounds and grammatical forms go when they go away turns out to
require an answer that responds to several levels of language and its use. From
a purely structural perspective, it could be said that wherever grammatical
forms go, they characteristically go away in fashions that are structured,
multiform, and, if not strictly speaking rational, then comprehensible. Once
begun, push-chain mechanisms like that which describes the Great Vowel
Shift organize disparate data into coherent and stable patterns that imbue
phonological change with a kind of simplicity and inevitability. And the
incipience of natural gender among Old English nouns that referred to
human males likewise fosters the understandable and structured impact of
discursive practices on the development of language. Where sounds and
forms go, from this perspective, depends on certain empirical tendencies
and limitations for languages and their changes. Methodologically, that is,
historical linguistics relies on the same premise that underwrites historical
inquiry in general. This is the Uniformitarian Principle, by which, as Lass
describes it, ‘Nothing (no event, sequence of events, constellation of proper-
ties, general law) that cannot for some good reason be the case in the present
was ever true in the past’.23 What is impossible in modern theories of
linguistic structure, change, and even novelty thereby becomes impossible
for historical periods of a language: stops may add or lose voicing, but they
cannot become vowels; vowels and consonants may change in quantity and
quality, but for a given language they cannot all disappear at once, leaving
the language with only glides; inXectional morphology may atrophy and alter,
but syntax cannot be completely random. And novelties may appear among
both individual forms and general linguistic principles, but they can do so
only in ways allowed by the principles or in ways that require reconWguration
of the principles.
From another perspective, where sounds and forms go is signiWcantly con-
nected to speakers’ attitudes. Most immediately, speakers may consciously
modify the direction of a change by (say) accelerating the leveling of inXectional

23 Lass, On Explaining Language Change (Cambridge: Cambridge University Press, 1980), 55. The
whole of Lass’s discussion on 45–63 is valuable, as is his further consideration of these issues in
Historical Linguistics and Language Change (Cambridge: Cambridge University Press, 1997), 24–32. See
also Suzanne Romaine, Socio-historical Linguistics: Its Status and Methodology (Cambridge: Cambridge
University Press, 1982), 122–3; William Labov, Principles of Linguistic Change, i, Internal Factors
(Oxford: Blackwell, 1994), 10–27, and ‘On the Use of the Present to Explain the Past’, in Philip Baldi
and Ronald N. Werth (eds.), Readings in Historical Phonology: Chapters in the Theory of Sound Change
(University Park, PA: Pennsylvania State University Press, 1978), 275–312.
48 Language Anxiety

morphology or by avoiding socially stigmatized vocabulary. Or, whether con-

sciously or unconsciously, they may collectively use one vowel realization more
than another and thereby produce change. But speakers also eVect change in a
more conceptual fashion. They may, that is, interpret some linguistic phenom-
ena as internal and natural (and therefore permissible) and other phenomena as
external and artiWcial (and therefore to be resisted and rejected from a language’s
grammar). Or they may conXate these categories, regarding a stylistic choice
such as ‘father and mother’ as grammatical in a way that ‘mother and father’ is
not, and in this way utilize grammatical claims to sustain a particular social
view of the relationships between males and females. Here is where the ontology
of language becomes a factor, for it seems to take its identity from speakers’
perceptions as well as from empirical data. From a purely structural point of
view, for instance, Smith’s deWnition of language change makes good sense by
tying change to speakers’ usage: ‘a sound change has taken place when a variant
form, mechanically produced, is imitated by a second person and that process
of imitation causes the system of the imitating individual to change’.24 Yet as
for many linguistic topics, the simple criterion of the number of domains or
users does not allow for a precise classiWcation built on the principle of majority-
rule. Otherwise, the widespread pronunciation of ask as [æks] or the use of
double negatives would occasion no comment. These limitations of use-based
deWnitions also appear in the evaluation of varieties of English as well as of
speciWc forms. Received Pronunciation thus continues to be cited as the most
prestigious and socially advantageous accent in the United Kingdom, the accent
from which all other accents necessarily deviate, even though estimates routinely
claim no more than 3% of the population speak what is itself a variable accent.25
From this vantage, then, where grammatical forms go depends on what
speakers understand them to be and where they want them to go. To describe
linguistic phenomena as change is to categorize them as modiWcations of
English’s descriptive grammar and to claim, implicitly, that one form has
become another, retaining its former self only as a historical trace and without
violating the language’s integrity. Middle English /i/ thus became Modern
English /ai/ and no longer exists except in its transformed state; the /i/ in
words like feet represents a new /i/, arising from a diVerent but related change.
To regard linguistic phenomena as violations of either descriptive or prescrip-
tive grammar, however, is to represent them not as change but as error;
pronunciation of a word with a historic Middle English /i/, like Wve, as if

24 Smith, Sound Change and the History of English, 70.

25 e.g., David Crystal, The Cambridge Encyclopedia of the English Language (Cambridge: Cambridge
University Press, 1995), 365.
 A Moveable Speech 49

the vowel were /o/ rather than /ai/ would thus generally be considered simply
wrong, however many speakers actually used the pronunciation and whether
the error is due to circumstance or imperfect acquisition. And in this case,
particularly with respect to prescriptive grammar, the categorization of error
can carry with it moral valuations dependent on whatever social meanings
have been invested in the forms. Here, one form remains as the standard
against which other forms can be identiWed as imperfect or corrupted.
Judgements like these thus derive not simply from empirical data and a
Wxed taxonomy of linguistic phenomena. One of the fundamental achieve-
ments of Labov’s pioneering sociolinguistics work, indeed, was the demon-
stration, in both New York City and Martha’s Vineyard, that the origins of
phonological diachronic change can lie in synchronic variants, their distribu-
tion, and valuation.26 Yet only retrospection allows such a demonstration,
since not all synchronic variation will result in diachronic change, and at any
given moment speakers have little way truly to predict which variation,
whether of structure or repertoire, will result in change. ClassiWcations of
variation, error, and change, then, also depend on speakers’ choices: over
which phenomena have social meanings, over what the consequences of those
meanings may be, and over how much variation is tolerable for both linguistic
form and social meaning. In eVect, the conceptual category of change vali-
dates linguistic transformation—it constitutes aYrmation that one form has
become another while still preserving the essential characteristics of the
language and that as such the form and its change preclude additional value
judgements. And yet as categorical and consequential as this aYrmation may
be, it is also fundamentally situational. What speakers choose to classify as
change, variation, or error, indeed, bespeaks their larger senses of their
language, their society, and the relations of the two. Speakers’ concomitant
judgements about a particular change—whether it’s good, bad, or indiVer-
ent—ultimately testify more for their views of themselves and their language
than for the change itself. Variation and change in the gender of pronouns
began to matter, then, when they were invested with sociolinguistic meaning.
Among linguists it has become axiomatic that many linguistic phenomena
have situational and not transcendent meanings. Discourse analysis has
shown that features like intonation or silence, depending on who is using
them, when and how, can sometimes supplement meaning in one way and
sometimes in another. Sometimes silence can signal power, other times
submission. In this vein, I do not believe there is anything conceptually
revolutionary in the claims I have made about change and variation. This

26 Labov, Sociolinguistic Patterns (Philadelphia, PA: University of Pennsylvania Press, 1972), 1–69.
50 Language Anxiety

same situational force that discourse analysis recognizes, I am arguing,

underwrites interpretations and categorizations of varying grammatical
forms and, ultimately, of what is and is not English. Whether particular
linguistic features are judged synchronic variants, diachronic developments,
errors, or inconsequentialities depends a good deal on who is doing the
judging and when.

Sociolects, dialects, history

Turning from the details of grammar to the larger structures they make up,
I want now to consider how the inevitability, regularity, and susceptibility to
social valuation that I have traced in the change of grammatical forms Wgure
in the change and variation that produce linguistic varieties. Inevitability
appears in the fact that, irrespective of the internal mechanisms that drive
grammatical change, social and regional variation arises from geography,
history, economics, and politics. To some extent, as I noted earlier, this kind
of variation can be passive. Faced with social and cultural phenomena for
which their language has no expression, speakers may quite unconsciously
borrow words and structures that strike them as better suited to describe the
world around them. This is why Anglophone visitors to North America
borrowed words like raccoon and toboggan from Algonquian.
Such variation can also be active, however, in the sense that speakers can
use varieties as well as forms for social deWnition. Like music, clothing, or hair
style, forms and varieties of a language are in part social display, and as such
they have meaning against a variety of sociolinguistic institutions and atti-
tudes, such as sixteenth-century prescriptions on the syntactic ordering of
father and mother, which reXect social gender more than linguistic structure.
Beyond whatever cognitive content they convey, that is, particular forms (e.g.,
pronunciations) and regional and social varieties can sometimes (but not
always) have pragmatic signiWcance with potential implications for social
status. In this way, whether speakers realize as much or not, linguistic forms
serve self-identifying purposes for speaker and listener alike. For the former,
they are ways to declare one’s solidarity with or separation from others by
willy-nilly evoking a particular variety and its sociolinguistic associations; for
the latter, drawing on the same heuristic framework, linguistic forms and
varieties serve as means to assess a speaker’s ethnicity, provenance, education,
reliability, and so forth. And in this way, so long as speakers move about, meet
one another, and engage in social interaction with one another—which is as
much to say, so long as they and their language are alive—change and
variation must take place. The presence of regional and social variation,
 A Moveable Speech 51

indeed, might well be regarded as illustrative of the Uniformitarian Principle;

if a monolithic language without any variation is impossible now—and it is—
so must it have been impossible at any moment in the past. The histories of
transplanted English in post-colonial regions in fact make just this point. By
consistently enacting the same basic processes in changing from an imposed
language to one with its own indigenous norms, structurally as well as
sociolinguistically, the English of regions like India and Nigeria seems to
bespeak a uniformly shared pattern for adapting and nativizing language.27
Together, internal linguistic variation, speakers’ needs to respond to new
experiences, and the uses of language for self-identiWcation have produced—
and continue to produce—a proliferation both of national versions of English
and of regional and social varieties of those same national versions. As a
national version, Australian English paradigmatically illustrates this conXux
of regular internal linguistic and external contextual pressures. The inXuence
of the former appears in the fronting of the vowels /^/ (up) and /A/ (father),
which seems to be common to both Australia and eighteenth-century London
and thus attests to the linguistic substratum of the language of southeast
England spoken by many of the original immigrants. It is sociolinguistic
context, however, that accounts for historical regionalisms like frisk for pick
and cly for pocket, both of which point to the storied role of prisoner
transportation in the country’s history. Similarly, contact with Aboriginal
languages, other non-English languages, and the region’s landscape is evident
in bettong, cooliman, and billabong—distinctively Australian English words
for distinctively Australian phenomena: a kangaroo rat, a wooden water
vessel, and a branch of a river.28 In Robert BurchWeld’s overview of how
national and regional varieties of English developed in general, ‘the introduc-
tion of English in diVerent physical and cultural environments has everywhere
produced a similar set of results: markedly distinguishable speech patterns in
each of the regions, considerable diversity of local vocabulary (the need

27 Welsh oVers an interesting qualiWcation about the persistence of regional varieties, for with the
language’s revival centered on schools and not at home, a standardized variety has spread at the
expense of previous regional ones. If the language were to be fully revived, however, one would expect
the re-emergence of regional varieties and sociolects. See Mari C. Jones, Language Obsolescence and
Revitalization: Linguistic Change in Two Contrasting Welsh Communities (New York: Oxford University
Press, 1998). On the development of transplanted English, see Edgar W. Schneider, Postcolonial English:
Varieties around the World (Cambridge: Cambridge University Press, 2007).
28 See John Gunn, ‘Social Contexts in the History of Australian English’, in Tim William Machan
and Charles T. Scott (eds.), English in Its Social Contexts: Essays in Historical Sociolinguistics (New York:
Oxford University Press, 1992), 204–29; and George W. Turner, ‘English in Australia’, in The Cambridge
History of the English Language, v, Robert BurchWeld (ed.), English in Britain and Overseas: Origins and
Development (Cambridge: Cambridge University Press, 1994), 277–327.
52 Language Anxiety

elements being chieXy drawn from the languages of the indigenous inhabit-
ants) and the essential sameness (with only limited exceptions) of the acci-
dence and syntax’.29 Put another way, social and regional varieties of English
have been inevitable and inevitably reXective of the circumstances of language
change and variation.
Regularity as well as inevitability appears most simply in the consistency of
divergences between corresponding grammatical forms among varieties.
These forms diverge not necessarily because speakers of one variety are trying
and failing to achieve the grammar of another but because their own variety
represents a nativization of that or another grammar. Among speakers of
South Asian English, for instance, Standard British English [t] and [d] are
characteristically realized with the tip of the tongue curled back (‘retroXex’),
in accordance with indigenous phonologies. Similarly, African American
Vernacular English speakers who delete Wnal stops regularly articulate present
(e.g., walk) and dental preterite (walked) forms in the same manner ([wOk])
and thereby reXect the inXuence of a substratum of creoles and West African
languages. Earlier I noted that the interpretive maneuver of deriving one form
from another—perhaps retroXex [t] from alveolar [t]—can misleadingly use
the objectivity of descriptive grammar to present what is ultimately a pre-
sumptive judgement about which form is normative. Such representation can
represent South Asian English in general, that is, as not only derivative but
wrong. An additional important point here is that just as rules like ‘use
retroXex [t]’ and ‘delete Wnal stops’ are themselves not absolutely regular
but variable, speakers of Standard British or American English, whether they
recognize it or not, realize converse rules like ‘use alveolar [t]’ and ‘add [d] or
[t] to form preterites’ with similar variability. And thus the crucial point is
that the same kind of regularity underwrites both prominent and historical
national varieties of English and regional or social varieties of comparatively
recent development. To paraphrase Mark Twain, not only are we not all trying
to talk the same way, but we couldn’t do so even if we did try.
Both regularity and inevitability also appear in the persistence of social and
regional variation across time. The named varieties of English may have
changed, as have some of the features that are used to deWne them, but the
fact of identiWable types of English characterized by unique combinations of
shared linguistic features and attitudes has remained constant. Presumably,
when Bede’s Angles, Saxons, and Jutes arrived in England, mostly in the Wfth
and sixth centuries, they did so speaking not a uniquely monolithic language

29 BurchWeld, ‘Introduction’, in BurchWeld (ed.), English in Britain and Overseas: Origins and
Development, 13.
 A Moveable Speech 53

but varieties of West Germanic that reXected their origins and culture as
much as Australian English reXects its own. Some variations in the extant Old
English record, indeed, such as that between the ‘retracted’ Anglian form
haldan and West Saxon healdan, may well owe to pre-migration, continental
linguistic diversity, and the continental proximity of the Angles to speakers of
North Germanic varieties (Old Danish and Old Swedish in particular) may
account for other dialect diVerences in Anglo-Saxon England.30
The earliest English writs and charters survive from the seventh and eighth
centuries, and already by this time the written record certainly reXects struc-
tured regional variation. This early record means, then, that there has never
been a time in the recoverable history of English when the language did not
embody variation and change. The Venerable Bede himself, though indirectly,
provides a vivid illustration of this point. On his death bed in 735, reXecting
on this life and ordering his soul for the next, Bede (according to tradition)
looked back from his Latin literacy to Anglo-Saxon oral traditions of his
youth and uttered a hymn about the unpredictability of inevitable death:
Fore them neidfaerae naenig uuiurthit
thoncsnotturra than him tharf sie,
to ymbhycggannae, aet his hiniongae,
huaet his gastae, godaes aeththa yXaes,
aefter deothdaege doemid uueorthae.31

Or that, at least, is how the hymn appears in Bede’s native Northumbrian

dialect in a version written down one century after his death. Several centuries
later still, a slightly diVerent version of the song circulated, this one showing
some forms that characterize West Saxon, the variety of Old English that
became a kind of tenth-century koine and that remains the form most
commonly taught and read today:
For ðam neodfeore nænig weorðeð
ðances snotora ðonne him ðearf sy,
to gehycgenne, ær his heonengange,
hwæt his gaste, godes oððe yfeles,
æfter deaðe heonen demed weorðe.

30 The reason why I suggest ‘retraction’ may be pre-migration is that it is virtually complete in even
the earliest Anglian texts. See Alistair Campbell, Old English Grammar (Oxford: Clarendon Press,
1959), 55–6. On the possible linguistic consequences of the Angles’ continental proximity to speakers of
North Germanic, see Smith, Sound Change and the History of English, 88–106.
31 A. H. Smith (ed.), Three Northumbrian Poems (London: Methuen, 1933), 42–3. In S. A.
J. Bradley’s translation, the poem reads: ‘Before the inevitable journey no man shall grow more
discerning of thought than his need is, by contemplating before his going hence what, good or evil,
will be adjudged to his soul after his death-day’ (Anglo-Saxon Poetry [London: Dent, 1982], 6).
54 Language Anxiety

Even for a reader not conversant with Old English, the structured variation
among these versions recalls the regularity of the phonological changes
I described above. Leaving aside diVerences in wording (e.g., gehycgenne
for ymbhycggannae and deaðe heonen for deothdaege), which may owe to the
vagaries of transmission in a predominantly oral culture, we see consistent
variation in orthography (ð for th and w for uu) and inXectional morphology
(e for ae, eð for it). At least some of the variation, additionally, would seem to
have phonological signiWcance, reXecting diVerences in Old English accents;
neodfeore for neidfaerae suggests as much, as does oððe yfeles for aeththa yXaes.
While much written Modern English may erase or conceal such variation,
it has continued and, in doing so, has proliferated national varieties of
English, much as it did in the continental Germanic homeland as well as in
Anglo-Saxon England. And this same inevitability and regularity that created
national varieties like Canadian or Nigerian English produces regional var-
ieties of particular national ones. When completed, the Wve-volume Diction-
ary of American Regional English will document the history, usage, and
vicissitudes of millions of words in American English alone.32
Again as with judgements of grammatical forms like pronunciations, the
cultural valuation of such sociolects and regional dialects has a situational
quality that casts ‘English’ as an abstraction into sharp relief. Above all, as
Jonathan Marshall has argued, the spread and identity of such varieties is not
simply a structural matter: ‘mental attitude, solidarity, or orientation is the
driving force behind choices made about levels of social network integration
and language use. This can account for a high degree of correlation between
network indices and language uses, as well as for low degrees of correlation.’ In
this way, a kind of ‘mental urbanization’ may conduce rural speakers to model
their language not on nearby varieties but on a desirable one found in far
away cities.33 In a related situational vein that likewise reXects the ‘mental
attitude’ involved in the valuation of varieties, pidgin forms of English that
are stigmatized in Africa may have no social meanings at all in other Anglo-
phone communities, while overtly non-prestigious sociolects like Geordie
(spoken in Newcastle) often have a good deal of covert prestige among their
speakers. The mere fact that such valuation has a history—that there is a
beginning and development in the social valuation of regional varieties of
English—also points to this same situational quality, and it is to this history
that I now turn.

32 Frederic G. Cassidy, chief editor, Dictionary of American Regional English (Cambridge, MA:
Belknap Press of Harvard University Press, 1985–), 4 vols., with one in progress.
33 Marshall, Language Change and Sociolinguistics: Rethinking Social Networks (New York: Palgrave
Macmillan, 2004), 228–9.
 A Moveable Speech 55

To begin with Bede, whose own death song records regional variation, the
Ecclesiastical History identiWes social variation in its account of Imma, one of
Elfwin’s thanes. Captured by King Ethelred, Imma feigns to be only a poor
peasant, but ‘his appearance, clothing, and speech’ allow others to recognize his
noble background.34 The important point in this early recognition of dialect
varieties is that while Bede acknowledges Imma’s sociolect, he oVers no value
judgement on it, nor does he assign social meanings of any kind to the varieties
being used. Similar recognition of regional variation in Anglo-Saxon England
appears in certain eighth-century manuscript evidence that reveals speakers
altering the forms of their native dialects in approximation of forms from
Mercia, whose political ascendancy at that time could be thought to render its
variety of English prominent, if not prestigious.35 Like the tenth- and eleventh-
century propagation of West Saxon forms through the scriptorium at Winches-
ter, this kind of linguistic activity embodies an ambivalence about regional
variation comparable to Bede’s. On one hand, it certainly indicates awareness
of regional and social variation; on the other, it reXects a lack of recognition of or
interest in the relative sociolinguistic merits of varieties or the pragmatic utility
of any one variety in particular. The implication of these shared sentiments is
that the inXuence of both Wessex and Mercia lay in their political power, not in
their varieties of Old English. And all this is to say that in the Anglo-Saxon
period, however obvious the variations of English may have been, a clear
distinction existed between them and their social valuations, since speakers do
not in fact seem to have assigned the latter at all.
By the Middle English period, especially the Wfteenth century, there are
indications that speakers had begun to conceptualize varieties of the language
within a linguistic repertoire of not only English dialects but also non-English
languages. The survival of works in multiple copies, for example, increasingly
reveals dialect translation, by which a text written in one regional variety was
rendered in another one, presumably for reasons of intelligibility and mar-
ketability. A colophon in a copy of Richard Rolle’s fourteenth-century Form of
Living even records that the text has been ‘translat out of Northarn tunge into
Sutherne that it shulde be the bettir vnderstondyn of men that be of the selve
countre’.36 Likewise, scribes sometimes drew attention to variation in English

34 See Bede, A History of the English Church and People, trans. Leo Sherley-Price (Harmondsworth:
Penguin, 1955), 244.
35 See Thomas E. Toon, ‘The Social and Political Contexts of Language Change in Anglo-Saxon
England’, in Machan and Scott (eds.), English in Its Social Contexts, 28–46.
36 M. B. Parkes and Richard Beadle (eds.), The Poetical Works of GeoVrey Chaucer: A Facsimile
of Cambridge University Library MS Gg.4.27, 3 vols. (Norman, OK: Pilgrim Books, 1979–80), iii, 55. See
also Cursor Mundi, ed. Richard Morris, EETS os 57, 59, 62, 66, 68, 99, 101 (London: Trübner, 1874–93),
56 Language Anxiety

by specifying the regional form they used, as did Dan Michel in his 1340
Ayenbite of Inwyt, for which he used the ‘engliss of kent’.37 Perhaps the most
famous (if still general) comments of this sort are William Caxton’s remarks
in his preface to the Eneydos, where he observes that ‘our langage now vsed
varyeth ferre from that. whiche was vsed and spoken whan I was borne’ and
then recounts the confusion confronting a merchant who, while in Kent,
uttered the northern form eggs rather than the regional eyren.38
Although sporadic comments like these do not imply a clear hierarchy or
social valuation of dialects, Middle English itself did help to sustain institu-
tionalized conceptions of England’s three most proliWc languages: English,
French, and Latin. In his 1387 Testament of Love Thomas Usk described with
particular clarity the distinct roles for each of these languages: ‘Let than
clerkes endyten in Latin, for they have the propretee of science, and the
knowynge in that facultee; and let Frenchmen in their Frenche also endyten
their queynt termes, for it is kyndley to their mouthes; and let us shewe our
fantasyes in such wordes as we lerneden of our dames tonge.’39 And the
preface to the Cursor Mundi distinguishes its primary, ignorant (‘lewed’)
readership from knowledgeable readers conversant with Latin and French:
And on Inglysch has it schewed
Not for þe lerid bot for þe lewed:
For þo þat in þis land wone
Þat þe Latyn no Frankys cone.40

It is not until the early modern period, with the concomitant spread of
printing, literacy, regularized orthography, and standardized language,
that widely shared social valuations of English dialects emerge, however.
Even as it lauds the emergence of English as a national language, for example,
George Puttenham’s 1589 The Arte of English Poesie oVers a sociolinguistic

ll. 20,061–4. On Middle English dialects more generally, including dialect translation, see Margaret Laing
(ed.), Middle English Dialectology: Essays on Some Principles and Problems (Aberdeen: Aberdeen Univer-
sity Press, 1988); James Milroy, ‘Middle English Dialectology’, in The Cambridge History of the English
Language, ii, N. F. Blake (ed.), 1066–1476 (Cambridge: Cambridge University Press, 1992), 156–206; and
Felicity Riddy (ed.), Regionalism in Late Medieval Manuscripts and Texts: Essays Celebrating the Publication
of ‘A Linguistic Atlas of Late Medieval England’ (Woodbridge: D. S. Brewer, 1991).
37 Dan Michel, Ayenbite of Inwyt, ed. Richard Morris, EETS os 23 (London: Trübner, 1866), 262.
38 Caxton, The Prologues and Epilogues of William Caxton, ed. W. J. B. Crotch, EETS os 176 (1928;
rpt. New York: Burt Franklin, 1971), 108.
39 Usk, Testament of Love, in Chaucerian and Other Pieces, vii, Complete Works of GeoVrey Chaucer,
ed. W. W. Skeat (Oxford: Oxford University Press, 1897), 2.
40 Cursor Mundi, ll. 5–8. Also see ll. 21–6 and 237–40. For more discussion of the conceptualization
of dialects in the Middle English period, see Machan, English in the Middle Ages (Oxford: Oxford
University Press, 2003), 86–96.
 A Moveable Speech 57

conceptualization of English varieties that bespeaks attitudes far advanced from

those of Dan Michel or Caxton. The language of the court and town, he
observes, is much superior to that of ‘marches and frontiers’ or even universities,
where Schollers vse much peeuish aVectation of words out of the primatiue languages,
or Wnally, in any vplandish village or corner of a Realme, where is no resort but of
poore rusticall or vnciuill people; neither shall he [the aspiring poet] follow the speach
of a craftes man or carter, or other of the inferiour sort, though he be inhabitant or
bred in the best towne and Citie in this Realme, for such persons doe abuse good
speaches by strange accents or ill shapen soundes, and false ortographie. But he shall
follow generally the better brought vp sort, such as the Greekes call [charientes] men
ciuill and graciously behauoured and bred.41

In eVect, Puttenham conXates social and regional variation by linking pro-

scribed varieties, ‘vplandish’ locales, and ‘vnciuill’ speakers, and, as Chapter 5
will consider in detail, this kind of linking develops and spreads in the
seventeenth and eighteenth centuries in ways that Wrmly attach valuations
to synchronic variation in general and certain varieties in particular. For the
Wrst time in a systematic and institutional fashion, then, the facts of linguistic
change and variation were thereby broadly implicated in social signiWcance
and institutional propagation; and if codiWcation proliferated grammatical
prescription, schools came to embed it in business, politics, and the processes
of social advancement. By the nineteenth century, as the next chapter will
discuss, a hierarchy of English varieties and their sociolinguistic implications
was well enough established that writers like Dickens and Twain could use it
for literary eVects.
Today, as Denis Preston’s studies of what he calls perceptual dialectology
demonstrate, such a hierarchy has become a naturalized part of speakers’
sociolinguistic horizons. Indeed, Preston’s work shows just how far the social
valuation of English varieties—as distinct from empirical recognition of their
existence—has developed since the Anglo-Saxon period. Over several years he
has surveyed speakers on what they judge to be the spectrum of best to worst
varieties of American English and also on where on this spectrum they would
place their own variety. What he’s found is that speakers’ perceptions of
the spatial limits of American regional dialects are not empirically based but
rather reXect their own geographic biases, while their evaluation of their own
variety often reXects more broadly shared views, even when those might
involve the devaluation of their own variety. Testifying still more forcefully
to the internalization of value judgements on variation is the widespread

41 Puttenham, The Arte of English Poesie (1589; rpt. Menston: Scolar Press, 1968), 120.
58 Language Anxiety

acceptance of the very notion of language as a mediation of social status

within a Wxed interpretive framework. ‘By placing language itself outside both
social environment and cognitive embedding,’ observes Preston, ‘folk lin-
guists in the USA are able to posit an easily accessed exterior standard, one
which good people will access with little eVort and place themselves where
they belong—in the mainstream of US society.’42 The inXuence of this
standard can be apparent whether it is endorsed or resisted. In the case of a
predominant working-class variety of English in New York City—variously
labeled Brooklynese or New Yawkese—characteristic speech patterns that
reXect the city’s immigrant history now compete with both the social stigma-
tization of the variety (as in forms like [Oi] for [G], roughly oi for er, in
thirty-third) and the inXux of new forms reXecting new patterns of immigra-
tion. To some speakers, such competition inspires greater dialect loyalty to a
stigmatized yet historical and unique variety; to others, it suggests progres-
sive, sociolinguistic assimilation. To all such speakers, the fact that this kind of
competition has meaning reXects naturalized social responses to what is in the
end, to return to Lass, neutral linguistic variation.43
The absence of standard language in the middle ages may make medieval
dialect diversity the most apparent in the written record of English,44 but
every period in the language’s history has been dialectal. Whatever impression
the regularity of modern written English may give, this regularity has done
nothing to decrease the persistence and proliferation of regional and social
varieties. In fact, given the factors that produce social and regional variation,
and given the one billion people now commonly estimated to speak English,
the modern period would seem likely to be quantitatively more dialectal than
the Middle English one. This claim can be pushed even further. Identifying,
counting, and comparing varieties from one epoch of English with those of
another is a fraught enterprise, complicated in the early periods by the relative
paucity of written records, in later periods by the absence of electronic media
for preservation, and in the contemporary period by the leveling of spoken
variation in written representations. Yet with perhaps two-thirds of the
world’s Anglophones residing outside of the United Kingdom and the United
States combined, current sociolinguistic conditions would seem to oVer more

42 Preston, ‘The Story of Good and Bad English in the United States’, in Richard Watts and Peter
Trudgill (eds.), Alternative Histories of English (London: Routledge, 2002), 148. Also see Preston,
Perceptual Dialectology: Nonlinguists’ Views of Areal Linguistics (Dordrecht: Foris Publicatons, 1989).
43 Deborah Sontag, ‘Oy Gevalt! New Yawkese An Endangered Dialect?’, New York Times, 14 February
1993, A1, 18.
44 e.g., David Crystal, The Stories of English (Woodstock, NY: Overlook Press, 2004), 194; and Seth
Lerer, Inventing English: A Portable History of the Language (New York: Columbia University Press,
2007), 99.
 A Moveable Speech 59

opportunities for dialect diversity than at any moment in English’s history.

Even if, for the sake of the argument, one accepts Chomsky’s ideal speaker-
listener and its implications about the foundational grammatical regularity
and sameness of English, there remains an abundance of grammatical and
discursive diversity among the linguistic varieties subsumed by this label. For
this reason, as the designation of a coherent and stable variety, Modern
English—even just South African or Australian English—represents an ab-
straction of a high order. And just plain English, with its almost cavalier
disregard of synchronic variation and diachronic change, is an abstraction of
a still higher order.45
Discussions of the history of English often contain illustrations that show
historical change as documented in a repeatedly reissued work such as the
Bible. Here, then, is Genesis 3: 1 in four versions stretching from the Anglo-
Saxon period, through Middle and Early Modern English, to Modern English:
Old English Heptateuch: Eac swylce seo næddre wæs geapre ðonne ealle ða oðre nytenu
ðe God geworhte ofer eorðan. And seo næddre cwæð to ðam wife: ‘Hwi forbead God
eow ðæt ge ne æton of ælcon treowe binnan Paradisum?’
Second WycliYte Version: But and the serpent was feller than alle lyuynge beestis of
erthe, whiche the Lord God hadde maad. Which serpent seide to the womman, ‘Why
comaundide God to Zou that Ze schulden not ete of ech tre of paradis?’
King James Version: Now the serpent was more subtil than any beast of the Weld which
the Lord God had made. And he said unto the woman, Yea hath God said, Ye shall not
eat of every tree of the garden?
New International Version: Now the serpent was more crafty than any of the wild
animals the Lord God had made. He said to the woman, ‘Did God really say, ‘‘You
must not eat from any tree in the garden’’?’

Comparisons like this usefully highlight signiWcant features of diachronic

change in English, including: orthography (treowe > tre > tree); morph-
ology (seo næddre > the serpent); lexicon (geapre > feller > more subtil
> more crafty); and syntax (God geworhte > God hadde maad > God had
made). They attest to continuity as well as development in the expressive
potential of English and its grammatical structure. Like matter-of-fact listings
of Indo-European cognates, what such comparisons do not do is acknowledge

45 For a variety of reasons, such as the diYculty in empirically identifying levels of linguistic
competence and the well-known tendency of speakers to misreport their own competence in one
way or another, hard and fast Wgures on the numbers of Anglophones today are impossible to
produce. See the data—and qualiWcations—in David Crystal, English as a Global Language, 2nd
edn. (Cambridge: Cambridge University Press, 2003), 59–71.
60 Language Anxiety

their tacit distinctions between change and variation or engage the theoretical
and social investment in such distinctions.
Bopp and Müller may have passed value judgements over alterations in
inXectional morphology, but, as I argued in the previous section, for the most
part long-term, expansive transformations of the sort that distinguish Middle
from Old English (or Germanic consonants from Indo-European ones) have
been categorized as change and therefore less subject to the charge of being
errors than are those linguistic phenomena that are considered synchronic
variations. Here again, however, the neutrality of change can be deceptive.
I would venture that of these four versions the structural diVerences between
all but the King James and New International texts of Genesis would seem to
exceed the diVerence between any two modern varieties, such as Australian
and contemporary British English. And I would also venture that there would
similarly seem to be greater structural diVerence among these four biblical
versions than there is between any two varieties of either Old English or
Middle English, or between any two current United States varieties. And yet
despite such diVerences, all these varieties are labeled ‘English’ on the con-
viction—and with the implication—that they arise through diachronic
change that is accepted as internal and natural for the language.
There is, of course, something profoundly circular in such reasoning:
diachronic change is that which maintains a language’s integrity over time,
while a language like English remains English over time, so long as its alter-
ations reXect diachronic change. And it is this fundamentally impressionistic
value judgement, I suggest, that works to preclude controversies comparable to
those associated with social and regional varieties of English from the end of
the seventeenth century. By extension, such a judgement fashions language
and its history as simple, objective facts. Whether or not a change is accepted as
historical, however, has social signiWcance for both it and the language. While
seo næddre ‘became’ the serpent without violating the integrity of English—at
least according to histories of English—some of the most trivial synchronic
diVerences can take on the greatest signiWcance, as when the popular com-
mentator John Simon described the construction ‘between you and I’, a
construction that dates to the early modern period, as ‘an unsurpassable
grossness’ that must be eradicated before it produces ‘every kind of deleterious
misunderstanding’.46
At the outset of this section I noted that judgements about linguistic forms
and varieties conditionally and contextually reXect the speakers who make
them. This same point can now be made more pointedly as a question: when

46 Simon, Paradigms Lost (New York: Potter, 1980), 18, 21. The emphasis is in the original.
 A Moveable Speech 61

is language, its change, or its variation a signiWcant part of social or cultural

identity and therefore subject to social valuation? The evidence of English
varieties, their histories, and their evaluations suggests that the answer is:
when someone to whom the issue matters and who has the authority to
inXuence others, whether or not that person is part of the relevant speech
community, says it does. It is in this way that the categorization and social
meaning of particular linguistic forms or varieties are assigned; they are not
transcendent or based on empirical, structural categories. However strong
some sentiments about language can be, clear distinctions between change
and variation are in fact often diYcult to maintain. To put the matter
concisely: the history of English is itself a value judgement about which
forms and varieties constitute the language and which do not. In the next
section of this chapter I extend this argument by considering how English,
through contact with other languages, not only underwent variation and
change but also aVected the contact languages, frequently including the
shift of their speakers to English.

When languages collide

In 1599, barely a decade after the defeat of the Spanish Armada and England’s
emergence as a world power, Samuel Daniel published a small volume of
poetry entitled Poeticall Essayes, which included the poem Musophilus, Con-
taining a General Defense of All Learning. Despite England’s stature and
growing prominence, some of the country’s most consequential global
achievements still lay in the future: the founding of the British East India
Company in 1600; the settlement of Jamestown, Virginia, in 1607; and the
arrival in New Zealand of Captain James Cook in 1769. Nonetheless, in a
display of optimism and foresight, Daniel waxed rhapsodic about the spread
of not only English power but also the English language:
And who, in time, knowes whither we may vent
The treasure of our tongue, to what strange shores
This gaine of our best glory shall be sent,
T’inrich vnknowing Nature with our stores?
What worlds in th’yet vnformed Occident
May come reWn’d with th’accents that are ours.47

In the Wrst Xush of English expansion, language and all the ideas and
institutions it enabled would be England’s gift to a less fortunate world.

47 Daniel, Musophilus: Containing a General Defense of Learning, ed. Raymond Himelick (West
Lafayette, IN: Purdue University Studies, 1965), ll. 957–62.
62 Language Anxiety

Expressing widely shared views, Daniel saw English as inherently a treasure and
glory and therefore, like English government, religion, and technology, a means
for improving the unknown, primitive world. Slightly over three centuries later,
Mahatma Gandhi took a rather dimmer view of this gift: ‘To give millions a
knowledge of English is to enslave them. The foundation that Macaulay laid of
education has enslaved us. I do not suggest that he had any such intention, but
that has been the result.’48
Such diVering views partly depend on the individual speakers, of course,
but also on these speakers’ diVering epochs and the varying outcomes of
English’s contact with other languages in Daniel’s and Gandhi’s lifetimes.
Whether individual or institutional, this kind of contact has always charac-
terized all natural languages, occurring whenever speakers travel for business,
pleasure, war, or adventure. And language contact and its eVects upon the
languages involved displays the same kind of inevitability and regularity that
I have traced in structural and dialectal variation and change. Since mono-
linguals predominate in many Anglophone populations, including those of
England and the United States, to many speakers the normativeness of
language contact and multilingualism may seem less apparent. Yet they are
institutionalized in the multiple oYcial languages of Belgium, Switzerland,
and South Africa, evident in the hundreds of languages that co-exist in India
and Indonesia, and personalized in the fact that nearly the entire adult
population of Israel is at least bilingual.49 Even in predominantly Anglophone
societies, however, multilingualism is the norm. According to one study, as
many as 307 diVerent languages are used by London school children in their
homes, while the 1996 Australian census records 240 languages and the fact
that over a quarter of the population of both Melbourne and Sydney speak a
language other than English at home.50 The 2000 United States census
indicates that nearly 18% of the population over the age of Wve speak a
language other than English at home, with over half this group (10% of
the total population) speaking Spanish; about 3.8% speak another Indo-
European language and roughly an additional 2.5% speak an Asian or PaciWc
Island language. All totaled, perhaps four hundred languages besides English
are spoken in the United States.51 In Canada, the 2001 census oVers similar

48 Quoted in Richard W. Bailey, Images of English: A Cultural History of the Language (Ann Arbor:
University of Michigan Press, 1991), 144.
49 On Israel, see Eliezer Ben-Rafael, Language, Identity, and Social Division: The Case of Israel
(Oxford: Clarendon Press, 1994).
50 Kingsley Bolton, Chinese Englishes: A Sociolinguistic History (Cambridge: Cambridge University
Press, 2003), 111; and Michael Clyne, Dynamics of Language Contact: English and Immigrant Languages
(Cambridge: Cambridge University Press, 2003), 23.
51 US Census, Table 5 ([Link]
 A Moveable Speech 63

numbers: approximately 82% of the population know only one or both of

the oYcial languages—English and French—and the remaining individuals,
though they may speak one of these languages, also speak another.52
A claim for the ubiquity of multilingualism is supported by the past of
English as well as its present. In the twelfth century, for example, Great
Britain’s majority language was certainly English, and many of its approxi-
mately three million residents were monoglots. But Latin was still spoken in
monasteries and schools and remained the dominant written language in
theology, science, philosophy, law, and other such ecclesiastical and civil
activities. In the southeast in particular, descendants of the Norman and
Angevin aristocracy spoke French, which was also increasingly used in legal
and ecclesiastical domains. Cornish was spoken in the southwest, Welsh in
most of modern Wales (with much of the population being monoglots),
Gaelic in most of Scotland, and Irish in most of Ireland. Beyond this, Flemish
would have been heard in Wales and Norse in the Shetland and Orkney
islands and also, possibly, in Yorkshire and the London area, where Hebrew
was spoken, too.53
But if England’s political expansion in the early modern period continued
rather than initiated English’s interactions with other languages, it also
accelerated such contact. And this acceleration signiWcantly involved not
only new contact languages (such as Maori, Zulu, and Navajo) but also new
contact situations, in which English served as the colonial language of tradi-
tion, government, education, and technology. At the same time, the contact
languages served as the mediums of indigenous cultures and the varying
desires of individuals to maintain their culture or assimilate with the Anglo-
phones. Such dynamics of language contact have been well documented as the
impetus that frequently turns the potential for change, which is always
present, into its reality.54 It was continental contact with North Germanic,

52 Language Composition of Canada: Highlight Tables, 2001 Census ([Link]

english/census01/home/[Link]).
53 R. I. Page, ‘How Long Did the Scandinavian Language Survive in England? The Epigraphical
Evidence’, in Peter Clemoes and Kathleen Hughes (ed.), England Before the Conquest: Studies in
Primary Sources Presented to Dorothy Whitelock (Cambridge: Cambridge University Press, 1971),
165–81; M. L. Samuels, ‘The Great Scandinavian Belt’, in Laing (ed.), Middle English Dialectology,
106–22; Austin Lane Poole, From Domesday Book to Magna Carta 1087–1216, 2nd edn. (Oxford:
Clarendon Press, 1955), 88 and 290; M. T. Clanchy, From Memory to Written Record: England,
1066–1307, 2nd edn. (Oxford: Blackwell, 1993), 201–2.
54 Classic statements are Uriel Weinreich, William Labov, and Marvin I. Herzog, ‘Empirical
Foundations for a Theory of Language Change’, in W. P. Lehmann and Yakov Malkiel (eds.), Directions
for Historical Linguistics: A Symposium (Austin, TX: University of Texas Press), 95–188; and M. L.
Samuels, Linguistic Evolution, with Special Reference to English (Cambridge: Cambridge University
Press, 1972).
64 Language Anxiety

Smith argues, that produced some Old English phonological variation,

and contact with Norse in England, for Curzan, that may have accelerated
changes in the status of grammatical gender. From these dynamics, accord-
ingly, arose linguistic as well as social transformations that once more reveal
not only the regularity and inevitability of change but also the easily elided
distinction between language change and the social meanings that accrue to it.
When English has come into contact with other languages, that contact
has often taken on the quality of a collision in which the non-English language
(or languages) experienced similar sociolinguistic dynamics, leading to essen-
tially the same curtailment and even eradication of domains and speakers.
Anglophones involved in the expansion of English have perhaps invariably
been the politically and technologically dominant group, for example, al-
though they have not also been (at least initially) numerically superior. In fact,
given their numerical disadvantages and inexperience in many colonial areas,
such as North America, the Caribbean, Africa, India, and the antipodes,
Anglophones often found themselves dependent on indigenes for informa-
tion on local geography, climate, and foodstuVs. Beyond this, in order to
maintain a colonial state, Anglophones depended on indigenous populations
for service, whether forced or at least nominally voluntary. In this way
Anglophones typically not only cultivated a degree of English competence
among some locals, even as they resisted acquiring for themselves full com-
petence in indigenous languages, but also, as in India, thereby restructured
a region’s prevailing linguistic repertoire.
Because of shared characteristics like these, the spread of English has often
been likened to that of Latin in the late Antique and Middle Ages. Through
the expansion of the Roman empire, that is, Latin experienced extensive
contact with a range of languages, including Greek, Oscan, Venetic, Etruscan,
Punic, Gaulish, Germanic, Hispanic, Aramaic, and Thracian. And yet of
all these, with few exceptions Greek was the only second language widely
acquired by speakers of Latin, and this was not because of any interest in
Greek ethnicity but because of that language’s cultural and political associ-
ations.55 For English, the list of contact languages is even longer—perhaps Wve
hundred indigenous languages in North America alone at the time of Euro-
pean settlement—and the resistance and even xenophobia towards them
sometimes as strong. Just as through contact with English most of these
Wve hundred languages are now moribund or extinct, so, too, did the intro-
duction of English to Australia, Canada, and New Zealand in particular lead

55 J. N. Adams, Bilingualism and the Latin Language (Cambridge: Cambridge University Press,
2003), 293.
 A Moveable Speech 65

to signiWcant reduction of indigenous languages in these areas. Even in

England, of course, contact between English and non-English languages
has had adverse eVects on the latter. As in other regions, economics and
technology inXuenced the shift to English, but so, too, did Anglophones’
explicit judgements about the languages they encountered. In his famous
essay in praise of Celtic literature, for example, Matthew Arnold nonetheless
echoed comments also made about Maori and Hawaiian by observing, ‘The
sooner the Welsh language disappears as an instrument of the practical,
political, social life of Wales, the better; the better for England, the better
for Wales itself ’.56
Arnold’s seventh-century countryman Oswald, king of Northumbria, took
a rather more sympathetic attitude towards the Celtic languages. After his
victory over the Welsh king Cadwallon and the reassertion of his own power,
Oswald invited the monk and bishop Aidan to come from the Irish monastery
at Iona and direct the conversion of Northumbria in 635. According to the
Venerable Bede’s account, Oswald served as Aidan’s translator, thereby em-
bodying the sociolinguistic transactions between English and Irish culture:
‘And while the bishop, who was not Xuent in the English language, preached
the Gospel, it was most delightful to see the king himself interpreting the
word of God to his ealdorman and thegns; for he himself had obtained perfect
command of the Irish tongue in his long exile.’57 In situations involving
such extreme diVerences in power and numbers—one inXuential king or
oYcial in the company of many indigenous people—the compulsion to
learn another language can weigh heavily on both Anglophones and non-
Anglophones. Indeed, this same situation arose among early European mis-
sionaries and colonists in North America, though for signiWcantly diVerent
reasons. Colonists taken captive of course had no choice but to learn the
indigenous languages. Yet missionaries, vastly outnumbered by the American
Indians they hoped to convert, had little more Xexibility: like Europeans taken
captive, they needed to learn the indigenous language in order to survive.58
But while speciWc missionaries may initially have learned non-English contact
languages in these cases, the ultimate result of such contact has been virtually
the same in every colonial situation: the expansion of English at the expense
of other languages.

56 Arnold, On the Study of Celtic Literature and On Translating Homer (New York: Macmillan,
1904), 10.
57 Bede, A History of the English Church and People, 147.
58 Karttunen, ‘Interpreters Snatched from the Shore: The Successful and the Others’, in Edward
G. Gray and Norman Fiering (eds.), The Language Encounter in the Americas, 1492–1800: A Collection
of Essays (New York: Berghahn Books, 2000), 221.
66 Language Anxiety

In such contact situations, change and variation as inevitable linguistic

phenomena contrast with the variable social meanings that speakers assign to
them. To the extent that Oswald’s knowledge of Irish suggests his magnanim-
ity—and Bede certainly seems to interpret it that way—it might be treated as
evidence of his power and his ability to reach out with his learning both to
those around him in exile and to his subjects. Yet this same presumptive
magnanimity was a non-issue among early Anglophones on the North
American continent, for it was trade with indigenous peoples that essentially
compelled many to learn the language of the very people they sought to
dominate. Even among European missionaries, who would have shared at
least some of Bede’s sentiments, the social signiWcance of indigenous lan-
guages varied in ways that reveal its contingent nature. While Protestants
sought to learn American Indian languages and to translate Scripture into
them as an extension of the Reformation’s project for individuals to know
God’s word in their own languages, Catholics pursued these same activities in
order to provide a bridge to Latin and the institutional hierarchy it sup-
ported.59 Contact between English and Celtic languages provides yet another
illustration of the contingency of social evaluations. It is commonly asserted
that eighteenth- and nineteenth-century Irish, Gaelic, and (to a lesser extent)
Welsh speakers shifted to English in part from socioeconomic desperation—
from a sense that the inability to speak English was socially and Wnancially
deleterious—and this explanation certainly has much to commend it. Yet
in the face of similarly enormous odds and despite similarly negative eco-
nomic consequences, some indigenous people, such as the North American
Navajoes, have maintained their language so well as to suggest on their part a
kind of indiVerence to or deWance of emerging linguistic and economic
norms. The learning of a second language, whether by a speaker of a socially
dominant or subordinate language, clearly can be a submissive as well as
an empowering gesture—the latter for reasons of economics or cultural
identity. In the same vein, refusal to learn another language can reXect the
dominance of some speakers (e.g., Anglophones) and the subjection or
deWance of others (e.g., Navajoes).
When colonial Anglophones have compelled the acquisition of English
by non-Anglophones, they have done so in ways that emerge from and sustain
this same conditional and contextual categorization of language change
and its meaning, irrespective of the change itself. Lord Macaulay’s educational
minute of 2 February 1835 is a case in point. Building on a century of

59 Edward G. Gray, New World Babel: Languages and Nations in Early America (Princeton, NJ:
Princeton University Press), 1999.
 A Moveable Speech 67

sentiment, this minute established government policy on the status of English

in subcontinental Indian education by directing funds for the instruction
of the native population in English language and literature. Its professed,
non-altruistic objective was the creation of a new Anglophone group of
indigenes who could act as intermediaries between the English and their
colonial subjects: ‘We must at present do our best to form a class who may
be interpreters between us and the millions whom we govern—a class of
persons Indian in blood and colour, but English in taste, in opinion, in morals
and in intellect.’60 In one interpretation, the minute was highly successful,
leading to the use of English in 60% of India’s primary schools by 1882; in
another, by transforming English from a foreign language to a dominant,
institutionalized one within an already multilingual society, the minute
fostered the linguistic slavery of which Gandhi spoke. And in either inter-
pretation, the purpose of inculcating or learning English diVers markedly
from the economic motives prominent in Celtic communities or the religious
ones among Native Americans.
The changing views of English held by the proliWc and inXuential Nigerian
novelist Chinua Achebe oVer another vantage on how the meaning and the
fact of language contact can diverge from one another. In his inXuential 1964
essay ‘The African Writer and the English Language’, published just four years
after Nigeria’s independence, Achebe viewed English (like French) as one of
the factors by which colonial governments forged African countries from
disparate ethnic groups. To use a nativized English was to be part of that
nationalizing process: ‘I feel that the English language will be able to carry the
weight of my experience. But it will have to be a new English, still in full
communion with its ancestral home but altered to suit its new African
surroundings.’61 By the early 1970s and after the Nigerian civil war had
perhaps dispelled any illusions about post-colonial life, Achebe had begun
to speak more forcefully about the cultivation of indigenous African beliefs
and practices and the rejection of those sustained by English. In 1982 he even
helped establish the bilingual journal Uwa Ndi Igbo: a Journal of Igbo Life and
Culture. Achebe’s experience is far from unique, however, for such shifts in
sentiment and the interpretation of linguistic phenomena have become wide-
spread in former British colonies in Africa. In his 1986 Decolonising the Mind,
in which he eloquently details how language shift served larger colonial goals

60 Quoted in Henry Sharp (ed.), Selections from Education Records, i, 1781–1839 (Calcutta: Superin-
tendent Government Printing, 1920), 116. See also Kachru, ‘English in South Asia’, in BurchWeld (ed.),
English in Britain and Overseas: Origins and Development, 500–8.
61 Quoted in Ezenwa Ohaeto, Chinua Achebe: A Biography (Bloomington, IN: Indiana University
Press, 1997), 101. Also see Bailey, Images of English, 170.
68 Language Anxiety

of economic and social dominance in Africa, the Kenyan writer Ngũgı̃ wa

Thiong’o thus expressed a rejection of English similar to Achebe’s. This book,
he says, ‘is my farewell to English as a vehicle for any of my writings. From
now on it is Gı̄kūyū and Kiswahili all the way.’62 Today, this kind of anxiety
over the spread of English as a global language and its disruption of local
linguistic repertoires is not restricted to post-colonial areas. It has become
increasingly strident, indeed, in continental Europe, where a recognition of
the utility of English for business and computer communication accompanies
a sense that such utility comes at the expense of cultural autonomy.63
For English itself, the eVects of its global contact have been equally fraught.
Beginning with predominant Anglophone populations of what Braj Kachru
has called the Inner Circle—the United States, the United Kingdom, Canada,
Australia, and New Zealand—English has spread increasingly as both a
transplanted language (as in India, Nigeria, and South Africa) and a supple-
mentary means of communication (as in South America or eastern Europe).64
Since the early modern period, whether through choice, coercion, or incen-
tive, non-Anglophones have acquired or even shifted to English at spectacular
rates and in the process reconWgured prevailing cultural relations among
education, Wnance, and ethnicity so as to include English alongside indigen-
ous languages.65 English has spread so much, in fact, that while estimates vary,
today, as I noted above, perhaps one-sixth of the world’s 6.5 billion people
know English as a primary or secondary language, and of this group, only
about 350 million reside in the United Kingdom or the United States.
Given its own structured diVerences of social and regional varieties, the
English that has contacted other languages in these many circumstances was
not necessarily or even likely the standard language of dictionaries and
grammar books but rather had its own peculiarities of lexicon and syntax,
just as surely as did the languages it contacted. One structured, variable
language, in other words, contacted other structured, variable languages in
ways that have been themselves structured and variable, and so in some cases
of what is blithely described as contact between English and another language,

62 Wa Thiongo, Decolonising the Mind: The Politics of Language in African Literature (London:
James Currey, 1986), xiv.
63 Suzanne Daley, ‘In Europe Some Fear National Languages are Endangered’, New York Times,
16 April 2001, A1, 10; and John Tagliabue, ‘In Europe, Going Global Means, Alas, English’, New York
Times, 19 May 2002, A15.
64 Kachru, ‘The Second Diaspora of English’, in Machan and Scott (eds.), English in Its Social
Contexts, 230–52.
65 For a useful overview of the spread of English and its variable roles in multilingual societies,
see Viv Edwards, Multilingualism in the English-Speaking World: Pedigree of Nations (Malden, MA:
Blackwell, 2004).
 A Moveable Speech 69

several regional and social variations of English and its contact languages may
have been involved. In New Zealand, for example, the label Maori subsumes
a number of distinct dialects, and in the United States patterns of immigra-
tion aVected some areas more signiWcantly than others. To say that Yiddish
inXuenced American English would certainly be true, but at least initially so
only in urban areas like New York City, where there were signiWcant numbers
of Yiddish speakers in the nineteenth century; the impact of Yiddish on the
English of the plains states would be negligible until print and electronic
media began to disperse words like schlock and schmuck.66 The stylized maps
in the Dictionary of American Regional English, which depict the concentra-
tion of words like cooter (‘freshwater turtle’) in the southeast United States,
make a similar point with visual clarity, for as former slave-holding territory
this is just the part of the country where one would expect to Wnd a word
related to Bambara and Malinké kuta. Unlike schmuck, however, cooter
has not entered general American speech. For all languages in a contact
situation, such Xuctuations in speech communities extend the normal, struc-
tured process of grammatical transformation and variation by group. And
these Xuctuations raise the same issue that appeared in my consideration of
sociolects and dialects: the regular processes of borrowing and grammatical
innovation developing alongside the distinct conditional and contextual
categorization of language change and its meaning.
The status of American English among early British critics provides an apt
example. Although English was Wrmly established in North America early in
the seventeenth century, it is not until the eighteenth century, as England’s
relations with its colonies deteriorated, that speakers came to comment on the
issue of whether American English was the same as British English, a dialect of
it, or its corruption. Perhaps motivated by the same sociolinguistic impulses
as my radio interviewer but reaching the exactly opposite conclusion, many
early commentators stressed the barbarity of American English, emphasizing
how much the language had already changed from that used in England. This
early modern and modern divergence of conclusions rooted in the same
motivations points to just how opportunistic judgements about change and
its meaning can be, and the same is true of comments by colonists like
Benjamin Silliman, who visited England late in the eighteenth century.
Of British Anglophones he observed, ‘they imagine it [United States English]
is a colonial dialect, with a corrupt and barbarous pronunciation, and a

66 Romaine, ‘Contact with Other Languages’, in The Cambridge History of the English Language,
vi, John Algeo (ed.), English in North America (Cambridge: Cambridge University Press, 2001), 181.
70 Language Anxiety

vocabulary, interspersed with strange and unknown terms of transatlantic

manufacture’.67 Yet for all this putative barbarity and distinctiveness, Silliman
was astounded (and perhaps secretly relieved) that he himself was not recog-
nized as an American but rather judged English on the basis of his pronun-
ciation. Drawing on presumably the same evidence, other critics distanced
American speech from British still more—if with less vituperation—by
regarding United States English as essentially a diVerent language instead of
a corruption of the one used in England. According to an anonymous 1807
piece in the Edinburgh Review, the root of Joel Barlow’s language in his epic
poem Columbiad ‘may be English, as that of the Italian is Latin; but the
variations amount already to more than a change of dialect; and really make a
glossary necessary for untravelled readers’.68 Just as opportunistically, while
both American and British analysts could also argue that the language of
the United States was a regionally or socially undiVerentiated variety that
diverged only in its general characteristics from the English of England, the
former might interpret this putative uniformity as a mark of a national
character, the latter as conWrmation of the predominant depravity of Ameri-
can speech.69 In 1832, for instance, Jonathon Boucher prefaced his Glossary by
stating, ‘With little or no dialect, they [i.e., Americans] are peculiarly addicted
to innovation: but such as need not excite our envy, whether we regard their
elegance, or their propriety’.70 And still other, more dispassionate critics,
including David Hume, saw the prosperity and population of North America
as the means for the worldwide spread of English and the increase in the
number of its speakers.
The global dispersion of English in the past two centuries has produced
new linguistic varieties that highlight in still other ways the disjunction
between regular, inevitable change and the value judgements that can be
assigned to it. As with language in the early modern period, contact with
both indigenous languages and the languages of other European explorers
and colonists—such as French, Dutch, Swedish, Spanish, and German—did
not alter natural processes of variation and change, but it did provide input
fundamentally diVerent from that which aVects an isolated Anglophone
community. And it sometimes did so in socially charged situations involving

67 Quoted in Bailey, Images of English, 129. On the barbarism of colonial English, see Allen Walker
Read, ‘Milestones in the Branching of British and American English’, in Milestones in the History of
English in America, ed. Richard W. Bailey (Durham, NC: Duke University Press, 2002), 3–21; see also
Read’s ‘British Recognition of American Speech in the Eighteenth Century’ in the same volume, 37–54.
68 Quoted in Read, ‘Amphi-Atlantic English’, in Milestones in the History of English in America, ed.
Bailey, 60–1.
69 See Bailey, Images of English, 130–57.
70 Quoted in Read, ‘British Recognition’, 42.
 A Moveable Speech 71

diVerences in race, ethnicity, level of education, and economic class. Jamaica,

for example, has a particularly rich and diverse linguistic history, involving
indigenous languages as well as those transplanted from Europe and Africa,
from all of which have developed both distinctive varieties of English and also
creole varieties. As a British colony and then post-colonial society, Jamaica
has equally experienced shifting social dynamics in which language has served
as a means of identifying cultural stability and also various threats to it. One
such perceived threat, in Jamaica and elsewhere in the Anglophone world, has
been racial diVerence. Already in 1858 Major Alan Chambre focused on
variation in English as an indication not only of intrinsically debased speakers
but also of the pernicious inXuence such variation can have on British speech
and, perhaps more importantly, society: ‘There is scarcely a black in a
hundred who speaks pure English, and the white people take no pains to
correct them. Sometimes they even adopt the barbarous idiom of the negro,
thinking to make themselves understood. The consequence is, their pronun-
ciation is abominable, and the rising generation, not withstanding [sic] the
pains taken to educate them, retain the villainous patois of their parents.’71
In Charles Leland’s 1876 Pidgin English Sing-Song, such blending of racial
judgement and linguistic variation acquires a strange, academic kind of
respectability: even though Leland himself was a well-known comic writer
who wrote (among other works) burlesque legends in a mixed German-
English language, his book became the best-known nineteenth-century hand-
book on Chinese English. Focusing less on race and more on the social class of
colonial Anglophones, an 1829 Australian commentator noted that the lan-
guage of new immigrants would do much to sustain the region’s debased
variety of English: ‘bearing in mind that our lowest class brought with it
a peculiar language, and is constantly supplied with fresh corruption, you
will understand why pure English is not, and is not likely to become, the
language of the colony’.72
It is perhaps easy to identify how comments like these Wlter non-linguistic
concerns through change and variation, but the same Wltering has sometimes
occurred more recently in the championing of the plural concept of ‘Eng-
lishes’. Pointing to a variety of structural and contextual features, some critics
of the past quarter century have described the transplanted varieties that
disturbed someone like Chambre not as ineVective attempts to replicate a
standard British or United States variety but rather as fully developed versions

71 Quoted in Barbara Lalla and Jean D’Costa, Language in Exile: Three Hundred Years of Jamaican
Creole (Tuscaloosa, AL: University of Alabama Press, 1990), 97.
72 Quoted in Gunn, ‘Social Contexts in the History of Australian English’, 211.
72 Language Anxiety

of English that respond to the social needs and contexts of those who speak
them. Chicano English, Singapore English, Asian English, Philippine English,
and Chinese English have thus all been identiWed as distinctive varieties
because they all have distinctive phonologies, lexicons, and syntax, because
they all have their own histories, and, in the case of Chicano English, because
it can be spoken as the birth language of individuals who are not bilinguals of
Spanish and English. If Chambre and other early critics Wltered prejudices
about race and class through their assessments of English change and vari-
ation, these kinds of analyses can Wlter a kind of identity politics. By respond-
ing directly to the discussions of earlier generations and championing the
non-linguistic rights and integrity of particular speakers, that is, they, too,
separate language from its sociolinguistic status and become driven by racial,
ethnic, and social concerns. In his valuable study of Chinese English, Kingsley
Bolton has thus noted that for nineteenth-century colonialists, ‘the notion of
‘‘pidgin English’’ was being actively promoted in Britain and the United States
in the conscious attempt to create a racially derogative stereotype of the
Chinese in the minds of the western public’.73 It is of course this stereotype
that the promotion of contemporary Chinese English—or Chicano English or
Singaporean English—just as actively seeks to erase, replacing it with a
diVerent view of the variety and its speakers.
Similar kinds of anxious, social meanings, whether derogatory or vindi-
cating, have been assigned to various kinds of interlanguages that have been
produced through contact between English and non-English languages.
Whether judged as code-switching, pidgins, or simply dialects with extensive
borrowing, Franglais, Denglish, Japlish, and Spanglish (for French-English,
German-English, Japanese-English, and Spanish-English) share certain gram-
matical principles (e.g., English nouns are the most heavily borrowed word
class) and pragmatic eVects: those who speak them often regard the varieties
as creative and contemporary, while traditionalists see them as linguistic
corruptions of both languages and diminishments of cultural identity, per-
haps every bit as threatening as those that indigenous American Indian
languages experienced in their own contact with English. Even as some critics
celebrate varieties like Spanglish as cultural phenomena with their own
history and socially deWning force that is neither entirely English nor entirely
Spanish, for Roberto González Echevarrı́a, ‘Spanglish, the composite lan-
guage of Spanish and English that has crossed over from the street to Hispanic
talk shows and advertising campaigns, poses a grave danger to Hispanic
culture and to the advancement of Hispanics in mainstream America.

73 Bolton, Chinese Englishes: A Sociolinguistic History, 186.

 A Moveable Speech 73

Those who condone and even promote it as a harmless commingling do not

realize that this is hardly a relationship based on equality. Spanglish is an
invasion of Spanish by English.’74
The assignment of social meaning to change and variation (at least in the
modern world) thus appears as consistent and even inevitable as the change
and variation themselves. As intensely as anxieties of one kind or another may
be channeled through what are value-free (to reprise Lass) linguistic phenom-
ena, the anxieties over change are not identical to the change, and in many
cases they betray an innocence about the structure and history of natural
languages like English. From a purely historical perspective, for example,
concerns over interlanguages present several diYculties. First, the interlan-
guage switching in Spanglish or Denglish is not random but structured,
following grammatical principles as regularly as did the First Consonant
Shift or the development of natural gender in English; and its pragmatic
motivations are as clear and consequential as those underlying the formation
of regional and social varieties of English around the globe. Second, intermix-
ing of languages has accounted signiWcantly for the growth and vitality of
English and other living languages. It is, in fact, perhaps the pre-eminent
impetus behind the lexical diVusion that many popular commentators identify
as English’s most distinctive and valuable trait.75 For English, the third person
singular indicative verbal inXection -s as well as the third person plural
pronouns they, them, and their are not native Anglo-Saxon forms but derive
directly from Old Norse, the language spoken by the descendants of Viking
raiders and colonizers in the Midlands and northern parts of England; a kind
of Norslish gave rise to them, as surely as Franglais has produced qu’est-ce que
tu veux pour lunch. Through mixing such as this, which keeps a language vital
and personally meaningful for its speakers, a language like Afrikaans can
remain alive in South Africa, despite its comparatively small number of

74 Echevarrı́a, ‘Is ‘‘Spanglish’’ a Language?’, New York Times, 28 March 1997, A19. For arguments in
support of Spanglish as a mediation of cultural hybridity, see Ed Morales, ‘Introduction: What I’m
Talking about When I Speak in Spanglish, or the Spanglish Manifesto’, in Morales (ed.), Living in
Spanglish: The Search for Latino Identity in America (New York: St Martin’s Press, 2002), 1–29; and Ilan
Stavans, Spanglish: The Making of a New American Language (New York: Rayo, 2003). More generally,
see Lizette Alvarez, ‘It’s the Talk of Nueva York: The Hybrid Called Spanglish’, New York Times, 25
March 1997, A1, 14; and Richard Bernstein, ‘A Snappy Slogan? In German? Don’t Smile. Try English’,
New York Times, 21 December 2004, A4.
75 e.g., Bill Bryson, The Mother Tongue: English & How It Got That Way (New York: W. Morrow,
1990); Howard Richler, A Bawdy Language: How a Second-Rate Language Slept Its Way to the Top
(Toronto: Stoddart, 1999); Kate Burridge, Blooming English: Observations on the Roots, Cultivation and
Hybrids of the English Language (Cambridge: Cambridge University Press, 2004); and Guy Deutscher,
The Unfolding of Language: An Evolutionary Tour of Mankind’s Greatest Invention (New York: Henry
Holt, 2005).
74 Language Anxiety

speakers, its role as only one of eleven oYcial languages, and its historical
associations with apartheid.76
A third diYculty with many contemporary concerns over interlanguages is
that the historical record suggests that such varieties are inevitable and
valuable features of human communication. In the colonial era of North
America, the principles of grammatical and pragmatic regularity motivated
the production of pidgins that were used not only between Europeans
and Native Americans but among the Europeans themselves.77 In the Middle
Ages England’s linguistic repertoire similarly came to include entire linguistic
codes structured like Spanglish or Denglish in the forms of a Latin-English
macaronic variety utilized in sermons and a French-English-Latin interlan-
guage employed in late-medieval business, alongside of which Latin, French,
and English each continued to exist.78 At that time, code-switching among
Latin, French, and English could also achieve signiWcant rhetorical eVects,
much like the style-shifting eVected by modern novelists. This is the case in
William Langland’s fourteenth-century poem Piers Plowman, where switches
from English to Latin both help foreground moments of thematic importance
and portend, perhaps overly optimistically, the eventual emergence of English
as England’s dominant language in all domains of business, education, and
government. Langland emblemizes the spiritual and intellectual faculties
that deWne humans in the character of Anima, for instance, who is also one
of the most linguistically gifted Wgures in the poem. She easily lapses into
extended Latin passages and quotations, such as the following:
Prelates of cristene prouinces sholde preue if þei myZte
Lere hem litlum and litlum et in Iesum Christum Wlium,
Til þei kouþe speke and spelle et in Spiritum sanctum,
Recorden it and rendren it wiþ remissionem peccatorum
Carnis resurreccionem et vitam eternam amen.79

76 Kimberly J. McLarin, ‘To Preserve Afrikaners’ Language, Mixed-Race South Africans Join Fray’,
New York Times, 28 June 1995, A4.
77 See Ives Goddard, ‘The Use of Pidgins and Jargons on the East Coast of North America’, in Gray
and Fiering (eds.), The Language Encounter in the Americas, 61–78.
78 See Siegfried Wenzel, Macaronic Sermons: Bilingualism and Preaching in Late-Medieval England
(Ann Arbor, MI: University of Michigan Press, 1994); and Laura Wright, ‘Macaronic Writing in a
London Archive, 1380–1480’, in Matti Rissanen et al. (eds.), History of Englishes: New Methods and
Interpretations in Historical Linguistics (Berlin: Mouton de Gruyter, 1992), 762–70, and Sources of
London English: Medieval Thames Vocabulary (Oxford: Clarendon, 1996).
79 Langland, Piers Plowman: The B Version, rev. edn., ed. George Kane and E. Talbot Donaldson
(Berkeley and Los Angeles, CA: University of California Press, 1988), XV.609–13. See further Machan,
‘Language Contact in Piers Plowman’, Speculum, 69 (1994), 359–85.
 A Moveable Speech 75

Syntactically, this is a diYcult sentence to describe, combining as it does

quotation, code-switching, and relic English (‘litlum’) in a rhyming, allitera-
tive, and thematic tour de force. As the conclusion to passus XV, these lines
are also in part an introduction to passus XVI and the Tree of Charity, the
most powerful mnemonic and theological icon in Piers Plowman and an
image that recapitulates not only Anima’s teachings but also those of Piers
Plowman in its entirety. It is thus with the ingenuity of an interlanguage—
of Latlish—that Langland pragmatically foregrounds this section as one of
the poem’s climaxes.

Change, shift, death

In order to account for the persistence of variation and change among
European languages, Leonard BloomWeld, in his ground-breaking 1933 book
Language, formulated a ‘principle of density’:
The reason for this intense local diVerentiation is evidently to be sought in the
principle of density. Every speaker is constantly adapting his speech-habits to those
of his interlocutors; he gives up forms he has been using, adopts new ones, and perhaps
oftenest of all, changes the frequency of speech-forms without entirely abandoning any
old ones or accepting any that are really new to him. The inhabitants of a settlement,
village, or town, however, talk much more to each other than to persons who live
elsewhere. When any innovation in the way of speaking spreads over a district, the limit
of this spread is sure to be along some lines of weakness in the network of oral
communication, and these lines of weakness, in so far as they are topographical
lines, are the boundaries between towns, villages, and settlements.80

Implicit in BloomWeld’s principle is the fact that people talk for all sorts of
reasons: to convey information, to ask questions, to deceive, to seduce, to
entice, and, simply, to get along. More generally, as I noted earlier, they speak
to identify themselves socially, in their linguistic forms as well as their topics,
and they use these same criteria to pass judgement about their interlocutors’
age, politics, education, credibility, familiarity, congeniality, honesty, and
ancestry. What BloomWeld points out are the linguistic consequences of
motivations like these. Whether they want to display solidarity with one

80 BloomWeld, Language (New York: Holt, 1933), 476. BloomWeld here lays the groundwork for the
theory of social networks, which has become a powerful and valuable way for understanding how
speakers interact with language and how it changes through these interactions, depending on how
closely tied particular speakers are to particular groups. See further Lesley Milroy, Language and Social
Networks, 2nd edn. (Oxford: Blackwell, 1987); James Milroy, Linguistic Variation and Change: On the
Historical Sociolinguistics of English (Oxford: Blackwell, 1992); and J. K. Chambers, Sociolinguistic
Theory: Linguistic Variation and Its Social SigniWcance (Oxford: Blackwell, 1995), 34–101.
76 Language Anxiety

another or to claim dominance or distinction, speakers who live in close

proximity necessarily use language to accomplish social as well as communi-
cative objectives, and this means that they use it in ways that aVect their
language’s grammar and pragmatics. Language, in turn, necessarily continues
to vary and change in regular if often unpredictable ways.
From this perspective, the processes conventionally understood as language
variation, change, and shift have much in common. All are responses to the
impact of time and geography; all proceed in regular, structured fashions; all
correlate language with group and individual identity; all are temporal phe-
nomena that conWrm and deWne the temporality of language; all have diver-
siWed and enriched English in general and in the idiolects of individual
speakers and literary artists; and all occur with a frequency that projects
inevitability. For reasons like these, in fact, it has become increasingly com-
mon for critics to approach a language not as a discrete system (or set of
systems) but as a continuum; like a creole, then, the varieties of a language
gradually shift, by speaker and form, from one to another, even including the
creole varieties that have developed from it.81
Perhaps not surprisingly, the consequences of such phenomena can pre-
clude easy distinctions among change, variation, and shift and even challenge
the categories themselves. Whatever their diachronic diVerences, for example,
shift and historical change can produce strikingly similar results. From the
viewpoint of structural complexity and arrangement, creole languages,
whether they arose instantaneously or developed gradually from pidgins,
share the characteristics of all natural languages, and so without any historical
background and with only the grammar and corpus of a language, distinc-
tions between interlanguage creoles and historically developed monogenetic
languages are likewise not easily drawn.82 Synchronic variation blends into
diachronic change when contact with an immigrant language leads to the
grammatical restructuring of a dialect, producing features that eventually
spread throughout the language; this was the case with the third person plural
pronouns they, them, and their. And, to return to the biblical examples that
I cited above, even the discreteness of an individual language across time can
sometimes be diYcult to maintain. We may identity the New International
Version of Genesis as the linguistic descendant of the version in the Old
English Heptateuch, but I would venture that for a modern reader the latter

81 For a careful case study that illustrates this very point, see John R. Rickford, Dimensions of a
Creole Continuum: History, Texts & Linguistic Analysis of Guyanese Creole (Stanford, CA: Stanford
University Press, 1987).
82 Salikoko S. Mufwene, The Ecology of Language Evolution (Cambridge: Cambridge University
Press, 2001).
 A Moveable Speech 77

is only marginally more intelligible—or even recognizable as English—than

would be (say) the Wrst Welsh version, initially published in 1588 and begin-
ning ‘Yn y dechreuad y creawdd Duw y nefoedd a’r ddaiar’.83 To that reader,
in fact, Modern English might nearly as easily have developed from Middle
Welsh as from Old English.
What does strongly distinguish language variation and change from shift,
death, and even simple error or inadequate linguistic achievement are the
responses of both the speakers themselves and the succeeding generations.
When viewed and described as natural or internal (i.e., structured grammat-
ical development), linguistic transition across time tends to be regarded as
change that does not raise controversy. Except in very arch ways, speakers
don’t lament—and linguists don’t regret—vowel mergers, morphological
leveling, or syntactic elaboration. Linguistic transition within time, judged
variation, can raise controversy, but it doesn’t do so necessarily. Regional
variation may evoke some social stereotypes but often passes without com-
ment. When it seems to morph into external, diachronic change, however,
as in the cases of Nigerian or Indian English, variation can enfold racial
and cultural issues in such a way that it does become quite controversial,
even though precisely the same process brought in the third person pronouns
without comment. And controversy can likewise surround synchronic socio-
lects, such as African American Vernacular English or Black South African
English.
M. L. Samuels has explained change as follows:
Every change is, at least in its beginnings, present in the variants of the spoken
language; it is the process of continuous selection that ensures its imitation, spread,
and ultimate acceptance into one or more systems. The selection takes place both at
the level of idiolect (and interlocutor) and at the level of system proper; and it may
consist of a preference for a commoner variant rather than a rare one, or vice versa.84

This is as concise and clear an account of the mechanics of change and its
implication in variation as I know. Yet I also think that as a deWnition
Samuels’s account is limited by its own restrictions to the mechanics of
linguistic change. If language truly were a value-free system, unconnected to
the desires, whims, and foibles of its speakers, the deWnition would be
exhaustive. But language—English—isn’t this kind of system. It’s a system
whose speakers do not simply articulate its elements but also assign meaning
(and controversy) to them in their daily lives. In many ways, indeed, it is in

83 Y Bibl Cyssegr-lan: Sef yr Hen Destament a’r Newydd (London: Robert Barker, 1630), sig. alr.
84 Samuels, Linguistic Evolution, 138.
78 Language Anxiety

speakers’ lived sociolinguistic experience that the clarity of technical distinc-

tions among variation, change, corruption, and shift becomes most obscure.
The mere ability of phenomena as structured and recurring as variation and
change to generate controversy points to the signiWcance of this experience.
So, too, does the fact that just as the teaching of language and the categor-
ization of its change and variation are not value-free, neither, it seems, is
identiWcation of its history.
Structural rearrangement has been so consequential as to render Old and
much Middle English unintelligible to most modern Anglophones, whatever
variety they speak. And yet as histories of English are conventionally written,
some varieties—and only some varieties, such as Standard British or United
States English as opposed to South Asian English—can claim a kind of
linguistic lineage from these historical versions. Such a claim both endorses
their primacy above other varieties and obscures how much they, too, have
changed over time. In eVect, history becomes a positive value judgement in a
way that shift is not, since the former implies continuity and integrity, while
the latter evokes all at once cultural rupture and loss from the speakers who
have shifted and (sometimes) only a kind of liberal guilt from those to whose
language the shift has occurred. Diachronic change simply has sociolinguistic
advantages that are lacking to shift and, to a large extent, variation. Indeed,
the classiWcation of new varieties as corruptions or even interlanguages (and
thus entirely new languages) is an interpretive gesture that recalls some
nineteenth-century British responses to American English. Linguistic shift
thereby comes to describe a non-Anglophonic population’s loss of its indi-
genous language but also to quarantine its newly acquired language from
the Anglophone mainstream. And in this way speakers of Chicano English,
for example, may be judged to speak neither Spanish, nor, unless one is
willing to allow the identiWcation of a variety through structure, history,
and so forth, a coherent, recognized variety of English.
There is something particularly insidious about where analyses like these,
however well intentioned, leave the concerned speakers. While speakers of
Maori, Welsh, and Native American languages—and in the United States, to a
much lesser extent, Spanish—have generally adopted English in place of their
historical languages, they have often sought to remain ethnically and cultur-
ally identiWed with their indigenous history and traditions. If linguistic
shift has occurred, this identiWcation necessarily must come through the
medium of another language—English—much as Anglophones can continue
to identify with the Anglo-Saxon and medieval past by reading (or perhaps
only knowing that they can read) Beowulf and the Canterbury Tales through
the medium of Modern English translations. The important diVerence is that
 A Moveable Speech 79

Anglophone populations in Britain, the United States, and elsewhere at least

in Kachru’s Inner Circle can assert a historical connection with Old English,
even if it is no more intelligible to them than would be Maori or Welsh.
And it is just such a connection that can be denied not only to Maori speakers
but also to speakers of varieties like Jamaican English, which may be no
structurally farther from Old English than is Standard United States English
but which, unlike Standard United States English, can be marked as a non-
standard interlanguage.
To the extent that since the Romantic period Anglo-American criticism has
fostered the notion that language and culture are immanent in one another,
there’s an additional diYculty here. If language is an essential determinant of
culture, then when language is lost, as through shift, so might culture be. By
extension, just as a shifted variety might be linguistically excluded from
English as an interlanguage or an inadequate level of achievement, so would
the shifted speakers be excluded from the cultural traditions of their original
language. If speakers of varieties recognized as English—such as Standard
British English—can claim the patrimony of historical style, language, and
literature, speakers of creole varieties or other interlanguages cannot. And if
such speakers are denied the patrimony of English, Romantic notions of
nation and language can equally deny them the patrimony of their own
indigenous languages. The implication would seem to be that there can be
no Irish culture independent of Irish, no Maori culture independent of Maori.
It might thus be argued that beyond the forcible ways in which Anglophones
have sometimes imposed their speech and culture around the world stands
another kind of violence and rupture: this set of interpretive sociolinguistic
strategies that variously advocate protection of indigenous people but also
serve to erase them in the onrush of Anglophone culture. In this way, where
languages go when they go away is into oblivion, and their speakers with them.
The larger, crucial issue here is how sociolinguistic meanings and values
like these are assigned to the phenomena associated with language change and
variation. Whatever structural and historical criteria may be used to diVer-
entiate among change, variation, shift, and corruption, I am maintaining, the
classiWcations themselves are to an extent value judgements—gestures of
sociolinguistic approval, disapproval, or disregard. It is not simply an empir-
ical decision but an exercise of judgement that certain forms of English qualify
as legitimate natural variation, other forms as linguistic corruptions, and
still others as evidence of a new interlanguage. In order to deWne what the
English language is, this judgement rests on a selection of some of the
hundreds of millions of structural forms uttered in the course of a millen-
nium and a half. And this deWnition, in turn, has profound consequences for
80 Language Anxiety

the social standing of various speakers, for the cultural signiWcance of the
varieties they speak, for their ethnic and national identities, and for their
ability to appropriate English literary and linguistic history as their own.
As distinct as linguistic phenomena may be from their interpretation, the
latter has a great deal to do with how the former not only are perceived but
also function in society. At the outset of this chapter I asked where linguistic
forms and languages go when they go away. The answer I would give now is
where they go—into new varieties, languages, or corruptions—and whether
they go away at all or simply are treated as the same forms and languages
in new guise depend a great deal on the partiality of judgements about
language, its variation, and its change. It turns out that linguistic change, as
distinct from shift or error, can sometimes be even less empirically identiWable
than pain.
In asking how sociolinguistic meanings and values are assigned, I am
of course partly asking who assigns them and why they are accepted. Who
judges which linguistic phenomena represent change, which variation, which
shift, and which errors, and how do they do so? Why do some kinds of
structured, irrational variation engender hostility, while other kinds do not?
Why do some languages or forms become other languages or forms, some
continue, and some disappear? Why are some varieties accepted as versions of
English whereas others are not? These questions point to fundamental issues
for understanding how language change and variation have occasioned so
much anxiety, which is itself as contingent as the phenomena it accompanies.
When a variation is judged a corruption, as with putting the cart of mother
before the horse of father, the anxiety can be the result of the assigned values.
But the anxiety can also produce such values, as in the case of British
educational policy in colonial India or American policy with indigenous
peoples, where anxiety over language contact and its eVects ironically led
to additional variation in the form of English acquired by indigenes and thus
additional anxiety. In this kind of contingent balance, it is often the case
that the responsibility for a particular sociolinguistic meaning lies less with
one person or group and more with a complex nexus of institutions, prin-
ciples, and actions. In the three chapters that follow, I look at three such
nexuses and the way they have shaped linguistic anxiety.
 3

Narratives of Change

Stories, confusion, and the Tower of Babel

In his discussion of the formal content and moralizing impulses of historical
narrative, Hayden White observes that the events of daily life, whether
intimate social interaction or epic international struggle, do not transpire in
the well-motivated, balanced, and thematized plot of a story. These kinds of
coherence are imposed by the historian, who selects events, arranges them in
a linear order, and necessarily provides cause-and-eVect explanations. As
the matter of narrative, such coherence is also the product of imagination.
For White, in fact, it is historical narration that renders the phenomena of the
world real and desirable by positing in these phenomena the form and
coherence characteristic of imaginative acts: ‘These events are real not because
they occurred but because, Wrst, they were remembered and, second, they are
capable of Wnding a place in a chronologically ordered sequence.’1 As an act of
imagination, historical narration must also posit its own closure, and the need
for such closure—for comprehensible completion, purpose, and meaning in
historical events—is ultimately moralizing: ‘The demand for closure in the
historical story is a demand, I suggest, for moral meaning, a demand that
sequences of real events be assessed as to their signiWcance as elements of a
moral drama.’2 But the moralizing and authenticating powers of the historian
can be evoked only when the imaginative narrative form assigned to human
experience is contested, only when the reality to be made desirable is con-
tested. ‘In order to qualify as historical,’ White contends, ‘an event must
be susceptible to at least two narrations of its occurrence. Unless at least
two versions of the same set of events can be imagined, there is no reason for
the historian to take upon himself the authority of giving the true account of
what really happened. The authority of the historical narrative is the authority

1 Hayden White, The Content of the Form: Narrative Discourse and Historical Representation
(Baltimore, MD: The Johns Hopkins University Press, 1987), 20.
2 White, The Content of the Form, 21.
82 Language Anxiety

of reality itself; the historical account endows this reality with form and
thereby makes it desirable by the imposition upon its processes of the formal
coherency that only stories possess.’3
While White’s concerns are with historical narratives, their applicability to
myths and stories leads me to use them as a point of departure in this chapter,
where I am concerned with the kinds of stories and myths individuals and
epochs have told to explain language change and variation, speciWcally as it
relates to the history of English. These narratives might have language as
either their subject or their ideological background, and the explanations
might be either direct or implicit, in the sense that the use of change and
variation for stylistic eVect presumes a particular interpretation of them. In
any case, such narratives of language change and variation are powerful ways
to organize and interpret sociolinguistic experience. They make real not
simply the inescapable facts of language use but, like historical narratives,
also the extra-linguistic social meanings that accrue to them. Beyond simply
acknowledging that change and variation take place, such narratives position
these phenomena in relation to historical events, cultural beliefs, and political
expectations. They state chronological connections between language and
social process, and in this way the stories provide the form and coherence
that give change and variation substance, implicitly classifying them (as
change, variation, shift, or error) and providing them with causes, conse-
quences, and moral meanings. What these stories contest, to stay with White’s
analysis, are other accounts of change and variation, which themselves, neces-
sarily, oVer other sociolinguistic meanings and other versions of reality. The
ultimate eVect of many of these stories for English, I will argue, is to use
language change and variation as means to advance particular social positions,
neutralize challenges to these positions, and displace discussion from social
issues to linguistic ones.
In the most general terms, myths can perform many extra-linguistic tasks.
Whether images, ideas, or fully developed stories, they can narrate history,
explain social practice, justify beliefs and prejudices, deWne aspirations, and
give shape to fear. If the Norse myth of the Ragnarok expresses hope for a new,
less violent world in its account of the Wnal, cosmic battle among the gods and
giants, the Judeo-Christian myth of Adam and Eve assures us of divine love
and explains why the world can sometimes be a painful and hostile place.
Myths may achieve their prototypical and pre-eminent shape in a particular
literary work—such as the Norse Völuspá or the Book of Genesis—but they
attain their force not through the inXuence of one work nor through any

3 White, The Content of the Form, 20.

 Narratives of Change 83

verisimilitude to historical events but through the way they respond to

cultural needs and circulate in a variety of formats. The Ragnarok and other
Norse cosmological myths, for example, echo a pervading medieval Norse
anxiety over the passage of time and daily life in the relentlessly harsh
northern world. And while the Norse cosmology is echoed in this anxiety, it
also helps to explain it. Formed from the body of a giant in a primeval
struggle, that is, the world necessarily has an agonistic quality that will be
realized on one last occasion in a climactic battle at the end of time. This
explanatory power is the quality that gives a story or picture mythic status:
not simple recitation of or allusion to the myth, but the fact that the myth can
be applied as an organizing principle to cultural experience, as a paradigm
that makes a variety of experiences real.
For language change and variation, no myth has been more powerful—has
explained more or has had more echoes not simply for the history of English
but for the history of other European languages as well—than the account of
the Tower of Babel, as told in Genesis 11: 1–9. According to this account, some
time after the Great Flood, ‘the whole earth was of one language, and of one
speech’. Concerned that they would ‘be scattered abroad upon the face of the
whole earth’ and thus their memory and achievement obliterated in their
dispersion, the people build a city with a tower ‘whose top may reach unto
heaven’. For His part, the Lord seems motivated by concerns about what a
united people might accomplish, whether in and of itself or as a challenge
to His own authority: ‘And the Lord said, Behold, the people is one, and they
have all one language, and this they begin to do: and now nothing will
be restrained from them, which they have imagined to do.’ In order to thwart
the construction of the Tower and thus the people’s ambitions, the Lord
resolves to ‘confound their language, that they may not understand one
another’s speech’, and as a result the people leave oV building the Tower and
disperse around the world. ‘Therefore’, concludes the account, ‘is the name of
it called Babel; because the Lord did there confound the language of all the
earth: and from thence did the Lord scatter them abroad upon the face of all
the earth.’ As bleak a future as this story seems to foretell, however, it oVers
some hope to the people: this same chapter concludes with the introduction
of Abram, later Abraham, the patriarch of Israel. If the people’s pride leads
to their dispersion, their dispersion in turn leads to the building of a nation.
Perhaps Babylonian in origin, this story represents the earliest, recorded
explanation for what the Middle Ages called the confusio linguarum, the
confusion of tongues, although analogues exist in other traditions. Through
folk etymology of the Hebrew roots, the account interprets Babel to mean
‘dispersion’ rather than the actual ‘gate of God’ (or ‘gates of the gods’) and
84 Language Anxiety

connects the story with Babylon, the city founded by Nimrod according to
Genesis 10: 10. In this way, what might have begun as simply a story of a
tower, perhaps even a tower of piety, was overtaken by a story of language
change.4 While Genesis never addresses the matter, later traditions presumed
that from the one language spoken at the Tower arose the seventy-two
languages that exegesis understood to have been scattered with the people
across the earth—seventy-two, both because it matched the number under-
lying the Septuagint and because it represented the total number of Noah’s
grandsons from Shem, Ham, and Japheth.5
It is not always recognized that the narrative of the Tower of Babel is in fact
the second mythic explanation for language change and variation in Genesis.
The previous chapter, Genesis 10, relates the multiplication of Noah’s des-
cendants (including Nimrod) in a Table of Nations, and in so doing refers
with parallel phrasing to the peoples and languages associated with each of his
sons. At the end of the chapter, for example, Genesis 10: 31 reads, ‘These are
the sons of Shem, after their families, after their tongues, in their lands, after
their nations’ (also 10: 5 and 10: 20). It is possible to reconcile the two
explanations in the way St Augustine does, by regarding the references in
chapter 10 as proleptic of the general sociolinguistic groupings that would
develop as a result of the events at the Tower.6 There was one language, in
other words, but when change and variation began, they did so along the
patterns foreshadowed in Genesis 10, with the descendants of Shem tending to
scatter in one direction and speak one kind of language, those of Ham in
another, and those of Japheth in a third. It’s also possible that the two
explanations are just that—two diVerent accounts of change and variation,
much as the four Gospels oVer sometimes inconsistent views of Jesus’s life.
More signiWcant than the divergence in narrative lines between Genesis 10
and 11 is the divergence in the stories’ implications for language, for this
divergence produces a dialogue that dominates subsequent narratives and
representations of language. It is this kind of dialogue, following White’s
analysis, that constructs the form, coherence, and reality of change and
variation. According to the account in Genesis 10, change and variation
exist as ordinary features of ordinary human experience. They help deWne

4 Arno Borst, Der Turmbau von Babel: Geschichte der Meinungen über Ursprung und Vielfalt der
Sprachen und Völker, 4 vols in 6 (Stuttgart: Anton Hiersemann, 1957–63), i, 116–17. Also see OED, s. v.
Babel. For accounts of Greek, Kenyan, and Aboriginal analogues, see Richard Bailey, Images of English:
A Cultural History of the Language (Ann Arbor, MI: University of Michigan Press, 1991), 93–4.
5 Borst, Der Turmbau von Babel, ii, 480.
6 St Augustine, The City of God against the Pagans, ed. and trans. R. W. Dyson (Cambridge:
Cambridge University Press, 1998), 702.
 Narratives of Change 85

who we are as human beings, both by reXecting our history and culture and
by taking place as naturally as does the succession of generations; children are
born, and languages change. In Genesis 11, conversely, change occurs as divine
punishment. For St Augustine, the Tower, which he allows may be a meta-
phorical representation of an entire city, was built as ‘an aVront to God’ and
symbolizes Nimrod’s ‘ungodly pride’. ‘Because the power of a ruler lies in his
tongue,’ Augustine continues, ‘it was there that Nimrod’s pride was con-
demned, so that he who refused to understand and obey God’s bidding was
himself not understood when he gave his bidding to men.’ Having scattered
human language, God leaves the Tower standing, and in this way He reconW-
gures its symbolic signiWcance. Observes Augustine, ‘The safe and true way to
heaven is built by humility, which lifts the heart up to the Lord, not against
the Lord’.7 A personal tower of modesty and submission thus replaces Nim-
rod’s literal tower of pride. Elsewhere, Augustine ampliWes the moral import
of the story by claiming that ‘certain proud’ men built the Tower, ostensibly so
that they might not be destroyed by another Xood: ‘For they had heard and
recalled that all iniquity had been destroyed by the Flood.’8 What they had not
heard, evidently, was God’s pledge not to destroy the world by Xood a second
time, and so their hubris in resistance to God is compounded by their
ignorance of His Testament and the folly of their belief that were God to
send another Xood, they could build a tower high enough to rise above the
water.
If the injunction against pride provides the moral axis of the Tower story,
however, its emotional axis is the all-too-human anxiety over being left
alone in the world, particularly in the pre-modern world with its limited
networks of social support. Indeed, this threat of being scattered would
seem to appeal to an almost primeval fear: the action the people speciWcally
dread and build the Tower to forestall is the very action with which the Lord
punishes them. As powerful as it is, such fear might well better account for the
persistence and emotional power of the story over time than does its explan-
ation of language change.9 Already in this etiological myth of language
change, then, anxiety over language thus stands in for other, greater anxieties:
over the forced abandonment of a homeland that is known and familiar
for one that is not, and over forced separation from an individual’s family
and friends.

7 St Augustine, The City of God against the Pagans, 703.

8 St Augustine, Tractates on the Gospel of John 1–10, trans. John W. Rettig, i of The Fathers of the
Church: A New Translation (Washington, DC: The Catholic University of America Press, 1988), 138–9.
9 Cf. Bailey’s comment: ‘This terror of being ‘‘scattered’’ is a key to understanding this central myth
of Western linguistic culture’ (Images of English, 93).
86 Language Anxiety

Within this interpretive framework, whenever language changes, it reminds

us all at once of our pride, isolation, and punishment, just as surely as the
sweat of toil and the pains of childbirth remind us of Adam and Eve’s fall
in the Garden of Eden and our own mortality. By extension, if before the
fall humans were without sin or death, before Babel they spoke one language,
which allowed such complete and perfect communication as to enable them
to build a tower that captured divine attention. For St Augustine, additionally,
such complete communication is possible again only through Christ and His
Church: ‘If pride created diVerences of tongues, Christ’s humility has joined
the diVerences of tongues together. Now what that tower had dispersed, the
Church binds together. From one tongue came many; do not be amazed,
pride did this. From many tongues comes one; do not be amazed, love did
this.’10 Total, invariable communication is thus mythologized at once as an
original attribute of humans and a quality that elevates us above human
limitations to the point where, to echo Tennyson, we might strive with
God. And hence the signiWcance of God’s choice to scatter language and
humanity rather than simply destroy the Tower: another Tower might be
built, but not, evidently, without a shared language.
Peter Bruegel the Elder’s famous depiction of the Tower of Babel strikingly
accommodates both of these narratives. Painted with oil on oak in 1563 and
approximately 45 inches by 61 inches, the picture shows an aristocratic
Nimrod surrounded by his councillors and facing several prostrate workers.11
Other craftsmen shape and lever stone blocks. In the distance to the left
appears a village with houses too numerous too count, while up the right
side of the painting a river extends into the distance; alongside the Tower
are docked several large ships, from which, presumably, supplies and workers
have been unloaded. The Tower dominates the central space of the painting,
stretching from Nimrod, who through foreshortening stands at the Tower’s
base, through the clouds in the picture’s upper margin. A ziggurat partly
inspired by the Colosseum, which Bruegel had seen in Rome, this Tower
climbs upward by levels graduated so slightly that, if the enormous propor-
tions were ever completed, it might indeed reach unto Heaven. Each level
develops in the same way: tall archways with two windows above them that are
grouped in pairs by buttresses and three additional windows spanning the
top of the buttressed space.

10 St Augustine, Tractates on the Gospel of John 1–10, 139.

11 In or around 1563 Bruegel painted another representation of the Tower of Babel, one less
developed and considerably smaller (approximately 24 inches by 30 inches) than the better-known
version I discuss here.
 Narratives of Change 87

What is striking about Bruegel’s picture from a sociolinguistic perspective

is its ambivalence about the origins of language change and variation. On
one hand, in its scale and complexity, Bruegel’s Tower exudes the hubris to
which Genesis 11 alludes. In comparison to the simplicity of the depicted
construction procedures, it additionally testiWes to what can be accomplished
through human cooperation (and perhaps coercion) of a sort that the med-
ium of a shared language allows. On the other hand, Bruegel’s Tower allows
for the realization of variation and change that Genesis 10 describes and
Genesis 11 initiates. With the innumerable houses stretching into the distance
behind it, the Tower’s many rooms would seem superXuous, prepared for a
population that either is already housed or does not yet exist. More provoca-
tively, the repeated design of individual chambers, each with exterior access,
images the separation and isolation of a world already linguistically divided.
The account of change and variation expressed by the Tower of Babel is
the better known of the two versions in Genesis. It certainly better lends itself
to visual depiction, so much so that Bruegel’s painting is perhaps best seen as
part of a tradition of early modern representations. Between about 1550 and
the beginning of the seventeenth century, in fact, 140 depictions of the Tower
were published.12 Yet the other version has had explanatory power as well. As
the expansion of the known world made it more and more diYcult to adhere
to the seventy-two languages predicated on Genesis, and as developments in
linguistics made change and variation more empirically comprehensible, so it
became more diYcult to posit one parent language for all the world. Colon-
ization of North America alone presented Anglophones with evidence of
hundreds of languages structurally dissimilar not only to European languages
but also to one another. In the early modern period, indeed, recognition of
the diversity of human speech proceeds at an almost logarithmic rate: by 1660
Andreas Müller had identiWed ninety languages, by 1787 Lorenzo Hervás y
Panduro over three hundred, and by 1806 Adelung and Vater Wve hundred.13
While this early modern identiWcation of so many languages provided
empirical arguments against Genesis 11, Enlightenment thought strengthened
these arguments with a theoretical underpinning that had much the same
eVect. The arbitrary connection between words and things that was advocated
by John Locke and other early modern thinkers, for instance, rendered
language artiWcial and therefore a polygenetic rather than monogenetic phe-
nomenon.14 Even if early modern Anglophones regarded indigenous peoples
12 Vivien Law, The History of Linguistics in Europe from Plato to 1600 (Cambridge: Cambridge
University Press, 2003), 262.
13 Law, The History of Linguistics in Europe from Plato to 1600, 218.
14 Umberto Eco, The Search for the Perfect Language (Oxford: Blackwell, 1995), 73–116. Also see
Chapter 5.
88 Language Anxiety

in North America and other sites of Anglophone colonization as savages,

moreover, they increasingly came to grant that all languages primarily reXected
their speakers’ shared cultural experience.15 And this, too, is as much as to
accept both the inevitability of variation and the absence of descent from a
single linguistic fountainhead.
The greatest inXuence of the Table of Nations, however, may be indirect—
as the competing version that motivates echoes of the Tower of Babel and that
makes its account, and the sociolinguistic anxieties it projects, real. Behind
every insistence on the decline of language, punctuation, and pronunciation
uneasily lurks the anxiety that such transformations may be natural and
inevitable. Behind every claim that language and variation are unnatural lies
the recognition that this might not be true. Between them, the narrations of
Genesis 10 and 11 thus dialogically compete over the causes of variation and
change, the mechanics of their development, the consequences of their exist-
ence, and their moral implications for life on earth.
Earlier I noted that mythic power depends not simply on the historical
accuracy of stories. It may partly arise in this fashion, although since myths
can also be outright fabrications, equally important are discursive traditions
that include explicit reference to a story as part of an argument. Yet the
greatest power of a myth lies in its responsiveness to lived experience, in the
way it can be used to organize and explain unfolding events that would seem
to have no chronological or causative connection to it. In the case of the
Tower of Babel and the alternative account of Genesis, the history of English
and other European vernaculars aVords multiple examples of all kinds of such
mythic aYrmation. Since the earliest Christian era, speciWc allusions to the
stories and their presumed historical accuracy have occurred in a variety of
contexts, as have sociolinguistic developments implicitly underwritten by
the myth’s account of the origins of variation and change. What many of
these narratives share, accordingly, is the implication that language change
and variation are wrong—even unnatural and immoral—and can thereby
serve as justiWcations for all the sociolinguistic practices that proceed
from this judgement.
The predominant approach of early Christian exegetes takes a very literal
reading of the story. Late in his fourteenth-century Confessio Amantis, for
example, John Gower follows Augustine in identifying sin as the cause of
Nimrod’s pride and linguistic division of humankind as its consequence:

15 For a discussion of how the North American experience aVected traditional thinking about
languages and language development, see Edward G. Gray, New World Babel: Languages and Nations in
Early America (Princeton, NJ: Princeton University Press, 1999).
 Narratives of Change 89

And over that thurgh Senne it com

That Nembrot such emprise nom,
Whan he the Tour Babel on heihte
Let make, as he that wolde feihte
Ayein the hihe goddes myht,
Wherof divided anon ryht
Was the langage in such entente,
Ther wiste no what other mente,
So that thei myhten noght procede.16

By this common medieval reading, Nimrod’s pride is archetypal, since it is sin

that produces division of all kinds:
For Senne of his condicioun
Is moder of divisioun
And tokne whan the world schal faile.17

Three centuries later, in Paradise Lost, John Milton shared and developed the
medieval moralizing impulses in the story, if not Gower’s religious sentiments
more generally. To Milton, Nimrod is driven by ‘ambition’, and the very
material he uses to build the Tower—‘a black bituminous gurge’ that boils
up from ‘the mouth of Hell’—morally orients his ambition as well as the
Tower itself. Here, however, Nimrod appears less as someone who might
challenge God and more as the object of scorn, and in this way the Tower
emerges as truly the act of folly St Augustine understood it to be. It is in
‘derision’ that God divides human language into a ‘hideous gabble’ and
‘hubbub strange’, and once language has been divided, ‘Great laughter was
in Heaven’. The Tower itself is, very simply, ‘Ridiculous’.18
Consideration of just what ‘the langage’ was that preceded the Tower’s
destruction exercised many medieval and early modern commentators, and
it, too, had a distinctly moral cast. According to St Augustine, in the pre-
lapsarian world of the Garden of Eden, humans and God managed both
external, vocal speech and also a kind of silent, inner communication. ‘He
speaks’, says Augustine of God, ‘in His own ineVable way. His speech is
explained to us in our fashion; but God’s speech is indeed more sublime
than ours. It precedes His action as the immutable reasons of the action itself,
and it has no audible and transient sound, but it has a power which endures

16 The English Works of John Gower, ed. G. C. Macaulay, EETS es 81 (London: Oxford University
Press, 1900), Prologue, ll. 1017–25.
17 Prologue, ll. 1029–31.
18 Milton, Paradise Lost, 12.38–62, in The Poems of John Milton, ed. Roy Flannagan (Boston, MA:
Houghton MiZin, 1998).
90 Language Anxiety

for eternity and operates in time.’19 In Eden and before the Tower, commu-
nication was direct and unmediated, uncorrupted by any intermediate signs
or lapses in interpretation. Some of this quality of complete and perfect
communication was lost with the fall, when humans, in a sense, fell from
direct knowledge of God and into Xawed and indirect communicative chan-
nels, including writing. To some exegetes, indeed, writing could therefore
be regarded as the earliest indication of the hazards of change and variation.
By these accounts, well before Nimrod built his tower, had Adam not sinned,
we would not have needed writing, the Bible, or exegesis to understand one
another or God with complete comprehension.20
While some commentators regarded the original Adamic language as lost
either in the Garden of Eden or at the Tower of Babel, prior to the seventeenth
and eighteenth centuries identiWcation with Hebrew was so typical as to be
commonplace. As Isidore of Seville simply put the matter in his inXuential
Etymologies, ‘the Hebrew language is the mother of all languages and litera-
tures’.21 This connection may have strained linguistic explanation, inasmuch
as Hebrew is neither the parent nor even genetic relative of any indigenous
European language—not to mention South African, Asian, South American,
North American, and PaciWc Rim languages. But a genealogy that postulated
an intrinsic, genetic connection between Hebrew and the European languages
did respond to historical and theological imperatives: the medieval need
to connect Europe to the culture and language of the Old Testament.
Such a connection served ideological purposes as well, implicating the
development of languages and all that entailed culturally within the develop-
ment of Christianity. To the twelfth-century English exegete John of Salisbury,
for example, pre-Babel Hebrew wasn’t simply originary; this original language
was also ‘more natural than the others, having been, so to speak, taught by
nature herself ’.22 By this reasoning, a fall from Hebrew in particular was
simultaneously a fall from God and nature. In De Vulgari Eloquentia Dante
similarly invested Hebrew, the putative original language, with an innately
moral quality, noting that only it was worthy enough for Christ to utter. ‘In
this form of speech Adam spoke,’ Dante observes, ‘in this form of speech all
his descendants spoke until the building of the Tower of Babel . . . and this

19 St Augustine, The City of God against the Pagans, 705.

20 Eric Jager, The Tempter’s Voice: Language and the Fall in Medieval Literature (Ithaca, NY: Cornell
University Press, 1993), 52–97.
21 Isidore of Seville, Etymologiarum sive Originum Libri XX, ed. W. M. Lindsay, 2nd edn. (Oxford:
Clarendon Press, 1962), i, 3.4.
22 Quoted in Jesse M. Gellrich, The Idea of the Book in the Middle Ages: Language Theory, Mythology,
and Fiction (Ithaca, NY: Cornell University Press, 1985), 99.
 Narratives of Change 91

form of speech was inherited by the sons of Heber, who after him were called
Hebrews.’ Dante goes on to imagine the confusio linguarum as a felix culpa,
however, for if it shattered pre-Tower linguistic uniformity, it also allowed the
history of faith to become written in the history of languages. ‘With them alone’,
he observes of the Hebrews and their continued use of the original language,
‘did it remain after the confusion, in order that our Redeemer (who was, as to his
humanity, to spring from them) might use, not the language of confusion, but of
grace’.23 That apparently simple formulations about the origins of language
have an ideological edge likewise appears in St Augustine’s discussion of trans-
lation in De Doctrina Christiana. There, stressing that a knowledge of languages
is a necessary pre-requisite to any scriptural analysis, he explains how dangerous
mistranslation can be, not simply to the reading of the Bible but, in light of
its role as the foundational text for Christianity, to the book’s larger implica-
tions. As he had said in his Confessions, while divine speech is eternal and
immaterial, human speech is temporal, the result and source of further errors.24
To read Scripture and thereby participate in the understanding and even
governance of the medieval world is to resist the errors in communication that
have characterized human experience since the fall and that now include errors
in written transmission: ‘for the attention of those who wish to know the divine
scripture must Wrst focus on the task of correcting the manuscripts, so that
uncorrected ones give place to corrected ones, assuming that they belong to
the same class of translation’.25 For St Augustine, then, sin led to Babel, which
led to the confusion of tongues, which led to failed communication in general,
which led to failed transmission and understanding of the Bible in particular,
which leads to the forces that challenge medieval Christian society. Through
this inXuential analysis, resistance to variation and change becomes resistance
to social dissolution and, simultaneously, conservation of the ecclesiastical
and political institutions that counter it.
A similar ideological edge informs the fourteenth-century comments of
Nicholas Oresme, who proceeds from recognition that a diversity of languages
works against civilian and political conversation to the aYrmation that a
people should never accept a king from another nation who does not speak
their language.26 And the Babel story serves such political purposes even more

23 Dante, De Vulgari Eloquentia, in Alex Preminger et al., ed. and trans., Classical and Medieval
Literary Criticism: Translations and Interpretations (New York: Frederick Ungar, 1974), 417.
24 Confessions, trans. R. S. Pine-CoYn (Harmondsworth: Penguin, 1961), 256–60.
25 St Augustine, De Doctrina Christiana, ed. and trans. R. P. H. Green (Oxford: Clarendon Press,
1995), 81.
26 See Serge Lusignan, Parler Vulgairement: les intellectuels et la langue française aux
xiiie et xiv e siècles, 2nd edn. (Paris: J. Vrin, 1987), 109.
92 Language Anxiety

so in the sometimes unconventional arguments of the early modern French

scholar and thinker Guillaume Postel (1510–81). His 1538 De Originibus seu
de Hebraicae Linguae et Gentis Antiquitate maintains not only that Hebrew
was the original language, spoken by Noah and his sons, but that Arabic,
Chaldean, Hindi, and even Greek developed from it. In his Linguarum
Duodecim Characteribus DiVerentium Alphabetum Introductio, published in
the same year, Postel further argues that in fact all languages share a common
origin, and from there, Umberto Eco notes, ‘he went on to advance the
project of a return to Hebrew as the instrument for the peaceable fusion of
the peoples of diVering races’.27 Within the mythological tradition of Babel, a
unity of language would correspond to the unity of the world and God, would
foster the unity of humanity, and would thereby demonstrate the transcend-
ent truth of Christianity to believers as well as non-believers. The histories of
language, the Church, and politics once more become homologous.
In these ways, then, Christian exegesis mythologizes language change and
variation as unnatural and sinful conditions; they may persist, in fact must do
so, but they persist both as conWrmations of a world that fell in Eden, at Babel,
and ever afterwards and as catalysts of the confusion towards which humans
seem drawn by their fallen nature. In a telling comparison that suggests
just how far down the chain of being linguistic dispersion has cast humans,
St Augustine claims that ‘if two men, each ignorant of the other’s language,
meet, and are compelled by some necessity not to pass on but to remain
with one another, it is easier for dumb animals, even of diVerent kinds, to
associate together than these men, even though both are human beings’.28
While the story of Pentecost oVers what is in some ways a competing
biblical account of change and variation, it shares with the Babel myth this
notion that such linguistic phenomena are unnatural. At Pentecost, when the
apostles were gathered together, ‘there appeared unto them cloven tongues
like as of Wre, and it sat upon each of them. And they were all Wlled with the
Holy Ghost, and began to speak with other tongues, as the Spirit gave
them utterance’. When they go out to speak, they encounter people ‘of every
nation under heaven’, who ‘were confounded, because that every man heard
them speak in his own language. And they were all amazed and marvelled,
saying one to another, Behold, are not all these which speak Galilaeans? And
how hear we every man in our own tongue, wherein we were born?’29 Here,
change and variation in the miraculous form of glossolalia may represent

27 Eco, The Search for the Perfect Language, 76. See 73–116 in general for discussion of the putative
status of Hebrew as the original language.
28 St Augustine, The City of God against the Pagans, 928.
29 Acts, 2: 1–12.
 Narratives of Change 93

divine gifts for faith rather than punishments for pride, but like the confusion
of tongues at Babel they originate outside of ordinary human experience,
they transform this experience, and they hold out the possibility of perfect,
unmediated communication. For St Augustine’s teleological historiography,
as a gesture of Church uniWcation Pentecost stands simply as Babel’s fulWl-
ment: ‘The apostles were sent to the nations: if to the nations, to all tongues.
The Holy Spirit, parted in tongues, united in the dove, signiWed this. On this
side tongues are parted; on that the dove joins them together.’30 Three
centuries later, the Venerable Bede invested even more spiritual potential in
a putative return to divine, unmediated communication. For him, not only
does Pentecost restore what Babel cast apart, but also it is this restoration that
enables humans to begin to recover the nearly God-like wisdom they were
to have had at creation: ‘The humility of the Church brings together again
the unity of languages that the pride of Babylon had scattered; spiritually,
however, the diversity of languages signiWes the gifts of holy favors. Indeed,
the Holy Spirit is not improperly therefore understood to have given the
Wrst gift of languages to men, for whom human wisdom is learned and taught
from without, in that it can be shown how easily men can be made wise
through the wisdom of God, which is within them.’31
Although the story of Babel oVers an overt reference point for discussions
of language in the Middle Ages and later, it is perhaps even more signiWcant
that its conceptualization of change and variation can underwrite and organ-
ize historical experience so as to demonstrate the very thing that gives the
story its mythic power. First ancient Greece and then Rome, though decidedly
not Christian, eVectively strove towards a pre-Babel condition in their im-
perial aspirations and their linguistic practices. To the Greeks, anyone
who didn’t speak Greek was æÆæ —foreign, with speech sounding like
ba-ba-ba and behavior that was crude and uncultivated. Many of these same
connotations appeared in the Latin borrowing barbarus, epitomizing Rome’s
own indiVerence and condescension to those outside the empire who shared
neither their language nor their customs. The world may necessarily have
been multilingual, but conceptually, Greece and Rome were not.
Lacking imperial ambitions of its own, the early Christian church con-
comitantly demonstrated little interested in the unity or diversity of speech,
but by the eleventh and twelfth centuries it, too, had begun overtly to conduct
itself in ways consistent with a Babel model of change and variation.32

30 St Augustine, Tractates on the Gospel of John 1–10, 138.

31 Bede, Expositio Actuum Apostolorum et Retractatio, ed. M. W. Laistner (Cambridge, MA: Medi-
eval Academy of America, 1939), 16.
32 Borst, Der Turmbau von Babel, ii, 366.
94 Language Anxiety

This shift in outlook owes to several developments, in each of which linguistic

attitudes supported political concerns. First, by that time Christianity had in
fact begun to exercise pan-European expectations, seeking to establish hier-
archical institutions that were networked to carry out its administration
and to cultivate shared practices and beliefs as a way of deWning, maintaining,
and expanding the faith. Second, such an exercise necessarily required the
assistance of writing. This period witnessed, indeed, the proliferation of
Bibles, commentaries, and other heuristic works, all of them written in
Latin and all of them helping to deWne a limited class of literati who alone
were able to participate fully in political and ecclesiastical administration.
And third, it is around this same time, not coincidentally, that various
vernacular traditions underwent signiWcant and diverse expansion: critical
grammatical works like the Icelandic First Grammatical Treatise and Sir
Raimon Vidal’s Provençal Razos de trobar date to the period, and to the
following centuries stylistic ones like the Icelandic Prose Edda or the Welsh
Bardic Grammar and literary traditions like Provençal lyrics and the Icelandic
sagas. In such a context, the eZorescence of painted representations of the
Tower, particularly after the eleventh century,33 aYrms concern over the
reality and development of multilingualism and belief in the importance of
a mythic common language for the functioning of the Christian world.
Outside of iconography and myth, this same anxiety manifested itself in
an insistence on Latin as the language of government, the church, and
political power, on the conduct of all kinds of business through writing,
and on the eVective exclusions of some groups (such as women) from both
kinds of sociolinguistic activity. In eVect, such procedures fashioned a world
that, while perhaps more eYcient than the one it replaced, ultimately endea-
vored to resist variation and change and instead to reproduce the imagined
communicative achievement of the time before Babel.
For an emblem of this world, I return to a passage I’ve already cited
(Paradise Lost, 12.56–61) in which Milton uses hubbub to describe the ‘hideous
gabble’ of post-Babel speech:
Forthwith a hideous gabble rises loud
Among the Builders; each to other calls
Not understood, till hoarse, and all in rage,
As mockt they storm; great laughter was in Heav’n
And looking down, to see the hubbub strange
And hear the din.

33 Eco, The Search for the Perfect Language, 17.

 Narratives of Change 95

Only a century old at the time Paradise Lost was written, hubbub is an
onomatopoeic word that described, initially, a ‘confused noise of a multitude
shouting or yelling; esp. the confused shouting of a battle-cry or ‘‘hue and
cry’’ by wild or savage races’. Its Wrst recorded use, according to the
OED, illustrates a twofold linguistic anxiety, comparing as it does the ‘con-
fused noise’ of African languages to that of Irish: the Ichthiophagi of Africa,
William Watreman observes in his 1555 translation of Boemus’s Fardle of
facions conteining the aunciente maners of AVrike and Asia, ‘Xocke together
to go drincke . . . shouting as they go with an yrishe whobub’.34 Beyond
this striking contemporary anxiety, which is characteristic of the early mod-
ern global expansion of English, Milton’s word choice joins the moral
and institutional imperatives of Christianity to the xenophobia of Greece
and Rome. The wicked, in other words, are at once damned and Æææ Ø.
That a simple word like hubbub can encapsulate such capacious traditions
says much about the acceptance and naturalization of the Babel account
of change and variation.

International language and perfect communication

Later, in a letter to Catherine the Great dated 26 May 1767, Voltaire claimed to
have heard a woman at Versailles observe, ‘What a great shame that the bother
at the tower of Babel should have got language all mixed up; but for that,
everyone would always have spoken French’.35 As stereotypical as such a
comment from the age of the ancien régime might seem, there is more than
hubris or xenophobia here. As I noted above, global European expansion in
the early modern period made it increasingly diYcult to adhere to the biblical
tradition of a monogenetic language fragmented into seventy-two other
languages at the Tower of Babel; it became obvious that there were (and
are) many more languages than seventy-two, and also that many of the ones
newly encountered in North America, Africa, and Asia could have had little
direct connection to European languages, not to mention Hebrew. At the
same time, Enlightenment belief in the power of reason and the intrinsic
merit of human beings rendered an unnatural, punitive origin for language
change and variation more diYcult to accept intellectually. Even in these
circumstances, however, the Tower model for change and variation has
remained both adaptable and persistent. That the events of Babel could

34 OED, s. v., hubbub. Cf. Seth Lerer’s discussion of hubbub in Inventing English: A Portable History
of the Language (New York: Columbia University Press, 2007), 149–50.
35 Quoted in David Crystal, Language Death (Cambridge: Cambridge University Press, 2000), 28.
96 Language Anxiety

have fashioned this kind of general template for organizing linguistic experi-
ence, long after spiritual much less literal belief in them has ceased to be
commonplace, suggests again that their appeal is not simply linguistic but
rather to some profoundly felt anxiety about separation.
Already in the Paradiso Dante had come to question Hebrew’s originary
status by having Adam observe,
The tongue I spoke had vanished utterly,
long before Nimrod’s people turned their hands
To the work beyond their capability.

God’s linguistic gift as Dante here conceives it, then, is not a speciWc language
but the ability to speak—something like universal grammar—and in this way
Dante redeems the status of all European vernaculars and implicitly invokes
Genesis 10 as the competing narrative that dialogically has helped fashion
the form and coherence of language change:
That man should speak is Nature’s own behest;
but that you speak in this way or that
nature lets you decide as you think best.36

The French aristocrat’s substitution of her own national language for the
Hebrew of tradition oVers still another use of the Babel story, not by following
Dante’s substitution of critical faculties for a natural language but by appro-
priating the story for her own nativist purposes. Her claim reXects a com-
promise between biblical tradition and Enlightenment thinking with which
many other early modern thinkers were comfortable.
Over a century earlier the Englishman Richard Verstegan had noted that
Goropius Becanus had claimed it was in fact Teutonic that was ‘the moste
ancient language of the world; yea the same that Adam spake in Paradise’.37
Still others traced to Eden, or identiWed as the original language, Spanish,
German, Swedish, Hungarian, and Celtic.38 More modest proposals (which
were also more reminiscent of Dante’s views) were advanced by critics like the
anonymous author of the 1689 etymological dictionary Gazophylacium Angli-
canum, who merely traced their own language to the linguistic big bang of
Babel. Before Babel, the author observes, ‘all the then World speke one and the
same Dialect, supposed to be Hebrew’, but Babel gave rise to many languages,
‘of which, the primitive language of the Nation was one, and, had it not been
corrupted, perhaps as good and intelligible as the best; but being so alter’d by

36 Dante, The Paradiso, trans. John Ciardi (New York: Mentor, 1970), XXVI.124–6, 130–2.
37 Verstegan, A Restitution of Decayed Intelligence (1605; rpt. Ilkley: Scolar Press, 1976), 190.
38 Eco, The Search for the Perfect Language, 95–103.
 Narratives of Change 97

the aforesaid Accidents, it has quite lost its primitive Glory, as well as that of
the French, and other Nations’.39 For his part, Verstegan was skeptical of such
arguments, though he allows that if Teutonic is not the Wrst post-Babel
language, ‘it cannot bee denied to bee one of the moste ancientest of the
world’.40 In this way, the myth of Babel came to accommodate the competing
narrative of Genesis 10 and, along with it, language variation and change,
English nationalism, and the early modern insistence on distancing the
present, positively or negatively, from what had come before.
Perfect languages oVer another example of the way in which sentiments
reXected in the Babel story have framed and made real variation and change.
Rooted in kabbalistic traditions as well as in Christian exegesis on the Garden
of Eden and the Tower, perfect languages are artiWcial constructions that
attempt to transcend the vagaries of communication produced by time and
space. Whether numerical, symbolic, or lexical, they seek to oVer a medium of
communication that is transparent, unmediated, and reXective of constant
universal truths, and in this way they foster community and intellectual
achievement. They seek, in short, to reproduce the unchanging language
from before Babel. Flourishing in the sixteenth and seventeenth centuries in
particular, the perfect languages developed by scholars like George Dalgarno
and John Wilkins have all the esoteric characteristics of a cultic secret code,
known only to a few and never acquiring many users, much less enough to
restore the presumptively lost linguistic world. Having at length identiWed the
natural categories from which all concepts in general derive, for example,
Wilkins proceeds to devise a symbolic ‘philosophic language’ that both
manifests the conceptual reality of the universe and in so doing provides for
perfect communication: ‘The Wrst thing to be enquired after, is to Wnd out
Wtting Marks for the common Genus’s or Heads in the former tables of
Integrals, which are there reduced to the number of forty.’41 There follow
seventy pages demonstrating this ‘philosophic language’, which is written in a
kind of geometric script in which characters receive additional lines in a Wxed
pattern reXecting the ‘tables of Integrals’.
By the late eighteenth century, there emerged still another heuristic
for change and variation that sustained the Babel model and its narrative
of perfect communication: the discipline of linguistics. Beginning with
Sir William Jones’s 1786 hypothesis of the common origins of Sanskrit,
Greek, Latin, Gothic, and Celtic, comparative linguistics in particular focused

39 Gazophylacium Anglicanum (1689; rpt. Menston: Scolar Press, 1969), a4r .

40 Verstegan, A Restitution of Decayed Intelligence, 192–3.
41 Wilkins, An Essay towards a Real Character, and a Philosophical Language (1668; rpt. Menston:
Scolar Press, 1968), 386. On Wilkins see further Eco, The Search for the Perfect Language, 238–59.
98 Language Anxiety

increasingly on the deWnition of principles of historical reconstruction and on

the classiWcation of languages that such reconstruction allows. Jones, to be sure,
built on the work of previous scholars, but it was his own ‘Third Discourse’ to
the Asiatick Society that he helped found in Calcutta that speciWcally served
as both a catalyst to nineteenth-century philologists like Bopp and Müller and,
for several schools of modern linguistic thought, a critical aYrmation of the
importance of what lies beyond variation and change. Bopp and Müller, for
example, devoted themselves to the reconstruction of Indo-European, the
parent language from which nearly every European language, as well as several
subcontinental Indian ones, evolved. Moving back through regular sound
changes like the Great Vowel Shift and the First Consonant Shift, philologists
traced modern language forms to medieval ones, medieval ones to antique ones,
antique ones to hypothetical ones, and hypothetical ones to a Babel-like time,
perhaps in the Wfth millennium bc and perhaps in the area between the Black
and Caspian Seas, when Indo-European speakers began to migrate and their
putatively single language began to diversify.
While I don’t mean to suggest either that comparative philologists believed
in a literal Tower of Babel or that their speciWc goal was the reconstruction of a
perfect language, I do mean to underscore that the descent of language as they
traced it harmonized well with the Babel model of Genesis 11. Similarly,
and from wholly diVerent critical perspectives, although neither structural
nor transformational linguistics share much methodologically with Indo-
European studies, much less biblical exegesis, they, too, concentrate on the
recovery of a moment beyond variation and change—the timeless synchrony
of Saussure’s langue or Chomsky’s ideal speaker-listener. In bracketing oV the
mistakes and misstatements of ordinary speech, both approaches imagine the
possibility of language uncomplicated by the realities of daily communication.
The Babel model embraces still other schools of linguistics. It may seem just
as counter-intuitive to link Benjamin Lee Whorf with Bopp and Müller and
then with Saussure and Chomsky, since the primary concerns of Whorf ’s
research and reputation were the individual and sometimes mutually exclu-
sive ways in which languages represent empirical phenomena. The Sapir-
Whorf hypothesis, to which he lends his name, embraces Genesis 10 in its
insistence that language is inWnitely variable and signiWcantly determinative
of speakers’ perceptions of their experience, so that speakers of diVerent
languages might be judged to live, eVectively, in diVerent worlds. And yet it
was Whorf who wondered about the viability of ‘restoring a possible common
language of the human race or perfecting an ideal natural tongue constructed
of the original psychological signiWcance of sounds, perhaps a future common
 Narratives of Change 99

speech, into which all our varied languages may be assimilated, or, putting it
diVerently, to whose terms they may be reduced’.42
Most recently, this same aspiration for a common language and unmediated
communication informed the controversial linguistics of Joseph Greenberg,
who sought to construct super-families of languages, encompassing, for ex-
ample, enormous numbers of American Indian languages, and from these
languages to reconstruct in turn a proto-world language.43 Like St Augustine,
Verstegan, Bopp, Chomsky, and even Truss—in imagining, as she does, perfect
punctuation for a perfect language—Greenberg begins with the presumption
that there is or can be or once was a uniWed language, providing stable and
unmediated communication to all speakers. In all these approaches, change
and variation and the way they reXect the realities of human social experience,
travel, and history are, in a word, wrong. They and the historical speciWcs
they articulate are to be resisted and undone. And if variation and change attest
to these speciWcs, then an unchanging language of perfect communication
would in several senses transcend them.
In this chapter so far, in tracing out ways in which the Tower of Babel has
oVered an epitome of thinking about change and variation and the anxiety
they generate, I have touched on the mythic, the exegetical, the fanciful, and
the scientiWc. I want to conclude this portion of my discussion by looking at
another kind of narrative, one that seems to embody all these qualities:
Esperanto. While there have been other artiWcial and artiWcially modiWed
languages before and since its creation—including Volapük, Latino Sine
Flexione, and Tutonish—none has had the success of Esperanto in terms of
numbers of speakers, institutional support, or public recognition.44 Like most
of these languages, Esperanto is meant to be auxiliary, supplementing but not
replacing indigenous languages. Also like most of these languages, Esperanto
points towards a world in which communication is immediate and straight-
forward and language is transparent and constant. As with the Tower of Babel
and perfect languages, therein lie several provocative contradictions.
While Esperanto may claim to be a world language dissociated from any
particular nationality, region, or group, for instance, it is very much the
creation of a single individual—Dr Lazar (or Ludwik) Zamenhof—and the

42 Whorf, Language, Thought, and Reality: Selected Writings, ed. John B. Carroll (Cambridge, MA:
Technology Press of Massachusetts Institute of Technology, 1956), 12.
43 See Greenberg, Indo-European and Its Closest Relatives: The Eurasiatic Language Family (Stan-
ford, CA: Stanford University Press, 2000). A great many linguists have called into question the
reliability of Greenberg’s conclusions and methods, which claim genetic connections among sounds or
words on the basis of morphological similarities that seem to be no more than chance.
44 The Conlang Directory of International Communication lists well over 150 invented languages.
See [Link]
100 Language Anxiety

circumstances of his life. Born into Tsarist Russia in 1859, Zamenhof grew up
in a polyglot society at a time of active suppression of ethnic groups. As a
Haskalah Jew, he was raised in both strongly intellectual and religious tradi-
tions, and while a medical student in Warsaw he became actively engaged with
Jewish groups involved in emigration to Palestine and the United States.45
Zamenhof ’s interest in the integrative qualities of language, religion, and
society emerged from the experiences of this speciWc time and place, and
already at the age of ten he wrote a Wve-act tragedy based on the Tower of
Babel and staged in Bialystok. By nineteen he had begun developing and
teaching early versions of Esperanto (meaning ‘one who hopes’), although
the earliest grammar did not appear until the 1887 Russian-language The
International Language: Preface and Complete Manual (for Russians).46 De-
scribed very brieXy, Esperanto is an absolutely regular language, with no
grammatical exceptions, minimal inXectional morphology, and a derivational
morphology that, like Wilkins’s ‘philosophical language’, allows for the easy
and consistent conversion of one part of speech into another. Semantically,
while it is not an avowed perfect language, it does seek to represent experi-
ential categories that transcend situation, time, and geography.47 In the
eastern European political climate of Esperanto’s formative period, domin-
ated by contentious and repressed ethnic and national identiWcation, the
popularity of this regularized language grew quickly, leading to two landmark
events in 1905: Wrst, Zamenhof published the Fundamento de Esperanto,
an extensive grammar that has served as the reference point for all discussions
of Esperanto and its growth, and, second, the First International Congress
of Esperantists took place in Boulogne.
From these beginnings, two issues have remained prominent in Esperanto
studies that have particular relevance to the anxiety associated with language
change and variation. The Wrst is grammatical, involving the nature of
Esperanto and the distinction between correct and incorrect forms. As a
living if artiWcial language, Esperanto very early became subject to the same
forces that change completely natural languages, moving away from what
Zamenhof had designed and raising the question of whether the language
should be left to the use of its speakers or monitored in some way to preserve
a particular, presumptively essential, character. Put another way, the question

45 Peter G. Foster, The Esperanto Movement (The Hague: Mouton, 1982), 51–2; Marjorie Boulton,
Zamenhof: Creator of Esperanto (London: Routledge, 1960), 23–7; Eco, The Search for the Perfect
Language, 324–6.
46 Foster, The Esperanto Movement, 50.
47 See further George Alan Conner et al., Esperanto: The World Interlanguage (London: A. S. Barnes,
1948).
 Narratives of Change 101

early Esperantists faced was to what extent should the language, created
almost instantaneously in its own Edenic moment, be allowed to advance
towards its own Tower of Babel? The Fundamento provided an answer if not
the resolution to this question by not only establishing the language’s basic
grammar but also articulating the essential paradox of a language meant to be
of the people that can be maintained only through authoritarian management
of what those people say. While it lays out all the grammatical rules, ‘in which
nobody has the right to make change’, it also claims that the language can be
adapted to accommodate new expressions, ‘as is done in any other language’.
Indeed, according to the Fundamento, ‘The material master of this language is
the whole world and anyone who wishes can publish in or about this language
all works that he wishes, and can make use of the language for every possible
kind of action’. The Fundamento complicated the issue of change and vari-
ation even further by going on to embed Esperanto, the pre-eminently neutral
language, in government policy. Once Esperanto is ‘accepted by the govern-
ments of the most important nations and such nations by a special law
guarantee to Esperanto certain life and use’, the Fundamento claims, then
‘an authoritative committee’ can be established to change ‘the foundation of
the language’, if necessary. Until that time, Esperanto ‘must strictly remain
absolutely unchanged’.48 In this division of responsibility, governments would
have the authority to approve neologisms but not alterations of the language’s
grammar. All by itself Esperanto thus came to express the historical dialogue
between narratives of change as natural and those of change as divisive.
By the Fourth International Congress, held in Dresden in 1908, Zamenhof,
while guarding his privilege as the inventor of Esperanto, was attempting to
disqualify himself as its pre-eminent grammatical arbiter. The language sub-
committee of the Congress there elected an Academy from its own member-
ship, charged with the task of conserving the basic principles of Esperanto and
also, on a two-thirds vote of its members, mandating change. In fact, no
radical changes were ever implemented, and throughout Zamenhof ’s life
(he died in 1917) tension remained between anti-reformists, who considered
the Fundamento absolutely unchangeable, and those who sought to allow
Esperanto to transform like any living language.49 In the process, in 1907, yet
another language developed from Esperanto: Ido, which attempted to re-
spond to the most frequent criticisms about concord, the accusative case,
and morphological derivation in Esperanto. Rooted in a more fundamental
tension over the nature of linguistic change and variation, the tension

48 Quoted in Foster, The Esperanto Movement, 90, 111.

49 See Foster, The Esperanto Movement, 154–66.
102 Language Anxiety

between reformists and anti-reformists of Esperanto could never be Wnally

resolved, because it rested on a paradox reminiscent of perfect languages and
the Tower of Babel: the maintenance of a putatively natural language by
decidedly unnatural means.
The second issue involving Esperanto that has signiWcance for the study
of change and variation is the claim, made early and insistently, that the
language is ethnically and politically disinterested, a claim that often took on
a millenarian quality. The ‘Declaration of Boulogne’ issued at the 1905
Congress, for example, pronounced Esperanto a ‘neutrally human language’
that ‘would give men of diVerent nations the possibility of understanding
between one another’ and could be a ‘peace-keeping language’ in countries
where there was linguistic contention. All this recalls Postel’s aspiration for
the revival of Hebrew. Indeed, in his own speech to the Congress Zamenhof,
invoking the Tower of Babel and Pentecost and ending with a prayer for God
to reunite humanity, spoke about a golden age when all people would
understand one another and there would be no more linguistic confusion
leading to war.50 Yet the notion of a politically neutral, living language, echoed
at the 1906 Second International Congress in Vienna and afterwards,51 was as
unreachable as centralized management of change and variation in a living
language. As there was tension between reformists and anti-reformists,
so there was tension between local, national Esperanto organizations and
the international movement centered on Zamenhof. The paradox of a polit-
ically neutral, living language is perhaps emblemized by the League of Na-
tions’ response to Esperanto. An avowedly neutral organization, the League
nonetheless consistently supported the language. In 1920 (at its Wrst meeting)
the League debated a proposal to make Esperanto a second language for all
and thus a de facto world language; in 1921 it legislated a study of the status
and viability of Esperanto; and in 1924 it passed a resolution that recom-
mended the use of Esperanto in telegrams.
More pointedly political, as Tsarist Russia had suppressed Esperanto as
a threat to its authoritarian rule, so the Soviet Union initially supported it
as a means to disseminate socialism around the world. In 1921 the Soviet
Union even formed the Soviet Esperanto Union for this very purpose.
This same international applicability led France to resist League of Nations
resolutions in favor of Esperanto and the Soviet Union itself, eventually, to
reject it and imprison Esperantists in the Gulag.52 Nazism, the other face of

50 Foster, The Esperanto Movement, 81–9.

51 Boulton, Zamenhof: Creator of Esperanto, 106.
52 Foster, The Esperanto Movement, 172–203.
 Narratives of Change 103

twentieth-century totalitarianism, invested Esperanto with equal political

force. In Mein Kampf, Adolf Hitler, coming from much the same sociolin-
guistic background as Zamenhof, had said: ‘As long as the Jew has not become
the master of the other peoples, he must speak their languages whether
he likes it or not, but as soon as they become his slaves, they would all have
to learn a universal language (Esperanto, for instance!) so that by this
additional means the Jews could more easily dominate them.’53 With its
avowed political neutrality attracting increasing political attention, in 1935
the German Esperanto Association excluded all Jews from its ranks and
aYrmed its desire to advance Nazi objectives; the Association even oVered
to translate Hitler’s works into Esperanto. Seen as the medium of communist
and international thinking antithetical to the goals of the Third Reich,
however, Esperanto was outlawed in Germany in July of 1936.
In short, Esperanto could not escape politics, not of France, Germany, or
Russia. Even the claim of political neutrality, as the eVorts of the League of
Nations indicate, itself proved political. The language has served as a cipher to
be invested with successive cultural meanings: initially, for Zamenhof, a
means to international peace; to the Tsarists, a weapon of ethnic revolution;
to early Soviets, a vehicle to disseminate socialism; to later Soviets, a mani-
festation of bourgeois internationalism; and to the Nazis, an instrument of
Jewish and communist separatism. For his part, Zamenhof increasingly
devoted himself to his millenarian concerns, moving his focus from Judaism
to Hillelism to what he called Homaranismo, a quasi-religious concept
expressing belief in the brotherhood of all human beings, rejection of racism
and ethnic prejudice, and acceptance of equal rights for all. In 1913, in fact, he
renounced control of Esperanto in order to devote his time and eVort to
Homaranismo.54
Esperanto is often regarded as the work of a crank, utilized only by cranks
for cranky purposes. It is for this very reason that I have dwelt at some length
on it. Grammatically, it does indeed have a rigidity that poorly reproduces the
expressive Xexibility of natural languages; and if its semantic suppositions
about universally shared truths and language’s ability to express them may
have seemed merely odd in the late nineteenth century, they are completely
untenable in a post-modern world. The very notion of an individual building
a language evokes the quirky charm of an eccentric intellectual and the
monomania of a messianic complex. Belief that such a language might ever

53 Quoted in Foster, The Esperanto Movement, 221; see also 218–19. On Nazi suppression of
Esperanto, see Boulton, Zamenhof: Creator of Esperanto, 208–18.
54 Boulton, Zamenhof: Creator of Esperanto, 178.
104 Language Anxiety

be widely used seems naive or disingenuous or both, and the entire enterprise
invests language academies with far more direction over usage than has
ever proved possible. For all these reasons, competency in the language—
not to mention Xuency—raises questions of priorities and time management,
questions that if anything become more insistent as the world, a century
after Zamenhof ’s death, seems closer to the dissolution that his millenarian
linguistic vision sought to hold at bay.
And yet, Esperanto shares a good deal with the biblical, exegetical, and
scientiWc traditions I have already considered. Like them, the language con-
ceptualizes change and variation as problems in themselves, which lead to
additional communication problems among speakers and thence to add-
itional social problems, like conXict and war. Even leaving aside Zamenhof ’s
belief in a global religion, which the leading Esperantists rejected anyway,
Esperanto conceives of a world in which perfect, unmediated communication
is possible, and, once possible, remains. Far from Leibniz’s best of all possible
worlds, our own world is a damaged and deteriorating one, with language
as both sign and cause of the current situation, yet it is also a world that has
the potential for improvement. It is not coincidental that Zamenhof should
have invoked Babel, for like St Augustine, early grammarians, and later
philologists, Zamenhof presupposed a fundamentally conservative, even re-
actionary, view of the world. Esperanto may be a language, but ultimately
what it resists, as its origin and history suggest, is the extra-linguistic sphere of
politics and ethnicity. In this vein, the very idea of a neutral language, central
to Zamenhof ’s thinking, is preposterous, for it is the absence of neutrality—
partiality of one kind or another—that gives speakers a reason to talk,
something to say, someone to say it to, and the means to say it. In construct-
ing his auxiliary language, Zamenhof, again like St Augustine, worked at
cross-purposes with the nature of change and variation in natural languages.
In many ways, then, Esperanto enacts a Babel-model of communication
and thereby epitomizes many of the issues I consider in this book. The
paradox that confronted early Esperantists—that of maintaining a natural
language through unnatural control—is the very paradox that animates the
dialogue of Babel. That being the case, it is striking that speakers of Esperanto
should be judged cranks, when those who champion global English and
rail against variant pronunciations, the inXux of non-Anglophones into
predominantly Anglophonic countries, bilingualism, the development of
new varieties through second language acquisition, and changes in usage—
all of which share Babel’s presuppositions about perfect communication and
its anxieties about change—are not. Indeed, if Zamenhof worked at cross-
purposes with change and variation, the same might be said of attempts to
 Narratives of Change 105

engineer pronoun usage, vocabulary, and contact between English and other
languages. It is also striking that the narrative of Babel model should provide
such explanatory power for responses to change and variation, since the
ultimate moral of Nimrod’s eVorts is that perfect communication is not
simply impossible; it is lethal. Like Icarus or Prometheus, Nimrod aspired
to be god-like and was destroyed for his eVorts. Perfect communication would
seem to be even more lethal than wings or Wre, since Icarus and Prometheus
failed as individuals but Nimrod’s downfall was the world’s as well. Encour-
aged by narratives of change and variation, we may resist the linguistic
realities of this world—the grammatical, pragmatic, and demographic Xuc-
tuations across time—in pursuit of a golden era wherein meaning was
transparent, language constant, and communication Xawless. But in so
doing we relive Nimrod’s pride and resist the extra-linguistic world in an
attempt to transform it in our own image. As the history of Esperanto
suggests, this world, ultimately, may be as irresistible as the linguistic one.

Diglossia, Early Modern English, and dialect-writing

While the Tower of Babel supplies a theoretical model for change and variation
in the English language—the linguistic equivalent of a uniWed theory—narra-
tives of Wction supply that model’s details. They can and do make change real
in two ways. First, following White’s analysis of narrative, Wctional stories oVer
competing ways of remembering and representing linguistic phenomena. In
the case of the origins of change and variation, the dialogue between Genesis 10
and Genesis 11 emblemizes the competition between representing these origins
in ordinary human experience (i.e., language is shown or presumed to change
as naturally as Noah’s descendants moved about) or in extraordinary divine
punishment of human pride. The competition represented in English Wction
over the centuries is more complex than that in the biblical account, however,
reXecting as it does the intimate connections between language, art, and daily
experience. As the sociolinguistic variables of class, ethnicity, nationality, and
age have historically continued to produce variation, so Wction, increasingly,
has served as a forum for more involved, persistent conXict and argument.
The second way by which Wction makes change and variation real involves
its mimetic qualities. To the extent that Wction seeks to evoke a world recog-
nizable by its readers, that is, it portrays human experience with at least
some degree of familiarity: characters need to act, think, and, most import-
antly for the current discussion, speak in fashions that resemble those known
(if not used) by readers. If a writer is to make any particular point about a
speech act like a church sermon, such as that the congregation is ironically
106 Language Anxiety

indiVerent to the message it hears, the portrayal of the scene must have
enough of the qualities of a real-life sermon for the reader to see the parallel.
And the same is obviously true of speakers’ discursive strategies, conversa-
tional structures, and other aspects of how language mediates human behav-
ior. As I will suggest below, this must be the case even in works of science Wction
or fantasy, which portray worlds that may be profoundly diVerent from the
ones their readers experience but which require some similarities for the reader
to make an invested identiWcation with them. By the same token, so long as a
threshold of familiarity is met, Wction may represent parts of the world—
including the role of language—as writers might fear or wish them to be, even
if these representations would seem clearly to diVer from readers’ lived ex-
perience. In view of Wction’s power to make change real, then, and in order
to suggest the range of both representations of linguistic variation and the
anxieties these representations can evoke, I want to survey a range of works
from throughout the chronology of English literature. My intention is not to
suggest an evolutionary development in Wctional treatments of change and
variation, though certain devices are historically speciWc. Nor is it to argue
that one particular treatment has been inevitable or even predominant; there
certainly are exceptions to what I talk about here. Rather, I want to sketch out
some of the conXicting narratives of change, variation, and the social issues
they reXect in order to lay open how Wctional narratives have served as
one means for speakers to organize their thoughts on language and channel
real-world responses to it.
I have already alluded to the biblical account of Pentecost, which shares
with the Tower of Babel the connection of change and variation to other-
worldly causes and consequences: both stories depend on God’s intervention
in human aVairs, elicited by the weaknesses of humans and directed, ultim-
ately, at their improvement. This same notion that change and variation are
non-normative and even supernatural informs several written traditions,
both English and non-English. One particularly well-developed tradition
shaped English historiography of the eleventh, twelfth, and thirteenth cen-
turies, a time following the Norman Conquest and the imposition of
a Francophone nobility. Like any other sizeable region, the British Isles
had always been multilingual, with English, Latin, Cornish, Welsh, and Gaelic
co-existing in various regions. The Norman Conquest introduced not simply
a new language to this repertoire—French—but also a more rigid structural
diVerentiation among languages known as diglossia, in which the use of
a speciWc language helps to deWne speciWc domains and speech acts. Conven-
tionally, Latin was the language of ecclesiastical and many legal domains,
French of the court and law, and English, the only language of most of
 Narratives of Change 107

England’s inhabitants, a largely oral language, heard in many places (includ-

ing the court and legal proceedings) but restricted in writing, at least in the
early Middle Ages, to ephemeral traditions of lyrics, romances, and chron-
icles. To be sure, this situation changed as the centuries advanced, with more
English writing appearing in more domains and with some multilingual
individuals exploiting for stylistic eVect the lexical, syntactic, and rhetorical
resources of Latin, French, and English. For example, John Gower wrote a long
narrative poem in each of these languages. His was a situationally strategic
approach to multilingualism and diglossia, leading him to oVer an interpre-
tation of the unnatural outcome of Babel (which I quoted above) that matched
neither his own multilingual background nor the well-established distribution
of languages in universities and business enterprises but that did Wt his
lament about the decline of English society in the days of King Richard II.
While Xuent in Latin and French as well as English, and while well versed in the
status of each within England’s linguistic repertoire, Gower could situate what
was in eVect the origin of his own multilingualism in Nimrod’s sinful pride,
simply because of the narrative demands of the Confessio.55
Although I do not want to minimize the nuances in the multilingualism
of someone like Gower, I do want to assert that the infrastructure of diglossia
continued to inform England’s linguistic repertoire until the waning of
the Middle Ages, oVering a theoretical model that assigned particular roles for
particular languages and therefore conservatively working against changes in
any language’s grammar and discursive functions. Diglossia also provided
a framework for sociolinguistic behavior and, in turn, for the mediation
of non-linguistic concerns through language in ways that could facilitate
the divergence of linguistic reality and discursive representation from one
another. For post-Conquest English historiography—written mostly in
Latin but also in French and English—the reality of diglossia did in fact contrast
with narrative accounts that represented England as predominantly monolin-
gual by passing silently over events in which English, French, and Latin would
have co-occurred. While chroniclers like Orderic Vitalis and Richard Devizes
noted, respectively, that William I failed in his attempts to learn English and
that the equally Francophone Queen Eleanor (the wife of William’s great-
grandson Henry II) lacked the ability to speak English,56 and while other writers

55 See further Machan, ‘Medieval Multilingualism and Gower’s Literary Practice’, Studies in Phil-
ology, 103 (2006), 1–25.
56 Orderic Vitalis, The Ecclesiastical History of Orderic Vitalis, ed. and trans. Marjorie Chibnall,
6 vols. (Oxford: Clarendon, 1969–80), ii, 257; Richard of Devizes, The Chronicle of Richard of Devizes of
the Time of King Richard the First, ed. and trans. John T. Appleby (London: Thomas Nelson and Sons,
1963), 59.
108 Language Anxiety

implicated diglossia in England’s social and geographic landscapes, most epi-

sodes in which language contact would have occurred pass without comment.
This is the case not only with daily business or religious encounters in England
but also in battles between the English and their Saracen foes on the Crusades
and even in battles that do not involve the English.
Such silence expresses the resistance to change and variation that Babel
mythologizes, but an even stronger expression comes from the depiction of
multilingualism as the byproduct of preternatural experiences like Pentecost.
If early historiographers bypass English–French contact, that is, they also do
much to deWne variation in general as outside the norm of ordinary human
experience and often occurring with dangerous as well as other-worldly
consequences. It is Orderic Vitalis who again provides an apt illustration in
his account of the French and English aYrmation of William I as England’s
proper king. ‘But at the prompting of the devil, who hates everything good,’
Orderic observes, ‘a sudden disaster and portent of future catastrophes
occurred. For when Archbishop Ealdred asked the English, and GeoVrey
bishop of Coutances asked the Normans, if they would accept William as
their king, all of them gladly shouted out with one voice if not in one language
that they would. The armed guard outside, hearing the tumult of the joyful
crowd in the church and the harsh accents of a foreign tongue, imagined that
some treachery was afoot, and rashly set Wre to some of the buildings’.57 The
joyous aYrmation of William may suggest the harmony and righteousness
of his reign, but the devil’s appearance brings together language variation,
social upheaval, and the supernatural, rendering all three as latent, literally
inXammatory threats to the prosperity of post-Conquest England. Other
chroniclers, such as Walter Map, Ralph Coggeshall, William of Newburgh,
and Gerald of Wales, correlate multilingualism with enchanted knights, evil
spirits, spectral children, and pygmies, while in attributing glossolalia to
St Bernard and others Gerald also evokes Pentecost and attaches hagiographic
and spiritual qualities to language variation:
There is the story of St. Bernard, who spoke to the Germans in the French language,
which they were wholly ignorant of, and Wlled them with such devotion and com-
punction that he called forth Xoods of tears from their eyes. With the greatest of ease
he softened the hardness of their hearts so that they did and believed all he told them.
Yet when an interpreter faithfully set forth to them, in their own tongue, everything
Bernard had said, they were not moved at all. It is clear from this incident that what
was accomplished was more the result of holiness than of words.58
57 Orderic Vitalis, The Ecclesiastical History, ii, 185.
58 Gerald of Wales, The Jewel of the Church, ed. and trans. John J. Hagen. (Leiden: E. J. Brill, 1979),
117. Hagen notes that this incident is also recorded in the Vita et Res Gestae Sancti Bernardi Libris
 Narratives of Change 109

St Bernard’s accomplishment may well have resulted more from holiness

than from words, but Gerald’s words testify as much for the displacement of
social anxiety to language as for Bernard’s sanctity. A Norman claim on the
English throne rested, in the Wrst instance, on a tenuous story (much
ampliWed by Norman historians after 1066) that Edward the Confessor had
committed the kingdom to William of Normandy, an arrangement that had
been accepted and then treacherously ignored by Harold Godwinsson,
Edward’s nephew and immediate successor. More than this, the Norman
claim rested on the success of the Conquest and subsequent reorganization
of the Anglo-Saxon church, courts, and political system, which included
positioning of Normans in the most inXuential positions. The Normans
may have felt the justice of their claim, and they certainly highlighted
Anglo-Saxon foibles in order to present themselves as God’s scourge, but
the historiographic presentation of variation and change betrays an anxiety
about their presence in England. Through discursive practices that did not
simply erase English–French contact but also displaced all multilingualism to
the realm of the dangerous and supernatural, Norman historiographers did
much to exclude variation and change from the orbit of ordinary human
experience. In the process they also did much to justify Norman political
activities. By muting the sociolinguistic consequences of the Conquest—the
inescapable fact of multilingualism and French–English contact in England—
they also muted much post-Conquest regnal upheaval, including the twelfth-
century anarchy during the reign of King Stephen, as well as the daily evidence
that the Normans had appeared in the country as invaders. Because of the
breach in writing, history, and tradition that they themselves produced, the
Normans would have beneWtted from the suppression of all such breaches,
and the representation of change and variation in post-Conquest chronicles,
whatever the chroniclers’ individual motivations, served this very purpose.
By the early modern period, England’s diglossia had weakened consider-
ably. The Reformation produced English prayer books and Bible translations
and mandated the use of English, rather than Latin, in religious services,
giving impetus to the use of English in other domains. English had already
begun to pervade business activities, and by 1600 it had largely replaced
French in courts of law; throughout the seventeenth and eighteenth centuries,
it became the dominant language in grammar schools and at Oxford and
Cambridge. And yet even in this period diglossia and the resistance to change
and variation that it modeled continued to animate rhetorical representations

Septem Comprehensae, by William of St Thierry. See further Machan, ‘Language and Society in
Twelfth-Century England’, in Irma Taavitsainen et al. (eds.), Placing Middle English in Context: Selected
Papers from the Second Middle English Conference (Berlin: Mouton de Gruyter, 2000), 43–66.
110 Language Anxiety

of language. For instance, even as the status of English continued to change,

with expansion of its domains and codiWcation eVorts that I will consider in
Chapter 5, the value of Latin was reiterated in various ways. Sir Thomas Elyot,
in his widely read and reprinted 1531 Book of the Governour, encouraged the
use of Latin ‘as a familiar langage’ for a noble’s son, with him ‘hauynge
none other persons to serue him or kepyng hym company / but suche as
can speake latine elegantly’.59 Much as would be the case with Standard
English in the centuries to come, Latin’s value was understood to be the
stature, durability, and constancy that accrued to it as a classical language and
that granted permanence and intelligibility to works written in it, as opposed
to those written in English. In his ‘Of English Verse’ the seventeenth-century
poet Edmund Waller thus opined,
But who can hope his lines should long
Last in a daily changing tongue?
While they are new, envy prevails;
And as that dies, our language fails.

When architects have done their part,

The matter may betray their art;
Time, if we use ill-chosen stone,
Soon brings a well-built palace down.

Poets that lasting marble seek,

Must carve in Latin, or in Greek;
We write in sand, our language grows,
And, like the tide, our work o’erXows.60

Waller goes on to instance Chaucer, whom late-medieval poets like Thomas

Hoccleve and John Lydgate had lauded for his ‘illumination’ of the English
language, as someone whose verse the passage of time and the change of
language have diminished and obscured. For Waller, change and variation in
English cause us to forget who we are and the world, in turn, to forget us.
Inasmuch as paeans to Chaucer’s accomplishments continued through the
early modern period and into the nineteenth century—to Spenser, he was
famously a ‘well of English undeWled’—there is of course a contradiction
between the putative originary status of Chaucer’s language and its obscurity,
between English as an established tradition and English as a Xawed medium
subject to time and changing in ways that impede communication. This

59 Elyot, The Boke Named the Gouernour (1531; rpt. Menston: Scolar Press, 1970), 30.
60 Waller, The Poems of Edmund Waller, ed. G. Thorn Drury (London: Lawrence and Bullen, 1893),
197–8.
 Narratives of Change 111

contradiction appears with particular prominence in Sir Francis Kynaston’s

Amorum Troili et Creseidae Libri Duo Priores Anglico-Latini, issued in 1635 and
containing, in facing-page format, the Wrst two books of Troilus and Criseyde
in English (printed in the already old-fashioned black letter font) and in a
Latin translation (printed in roman italic). A sequence of prefatory poems by
Kynaston and his acquaintances emphasizes that it is Chaucer’s poetic and
rhetorical achievement that justiWes the volume. Kynaston himself begins his
preface by referring (in Latin) to Chaucer as a ‘venerable and ancient poet’,
who is ‘the ornament of this island and distinguished glory of poetry’. Yet
Dudley Digges (likewise in Latin) notes that by writing in English Chaucer,
unlike Homer or Virgil, has restricted his own worthy reputation: ‘The fame
of such a name ought j to lie before the world, not just an island.’ If Chaucer’s
reputation motivates the continued reading of his poetry, then, the imper-
manence of language eVectively victimizes the Troilus and impedes commu-
nication between poet and reader. Chaucer, Kynaston notes, is ‘not only
growing old, diminishing in value beneath the obsolete and already scorned
clothing of the ancient English idiom, but—how sad!—wholly wasting away
and nearly dead’. Later, in his dedication to Book One, he avers, ‘I desired the
preservation from ruin and oblivion of his gem of poems, which was nearly
lost and scarcely understood by us (at least as the favorite of none) because of
ignorance of the obsolete words in it which have fallen into disuse’. In
response, Kynaston’s objective is to resist such obsolescence by securing
Chaucer ‘with the lasting support of Roman eloquence’ and rendering him
‘again stable and immobile through all ages (however many we have left)’.61
Ironically, then, even as in the Troilus Chaucer earnestly entreats his audience
to accept the fact that in language as well as in love, there is ‘chaunge’ in the
course of a millennium, Kynaston (and others) advanced the poet’s claim to
greatness by citing this same ‘chaunge’ as both a sign of the world’s transience
and a justiWcation for their own literary ministrations.
While the Amorum Troili et Creseidae Libri is not a unique early modern
production—in 1690 William Hogg published a volume containing Latin
translations of Milton’s Paradise Lost, Paradise Regained, and Samson Ago-
nistes—it is certainly unusual in the individual and collective eVort it em-
bodies. To this end, Kynaston and his coterie may well have been motivated
largely by novelty, by a desire to see whether an exercise as peculiar as

61 Kynaston, ed. and trans., Amorum Troili et Creseidae Libri Duo Priores Anglico-Latini (London:
John LichWeld, 1635), sigs. 1r , y1r ,
r , y1v , and A3v . See further Richard Beadle, ‘The Virtuoso’s Troilus’,
in Ruth Morse and Barry Windeatt (eds.), Chaucer Traditions: Studies in Literature in Honour of Derek
Brewer (Cambridge: Cambridge University Press, 1990), 213–33; and Machan, ‘Kynaston’s Troilus,
Textual Criticism, and the Renaissance Reading of Chaucer’, Exemplaria 5 (1993), 161–83.
112 Language Anxiety

translating an English poem in accentual rhyme-royal stanzas into syllabic

Latin verse could be carried oV; and truth be told, they do just this rather well.
At the same time, set against a background of shifting social classes, the
incipient cultivation of nationalism, economic instability, and religious and
political upheaval—set in a century that began with the uncertainty sur-
rounding Elizabeth’s succession, passed through the interregnum and Civil
War, to the Restoration and Glorious Revolution—Kynaston’s concerns with
the dangerous mutability of language echo the concerns of others. Robert
Crowley, the Wrst editor of Piers Plowman, had observed already in 1550 that
English had changed in so many ways since the poem’s fourteenth-century
composition as to leave Langland’s language diYcult and obscure: ‘The
Englishe is according to the tyme it was written in, and the sence somewhat
darcke, but not so harde, but that it maye be vnderstande of such as wyll not
sticke to breake the shell of the nutte for the kernelles sake’.62 Even earlier, in
the preface to his 1532 edition of Gower’s Confessio Amantis, Thomas Berthel-
ette tried (albeit halfheartedly) to put something of a positive spin on
language change, asserting that it was deWciencies in English vocabulary
that had led early writers to borrow words from ‘latyne / frenche, and other
langages’. But such borrowing ultimately only obscured communication
for readers who were unable to read these languages, while the creation of
new vocabulary does not set ‘a president to vs / to heape [new words] in /
where as nedeth not / and where as we haue all redy wordes approued and
receyued of the same eVecte and strength’.63 Language change elicited similar
(if similarly overstated) anxiety in discussions of English literature well after
Kynaston’s translation, as in 1700, when in the preface to his Fables John
Dryden described Chaucer as ‘the Father of English poetry . . . a perpetual
Fountain of good Sense; learn’d in all Sciences’, but went on to note that in the
time since his death, Chaucer’s language has become ‘so obsolete, that his
Sense is scarce to be understood’. ‘Chaucer’, Dryden confesses, ‘is a rough
Diamond, and must Wrst be polish’d, e’re he shines.’64
So, contemporary discursive traditions as well as novelty may indeed
partly underlie the Amorum Troili et Creseidae Libri. At the same time, within
the framework of the Babel model and early modern social upheaval, the
volume evokes other anxieties associated with language change and variation.
Kynaston himself, obviously, was an aristocrat. Born in 1587 he took degrees

62 Crowley (ed.), Piers Plowman (London: Crowley, 1550), iir---v .

63 Berthelette (ed.), Jo. Gower de Confessione Amantis (London: Berthelette, 1532), aaiiv . See further
Machan, ‘Thomas Berthelette and Gower’s Confessio’, Studies in the Age of Chaucer 18 (1996), 143–66.
64 Dryden, The Works of John Dryden, ed. Edward Niles Hooker and H. T. Swedenberg, Jr. (Berkeley,
CA: University of California Press, 1956–), vii, 39–40.
 Narratives of Change 113

from both Oxford and Cambridge, was knighted in 1618 by James I, and in
1625 was made an esquire of the body to the new King Charles I. In 1635, with
funding from the king, he founded the Museum Minervae, an institute
devoted to the education of the sons of the nobility, and continued to write
sonnets and romances until his death in 1642. Kynaston was thus wealthy, well
connected, and professionally invested in the crown and high culture of
England. If like his contemporaries he exaggerates the obscurity of Middle
English in view of the continual change and variation in English, he also
beneWts in several ways from the exaggeration. For one thing, the ravages of
time require the ministrations of men like Kynaston, specially educated and
qualiWed to maintain continuity with the past and bring that past to the
present. For another, men like Kynaston, uniquely qualiWed as they are,
hold a prerogative on assertions about how language relates to history and
society and about which issues are linguistic and which extra-linguistic. They
represent a social group whose membership is closed but empowered to
make sociolinguistic pronouncements for all speakers—in eVect, to deWne
the history of the language. And for a third, linguistic conservatism in the
face of the disruptions of variation and change can serve as a model for a more
general conservatism of political structure and social values. Because of
his education and position, Kynaston would proWt as surely from stability
in English society as he would from the kind of stability he advocates in
language—for however many (as he ominously observes) ages as we have left.
Not coincidentally, I think, the period in which Kynaston produced
his translation is also the one that witnessed the increased emergence of the
use of regional and social dialects for literary eVect. I say ‘for literary eVect’
because so long as English lacked a standard variety—codiWed in grammar
books and dictionaries, inculcated in schools and universities, and constitu-
tive of powerful domains of business, law, and government—all English
writing was necessarily dialectal. As I noted in the previous chapter, prior to
the modern period written variations among regional dialects could be
copious and consistent enough to enable a poem written in one dialect,
such as Bede’s ‘Death Song’, to be translated into another. This kind of dialect
translation persisted throughout the Middle English period, but the use of
dialect diVerences for literary eVect dates only to the fourteenth century and is
sporadic for some time thereafter. In the Second Shepherds’ Play of the
Wfteenth-century Towneley Cycle of mystery plays, for example, the sheep-
stealing Mak, who is a farcical character and the victim of others’ jokes
and machinations, speaks in an erratic southern variety, setting him oV
from his fellow characters, whose English is northern. One of the Shepherds
asks of him,
114 Language Anxiety

Bot, Mak, is that sothe?

Now take outt that Sothren tothe,
And sett in a torde!65

Caxton’s preface to his 1490 edition of the Eneydos oVers a similar example,
when he tells the story of a merchant who said the northern eggs instead of the
southern eyren and was accused of speaking French.66 Even if true, this story
has a local stylistic eVect similar to that of Mak’s ‘Sothren tothe’: it enhances a
self-conscious passage on the importance of Caxton’s new edition and pro-
vides emphasis in the way metaphors or other tropes might. Also like Mak’s
southern tooth, I would add, Caxton’s vignette shows variation and change as
disruptive and dangerous: Mak, after all, is a sheep stealer, and the merchant is
angered by his exchange, which epitomizes the delay and business costs of
having to remain in the Thames estuary until the wind shifts and he can sail to
Zealand. Change and variation make even Caxton melancholic. His nostalgic
summarizing comment, again, is that ‘our langage now vsed varyeth ferre
from that. whiche was vsed and spoken whan I was borne’.
Chaucer’s ‘The Reeve’s Tale’ oVers a more sustained engagement with the
expressive capabilities of language variation, describing as it does a bawdy and
tempestuous encounter between a southern miller and his family and a pair of
northern Cambridge clerks. The contrast between the characters and their
speech patterns appears in the morphology, vocabulary, and lexicon of their
Wrst exchange:
Aleyn spak Wrst: ‘al hayl, Symon, y-faith!
Hou fares thy fair doghter and thy wyf ?’
‘Aleyn, welcome’, quod Symkyn, ‘by my lyf !
And John also, how now, what do ye here?’
‘Symond’, quod John, ‘by God, nede has na peer.
Hym boes serve hymself that has na swayn,
Or elles he is a fool, as clerkes sayn’.67

While speech diVerences among the characters are not topics in ‘The Reeve’s
Tale’, they do pervade it and necessarily seem to invite consideration of
language variation as integral to the themes of the story, which revolve around
pride, malice, violent sexuality, and overweening ambition. In this regard,

65 A. C. Cawley (ed.), The WakeWeld Pageants in the Towneley Cycle (Manchester: Manchester
University Press, 1958), 48.
66 William Caxton, The Prologues and Epilogues of William Caxton, ed. W. J. B. Crotch, EETS os 176
(1928; rpt. New York: Burt Franklin, 1971), 108.
67 The Riverside Chaucer, ed. Larry D. Benson, 3rd edn. (Boston, MA: Houghton MiZin, 1987),
Canterbury Tales, 1.4023–8.
 Narratives of Change 115

Chaucer’s aberrant use of language—his mixing of regional varieties—oVers

the linguistic equivalent of the social upheaval that the tale describes. In
language as well as theme, that is, ‘The Reeve’s Tale’ depicts a frighteningly
imbalanced world, maintained neither by the strictures of proscribed sexual-
ity or social rank nor by the linguistic stability manifested in diglossia. Change
and variation thereby become as alarming, even apocalyptic, as they are for
Kynaston or the twelfth-century chroniclers.68
When I say that the early modern period witnessed an increase in the use of
social and regional variation for literary eVect, then, I might better say that it
witnesses its creation. The eVorts of Chaucer, Caxton, and the Second Shep-
herds’ Play do not oVer a consistent presentation of linguistic variation’s
rhetorical potential, even if they do, in various ways, suggest it produces the
same kind of anxiety that I have discussed elsewhere in this chapter. By
comparison, just as the early modern period featured the proliferation of
grammars, dictionaries, and rhetorical manuals, all further helping to deWne a
Wxed written standard, so its writings displayed increasingly nuanced use of
variation and change. As I will argue in Chapter 5, this connection is not
fortuitous, since the identiWcation of non-standard language, not to mention
the rhetorical exploitation of it, requires the presence and acceptance of a
standard language. It is at this point that, in the terminology of William
Labov, variation and change cease to be indicators, becoming instead mark-
ers. Labov has in fact identiWed a three-stage development for the rhetorical
signiWcance of regional and non-standard forms, and these can be traced in
English literature.69 The initial stage is when a form is an indicator, evoking
the speech patterns of some social group, whether social or regional, and
connecting a speaker to that group. In the above passage from ‘The Reeve’s
Tale’, thus, Chaucer restricts to the Cambridge clerks the verbal inXection -s
(rather than -eth) and the graph a, suggesting a pronunciation [a] or [A], in
words where some Middle English varieties had o, presumably for [o] (e.g.,
the Wnal vowel in John’s na as opposed to Symkyn’s also). He thereby evokes
speech associated with the north of England both through works and manu-
scripts produced there and through dialect translation of such features in
non-northern works.

68 For further discussion, including consideration of why the regional variety of ‘The Reeve’s Tale’
lacks the humor in and of itself that is sometimes attributed to it, see Machan, English in the Middle
Ages, 112–38.
69 Labov, ‘On the Mechanism of Linguistic Change’, in John J. Gumperz and Dell Hymes (eds.),
Directions in Sociolinguistics: The Ethnography of Communication (New York: Holt, Rinehart, and
Winston, Inc., 1972), 512–38.
116 Language Anxiety

But as the desultory and disparate quality of such representations suggests,

this kind of variation achieves only local rhetorical eVects rather than evokes
broadly shared conceptualizations of a variety within what might be called the
ecology of languages, varieties, and domains in a speech community. It begins
to do so in Labov’s next stage, that of marker. Here, not simply are forms and
varieties associated with a particular group but that association and the group
themselves have recognizable sociolinguistic meanings—involving social rank
or credibility, for example—which representations of the language can ma-
nipulate. Shakespeare’s depiction of the northern language of the Scottish
captain Jamy in Henry V oVers a case in point: ‘It sall be vary gud, gud feith,
gud captens bath, and I sall quit you with gud leve, as I may pick occasion.
That sall I, mary.’ While this language shares certain graphic features with
‘The Reeve’s Tale’ (as in bath for both), it goes beyond it through its contri-
butions to a comic subplot involving the Irish captain Macmorris and the
Welsh captain Fluellen, whose language is similarly spelled so as to evoke their
own regional accents. More generally, such linguistic variation draws on
traditions of comical Celtic stereotypes on the stage and contributes to the
play’s nationalizing themes centered on an England surrounded, in several
senses, by a Celtic fringe. Similar nationalizing (even anti-foreign) traditions
underwrite the presentation of the French princess Katherine, who, in a
memorable passage, confuses English foot and count for homophonic French
obscenities: ‘De foot et de count! O Seigneur Dieu! ils’ont les mots de son
mauvis, corruptible, gros, et impudique, et non pour les dames d’honneur
d’user.’70 While muted and turned to comic eVect, variation in such early
modern representations still conjures images of aberration and deviance from
linguistic and social stability. One of the central concerns of Henry V, it should
be noted, is the early modern fashioning of an English nation from disparate
social groups, and this can take place once Jamy, Macmorris, Fluellen, and
even Katherine speak the same English.
It is thus not coincidental, to oVer one Wnal example, that alongside Henry
V some of the earliest uses of linguistic variation for rhetorical eVect are
sixteenth- and seventeenth-century dictionaries of cant and slang. When an
increase in vagrancy attendant on decreases in subsistence farming and other
social instabilities led to concomitant increases in crime, these works
appeared as, in eVect, defensive weapons, identifying a thieves’ sociolect
and helping magistrates and honest citizens by keeping them informed of
secret methods of communication directed against them. If for Shakespeare

70 Henry V, [Link].94–6 and [Link].48–50. Quoted from The Complete Works, ed. Alfred Harbage
(London: Penguin, 1969).
 Narratives of Change 117

variation in language projected a kingdom in need of uniWcation, handbooks

of cant suggest that deviance in language accompanied deviance in behavior.71

Modern Wction and language change

The third stage of Labov’s model occurs when, framed by the norm of a
standard written language and discursive traditions that shape the use of non-
standard language, forms and varieties become stigmatized. They are then
stereotypes with instantly recognizable social implications. If such implica-
tions provide opportunities for rhetorical exploitation, Wguring in character-
ization and so forth, they also reinforce the stigmatization of the variation.
In response to the way stereotypes foster the continued subordination of
a particular variety or those who use it, moreover, speakers may in fact
decrease their usage of subordinated forms, but the power of stereotypes is
such that they can continue to deWne a speech community even after any
empirical basis for them has disappeared.
Throughout his novels, for example, Dickens uses spellings like wot for
what to represent pronunciations that voice the initial glide, i.e., [wat], rather
than preserve the historically voiceless [ at] or even [hwat], suggested by the
w
Old English form hwæt. And he uses this spelling and its implied pronunci-
ation only for low-class speakers, such as cockneys, with whom the pronun-
ciation had become negatively identiWed in the eighteenth century. Yet by the
middle of the nineteenth century, when Dickens was writing, nearly all British
varieties and speakers voiced an initial [w] (creating homophones of witch
and which), and this pronunciation remains widespread, even typical, today.
While most speakers of his era—including, presumably, Dickens himself—
would have said [wat] and not [ at] or [hwat], the voicing evoked by
w
wot remained conceptually a novelty and distinction of only lower-class
speech.72 As Lynda Mugglestone observes, the success of techniques like
this—and not only this but also Wnal -in for -ing and the presence or absence

71 See further Julie Coleman, A History of Cant and Slang Dictionaries, i, 1567–1784 (Oxford: Oxford
University Press, 2004). For additional examples and discussion of the use of regionalisms for literary
eVects in this period, see Manfred Görlach, Aspects of the History of English (Heidelberg: Carl Winter,
1999), 94–161.
72 See E. J. Dobson, English Pronunciation 1500–1700, 2nd edn. (Oxford: Clarendon Press, 1968), ii,
974; Michael K. C. MacMahon, ‘Phonology’, in The Cambridge History of the English Language, iv,
Suzanne Romaine (ed.), 1776–1997 (Cambridge: Cambridge University Press, 1998), 466–7; Joan
C. Beal, English Pronunciation in the Eighteenth Century: Thomas Spence’s ‘Grand Repository of the
English Language’ (Oxford: Clarendon Press, 1999), 176–80; and Roger Lass, ‘Phonology and Morpho-
logy’, in Lass (ed.), The Cambridge History of the English Language, iii, 1476–1776 (Cambridge:
Cambridge University Press, 1999), 124.
118 Language Anxiety

of initial h and post-vocalic r—‘depends not on the Wdelity which may or

may not exist between the written and spoken forms selected but instead on
the clear perception of notions of form and deviation which are oVered to
the reader’.73 And they depend, too, on readers’ recognition of the stylistic
device, which is a kind of recognition that developed over the course of the
eighteenth and nineteenth centuries and on which Dickens could depend
implicitly.
This is the stage at which writing in dialect has become dialect writing: a
fully developed system that isn’t necessarily or even usually accurate (as
Dickens’s wot suggests) but that maps recognizable social valuations onto
an artiWcial arrangement of spelling, italicization, capitalization, hyphenation,
and punctuation. And it is a stage so thoroughly ingrained in modern
Wctional narratives that its history and development can be taken for granted.
Easy generalizations about literature written from the eighteenth through the
twenty-Wrst centuries certainly fail, since non-standard spellings throughout
this period suggest regional and social variation in many ways, directed at
diVerent literary eVects. One could not detail a line of continuity and devel-
opment from, say, Mrs Malaprop in Sheridan’s The Rivals, through Sir Walter
Scott’s Heart of Midlothian and Mark Twain’s Adventures of Huckleberry Finn,
to William Faulkner’s The Sound and the Fury, and Wnally to Toni Morrison’s
Beloved and Irvine Welsh’s Trainspotting. What many such works do share is
the general association of non-standard spellings with particular pronunci-
ations and, by extension, speakers and groups, as well as a link between this
association and particular social attitudes, qualities, or beliefs. And unlike in
‘The Reeve’s Tale’ or even Henry V, these connections are not novelties that
might require foregrounding or overt explanation. Rather, they constitute a
kind of horizon of literary and social expectations on which writers can draw,
knowing full well that their stylistic methods and thematic implications
will be instantly recognizable. In a sense, modern dialect writing is less a
way to think about change and variation than a reXection of how it has been
thought about. This will be so even if—perhaps especially if—writers work
against type, as did Thomas Hardy in his representations of a Wessex dialect.
A correspondent of Sir James Murray, the general editor of the OED, Hardy
took particular interest in language, regarding regional dialects as ‘intrinsic-
ally as genuine, grammatical, and worthy of the royal title [of English] as
is the all-prevailing competitor which bears it’.74 Hardy’s non-standard,

73 Mugglestone, ‘Talking Proper’: The Rise of Accent as Social Symbol, 2nd edn. (Oxford: Oxford
University Press, 2003), 119. See the whole of her discussion on 95–134.
74 Quoted from Hardy’s personal writings in Dennis Taylor, Hardy’s Literary Language and Victor-
ian Philology (Oxford: Clarendon Press, 1993), 160.
 Narratives of Change 119

Wessex-speaking characters, then, achieve their moral force precisely because

they can be seen to undermine Victorian England’s received expectations
about class, integrity, and language.
Works in which dialect is a Labovian stereotype often share another quality,
and that is the displacement of non-linguistic social anxieties to variations in
language that I have described throughout this chapter. As keen as Hardy was
about language, for example, novels like Jude the Obscure ultimately rage
against social institutions and the limitations of a class system that they
support, not merely individuals’ speech patterns. Regionalisms may be ‘genu-
ine’, but neither they nor their speakers are socially normative, and if their
variance from normative language channels the intrinsic virtue of their
speakers, it equally channels the tensions of social class. Such is the case,
too, with Dickens, whose novels prominently employ stereotypes (in Labov’s
sense) that, whether they overtly sustain or challenge social hierarchies,
associate language variation with broader kinds of disruption. Great Expec-
tations, for instance, opens with scenes of intimacy between the youth Pip
and the illiterate blacksmith Joe Gargery, which take on additional emotional
force from the contrasting violent scenes with the escaped convict Magwitch
or Mrs Joe, stiXing ones with Mr Pumblechook, and eerie ones with Miss
Havisham. Pip shares with Joe resiliency, honesty, and a compassionate spirit
born of brutal backgrounds, all of which are expressed in the scenes describ-
ing Pip’s eVorts to learn to read and in the non-standard representations
of Joe’s speech: ‘ ‘‘Well!’’ Joe pursued, ‘‘somebody must keep the pot a biling,
Pip, or the pot won’t bile, don’t you know?’’ ’ As the story progresses and
Pip realizes a change in character along with his great expectations, the simple
integrity focused on Joe and emblemized in his non-standard speech con-
trasts sharply with the aVectations of wealth and power in London. ‘I’m
wrong in these clothes’, he tells Pip on a visit to London, ‘I’m wrong out of
the forge, the kitchen, or oV th’meshes’.75 It may be an ironic twist of the
dialect stereotype that it is Pip’s variation from regional to standard speech
that accompanies his self-destructive change in values, but the association
of linguistic variation with social disruption still follows the pattern that
I have been tracing.
Dickens’s Nicholas Nickleby reveals the range of expressive possibilities for
dialect-writing as well as the ways in which it can direct comments on
language change and variation to social ends. For Mr Mantalini, a minor
character who provides comic relief in his tumultuous marriage and philan-
dering ambitions, the pronunciation of a single word (damn) serves as a kind

75 Dickens, Great Expectations, ed. Angus Calder (London: Penguin, 1965), 77, 246.
120 Language Anxiety

of signature. In the course of one brief conversation he remarks, ‘Oh dem!

. . . Demnition discount . . . Demd scarce. . . . demd triXing . . . Oh demmit!’
In contrast is the language of Lord Frederick Verisopht, who Wrst appears
as a shallow rake but who eventually rejects his malicious mentor Sir
Mulberry Hawk and is in fact killed by him in a duel. Irrespective of particular
words, he displays a habit of clipped speech and lengthened vowels. ‘What–
the–deyvle?’ he exclaims at one point, continuing during the same dinner
with ‘it’s not a wa-a-x work’, ‘How de do’, ‘deyvlish pitty’, ‘the most knowing
card in the pa-ack’, ‘how can you a-ask me’, ‘Gad, so he has’, ‘deyvle take me’,‘it
wouldn’t be a good pla-an’, ‘it’s too ba-ad of you’, and ‘ma-ake one eVort’.
While Mantalini’s dem comes across as a kind of verbal tic or prop, the
spoken equivalent to a comedian’s cigar, the more nuanced portrayal of
variation in Lord Verisopht’s language mimics the snobbish aVectation
of the monocle that he moves from eye to eye as he talks. And in that nuance
linguistic variation, which for Mr Mantalini evokes only laughter, projects a
disingenuous and even manipulative quality.
The language of John Browdie, a Yorkshire grain merchant, varies in still
more lexical and phonological detail from Standard English. Upon giving
a destitute Nicholas money to keep him on the road, John exclaims, ‘Dean’t be
afeard, mun . . . tak’ eneaf to carry thee whoam. Thee’lt pay me yan day,
a’ warrant.’76 John shares with Joe Gargery a genuine and decent spirit; he
is delighted to aid Nicholas’s escape from the miserly schoolmaster Squeers,
whom Nicholas has just beaten, and later in the novel he will enable the escape
of the feeble-minded and mistreated Smike, confront Squeers himself, and aid
the Xight of the abused boys of Squeers’s Dotheboys Hall. In this regard, his
non-standard speech, like Joe’s, suggests the absence of aVectation and deceit.
At the same time, however, the variation in their language marks both
characters as uneducated, Wnancially limited, and socially circumscribed.
They may be decent men, but their language indicates that, unlike Pip or
Nicholas, they will never succeed in London’s normative world. Put more
directly, decency appears as socially marked as regionalisms.
While in many ways Hardy, Dickens, and other nineteenth-century novel-
ists aYrmed prevailing notions about the social instability associated with
change and variation, dialect-writing itself came into criticism as a rhetorical
device that undermined the stability of written standards and what they imply

76 Dickens, Nicholas Nickleby, ed. Mark Ford (London: Penguin, 1999), 407–8, 230–5, 161. For some
examples of mid-nineteenth-century American dialect writing put to rhetorical (including comedic)
eVect, see F. G. Cassidy, ‘Geographical Variation of English in the United States’, in Richard Bailey and
Manfred Görlach (ed.), English as a World Language (Ann Arbor, MI: University of Michigan Press,
1982), 189–95.
 Narratives of Change 121

about social stability. In the United States, whose population increased

dramatically through immigration at this time, the language of regional
writers and immigrants concentrated a number of political issues, including
nationalism, education, ethnic and racial identity, geographic diVerence, and
Wnancial opportunity. Writers like Mark Twain, Theodore Dreiser, Willa
Cather, and Upton Sinclair found themselves in the double-edged position
of all writers who employ non-standard language for stylistic eVect: dialect-
writing could be a way to challenge accepted sociolinguistic views and
thereby aYrm that the non-standard language was worthy of representation,
but to do so inevitably involved drawing on comedic and disparaging tradi-
tions that, just as inevitably, it perpetuated. If dialect-writing has celebrated
ethnicity and challenged social stereotypes, it thereby has also aYrmed the
implication of social disruption in linguistic variation.
A particularly voluble authority on this subject was The Dial, a literary
journal published in Chicago from 1880 to 1929. One of the leading propon-
ents of international modernism in its later years, The Dial devoted many
early editorials to regional and non-traditional writing, a topic that resonated
particularly in Chicago, which was the focal point and catalyst of much
nineteenth-century non-traditional writing in the United States. Despite its
advocacy of modernism, The Dial was a largely conservative review, anchored
by a conviction in the prestige and importance of a shared Anglo-American
culture. Accordingly, The Dial echoed Waller and Kynaston on the pre-
eminence of immutable language by arguing against non-standard writing
on the grounds that it failed to achieve the transcendency characteristic of
great literature. One editorial asked, ‘Does the speech of Tommy Atkins and
Marse Chan, the dialect of Drumtochty and Donegal, the locution of the
Hoosier farmer and the Bowery tough, have anything of the antiseptic quality
that preserves a story or a poem and enables it to delight successive gener-
ations of readers?’ Another lamented the state of reading and writing in
American universities, warning of the dire consequences attendant upon the
liberties taken with English ‘morphology, phonetics, syntax, and meaning, for
no more adequate reason than the supposition that such linguistic butchery is
humorous’.
An anxious link between linguistic and non-linguistic concerns appears as
well in The Dial. Citing Athens and Transcendentalist Boston as models of
what a homogenous population can accomplish, one writer wondered: ‘Can a
really great literature grow up in the midst of a heterogenous population,
and how far are we Americans a heterogeneous people?’ More pointedly,
another editorial in eVect glossed the xenophobic, even racist, connotations
of ‘heterogenous population’:
122 Language Anxiety

There are few features of the recent literary situation as noteworthy as the large
production and wide vogue of writings which exploit some special form of idiom
and rely for their main interest upon the appeal of curiosity thus made. The idiom
of the sailor and the soldier, the rustic and the mechanic, have elbowed their way
into literature, and demand their share of the attention hitherto accorded chieXy to
educated speech. The normal type of English expression has to jostle for recognition
with the local and abnormal types of the Scotchman and the Irishman, the negro and
the baboo, and, in our own country particularly, with such uncouth mixtures as those
of the German-American and Scandinavian-American.77

While the emphasis on the relation between an individual’s democratic

expression and the collective character of a nation may have cast this argu-
ment on literary voice in a distinctively American way, the anxiety about
change and variation is familiar from other writers and periods. By the logic
of these kinds of argument, linguistic instability and imprecision result from
the aVectations of particular marginal groups (such as university students)
and, more signiWcantly, from immigrants and ethnic communities; this kind
of social instability undermines national identity as well as the clarity of
speech and writing; various national goods would be served by curtailing
change and variation; an eYcient way to curtail these linguistic phenomena
would be to curtail the process that putatively gives rise to them—immigra-
tion in particular, as well as, perhaps, access to higher education. It is
this same access, of course, that loosened the conceptual hold of men like
Kynaston and Arnold on the history of English and contributed to the
variation that critics like Honey, Truss, and McWhorter Wnd so disturbing.
As the literary use of non-standard and regional writing has proliferated
since the early twentieth century, the double-edged eVect of perpetuating
stereotypes in order to resist, exploit, or expose them has persisted. Novels
like Trainspotting or James Kelman’s How Late It Was, How Late use regional
variants as eVectively their normative language, requiring of readers both
linguistic agility and the empathy to apply it. A fairly typical narrative passage
in the latter reads: ‘And there were shoppers roundabout; women and weans,
a couple of prams with wee yins, all big-eye staring at him; then a sodjer was
here and trying not to but it looked like it was too much of an eVort and he
couldnay stop himself, he stuck the boot right in, into Sammy’s belly, then
another.’78 In a passage such as this, non-standard spellings evoke regional
pronunciations, word choice regional dialect more generally, and syntax the

77 All passages quoted from Lisa Woolley, American Voices of the Chicago Renaissance (Dekalb, IL:
Northern Illinois University Press, 2000), 24, 25, 23. It should be noted that this Dial is not the one
founded in 1840 in Boston by Ralph Waldo Emerson, Margaret Fuller, and others.
78 Kelman, How Late It Was, How Late (London: Secker and Warburg, 1994), 6.
 Narratives of Change 123

style of conversation. Unlike in Nicholas Nickleby, however, in How Late

It Was, How Late regionalisms mark not one or more characters within
the work but the work itself. Linguistic and social variation still correlate
with one another, but here it is the entire world of the novel, which includes
sociolinguistic stratiWcation among its non-standard speakers, that deviates
from sociolinguistic norms.
A work like Chinua Achebe’s Anthills of the Savannah deploys non-standard
language more sparingly—using allusions and lexicon to summon the exoti-
cism of culture and context without provocatively, even belligerently, chal-
lenging the comprehension of many readers: ‘Agwu does not call a meeting
to choose his seers and diviners and artists: Agwu, the god of healers; Agwu,
brother to Madness! But though born from the same womb he and Madness
were not created by the same chi.’79 Similarly, Patricia Grace, who is widely
praised for the distinctive Maori voice of her Wction, writes stories that are
imbued with Maori myths, culture and social practices, but creates her
distinctive voice through fairly restrained use of proper names and allusions:
‘Now they’ve come to this. It Wts, you see. We have a twenty percent Maori
here. We’ve tried to put programmes in place, kapa haka, etcetera, but it’s not
enough. No depth. We’re keen to have a whanau class, whether it be total
immersion or bilingual. With the children coming from Kohanga we’ll have
numbers enough, enough kids for an extra teacher.’80 More whimsically, in
Malcolm Pryce’s Wctional renderings of Aberystwyth, language variation
serves as one more marker of a post-modern world everywhere Wlled with
drollery and surrealism: ‘She switched to Welsh. ‘‘Beth ydych chi eisiau? Dydw
I ddim yn siarad Saesneg. . . .’’ I could speak in tongues, too. ‘‘Edrychwch
Hombré, agorwch y drws! por favor’’.’81 Writers like Grace may well seek to
undo social stereotypes with dignity, as some critics have argued, using
variation to eVect opposition to continued colonization or to enact cultural
hybridity.82 And Pryce may attempt much the same through humor. But the
linguistic stereotypes they use achieve much of their force through their
associations with the demeaning implications of earlier dialect-writing. And
like Dickens or Hardy, they cannot challenge these implications without
aYrming them. In all cases, social disruption still accompanies linguistic
variation.

79 Achebe, Anthills of the Savannah (New York: Anchor, 1987), 114–15.

80 Grace, Baby No-Eyes (Honolulu, HI: University of Hawaii Press, 1998), 146.
81 Pryce, Last Tango in Aberystwyth (London: Bloomsbury, 2003), 5.
82 e.g., Chadwick Allen, Blood Narrative: Indigenous Identity in American Indian and Maori Literary
and Activist Texts (Durham, NC: Duke University Press, 2002); and Evelyn Nien-Ming Ch’ien, Weird
English (Cambridge, MA: Harvard University Press, 2004).
124 Language Anxiety

The key point about such contemporary dialect-writing is this: while each of
these cases, to varying degrees, may aggressively challenge dominant culture
and its views on language, evoke the multilingual character of the Anglophonic
world, or celebrate diversity, they can do so only through explicit and self-
conscious recognition of change and variation. And such recognition again
fails to be politically neutral, for modern Wction in non-standard language
ultimately seeks political transformation as actively as did Zamenhof or
St Augustine. While Zamenhof advocated language as a means to foster
community, for example, Achebe and Grace see it as an expression of individ-
ual, resistant culture. Further, the same dialogue about variation that animates
Christian exegesis obtains here. If Dickens and The Dial, in their own very
diVerent ways, articulate sociolinguistic yearnings for the world of Genesis 11,
wherein communication was direct and complete, Kelman, Achebe, and Grace
aYrm variation with the competing narrative of Genesis 10.
Futuristic novels oVer their own versions of these competing narratives.
While language and its variations may be even more prominent in these kinds
of narrative than in those I have already considered, the utopian or dystopian
world of such books necessarily focuses on social concerns, thereby providing
a now familiar link between linguistic and disruptive extra-linguistic change.
Here, too, language variation and change can thus serve as symptoms or
concomitants of extra-linguistic ideas, practices, and institutions. An early
version of the use of change for this purpose occurs in Evelyn Waugh’s short
story ‘Out of Depth’, originally published in 1933. In it, a wealthy American,
through the magic of a doctor with the transparent surname of Kakophilos,
Wnds himself transported to London Wve centuries hence. In the area of
present-day Piccadilly Circus, by then reduced to a marsh with mud huts
and naked children, he asks whether he is still in London: ‘The men looked at
each other in surprise, and one very old white bear giggled slightly. After a
painful delay the leader nodded and said, ‘‘Lunnon’’.’ As the men squat to
discuss what to do with their captive, the narrator observes: ‘Occasional
phrases came to him, ‘‘white’’, ‘‘black boss’’, ‘‘trade’’, but for the most part
the jargon was without meaning.’ Adding to the extra-linguistic import of
Waugh’s story is the fact that in this futuristic world, the grunting, primitive
men are white, while the uniformed group who take the captive in trade, the
group who have won some sort of racial war, are black.83 In this vision of
the future, the collapse of culture, language, and racial hierarchies has been
simultaneous and complete.

83 Waugh, The Complete Stories of Evelyn Waugh (Boston: Little, Brown, and Co., 1998), 136.
 Narratives of Change 125

The Nadsat of Anthony Burgess’s A Clockwork Orange oVers perhaps the

best-known version of a sociolinguistically disintegrating world, in which the
slang of rebellious youths has combined with the Russian of England’s Cold
War enemy to produce a version of English, in structure and style, severed
from its cultural and historical roots and articulating violence, disillusion,
and boredom: ‘It was nadsats milking and coking and Wllying around (nadsats
were what we used to call the teens), but there were a few of the more starry
ones, vecks and cheenas alike (but not of the bourgeois, never them) laughing
and govoreeting at the bar.’84 A generation later, in describing a world reduced
by nuclear holocaust to primitive, warring bands, Russell Hoban’s Riddley
Walker similarly utilizes linguistic change to underscore individual and insti-
tutional devastation: ‘Peopl talk about the Cambry Pul theywl say any part of
Inland you myt be in youwl feal that pul to Cambry in the senter. May be its
jus in the air or may be where the Power Ring ben theres stil Power in the
groun.’85 In both works, as if in fulWllment of what twelfth-century Norman
chroniclers and Chaucer’s ‘Reeve’s Tale’ prophesy, language change is at once
sign, cause, and consequence of apocalyptic social upheaval.
From a utopian perspective in his 1933 The Shape of Things to Come, H. G.
Wells imagines a future in which a world-state will be created in 2059,
foreshadowed by the League of Nations and ending what he calls the Age of
Frustration. Among the concerns of the new world order is the elimination of
national and racial prejudice: ‘Next, a lingua-franca had to be made universal
and one or other of the great literature-bearing languages rendered accessible
to everyone.’ While writing of the future, Wells is necessarily a product of his
present, and to identify this lingua franca he turns to the then-popular eVorts
of C. K. Ogden to develop a simpliWed version of English—one with only 850
words, just 18 of which were verbs—that might facilitate the acquisition of
English as a second language and thus its spread around the globe. ‘One of the
unanticipated achievements of the twenty-Wrst century’, Wells thus later
observes, ‘was the diVusion of Basic English as the lingua franca of the
world and the even more rapid modiWcation, expansion and spread of English
in its wake.’ As with contemporary arguments for the spread of Basic English
as a global language, English of the future, to Wells, achieved its status not
through any language planning but simply by virtue of the fact that, com-
pared to other European languages, it ‘was simpler, subtler, more Xexible and
already more widely spoken’. A language academy in the form of the Diction-
ary Bureau solves the dilemma of Esperanto by monitoring further change,

84 Burgess, A Clockwork Orange (London: Penguin, 1972), 24.

85 Hoban, Riddley Walker: A Novel (New York: Summit Books, 1980), 106.
126 Language Anxiety

with the result that English of the twenty-Wrst century has far greater ‘delicacy
and precision of expression’ than did English of the eighteenth and nineteenth
centuries—though it’s still not perfect—and all speakers can now ‘understand
one another and they are all in one undivided cultural Weld’.86 Writing
between the world wars, Wells (like Zamenhof) imagined that language
would change to reXect the eradication of the nationalistic threats then
present in Europe. For Wells, Babel stands as a desirable and perhaps reach-
able future, while for Waugh, Burgess, and Hoban it seems to have receded,
regrettably, far into the past, leaving a future as bleak, scattered, and isolated
as the moment after the confusio linguarum. What all four writers share is
the desire for a world of total communication and, more importantly, a sense
that the structure of languages serves as a measure of how close or far away a
world of social and political stability may be.

Back to Babel
And so, metaphorically and conceptually, I come back to Babel. As an
explanation of language change and variation, the story of Nimrod’s Tower
has oVered, throughout the history of English, a powerful theoretical frame-
work for organizing conceptions of language and for understanding both the
past and the present. Indeed, perhaps the greatest testament to the explana-
tory power of this model is through its explicit extension into Christian
exegesis and the Enlightenment, as well as through the unacknowledged
justiWcation it provides for the stylistic choices of individual writers. The
competing treatments of these writers and, more generally, the competing
models oVered by Genesis 10 and 11 have made change and variation real in
White’s sense: they have made linguistic topics matter at more than simply an
intellectual level. As the Babel model has informed literary representations of
language across time, in fact, it has Wgured in extra-linguistic concerns as
disparate as medieval hagiography, early modern economics, nineteenth-
century immigration, and the rise of totalitarianism in the twentieth century.
From another, explanatory perspective that I have been sketching out so
far, the specter of Babel allows for the displacement of such concerns to
language. In this way, language change and variation serve as pre-emptive

86 Wells, The Shape of Things to Come (New York: Macmillan, 1933), 381, 415–18. A polymath, Wells
seems to have been particularly intrigued by language and its social implications. For discussion of the
variety of ways in which science Wction writers have responded to the issues of language change and
variation, whether in time-travel narratives or in narratives set in the future, see Walter E. Meyers,
Aliens and Linguists: Language Study and Science Fiction (Athens, GA: The University of Georgia Press,
1980), 12–37. Also see Bailey, Images of English, 215–35.
 Narratives of Change 127

issues, represented as more general and therefore more consequential; and

by substituting for issues of economics, race, nationalism, and so forth, they
in eVect preclude those issues. As the twenty-Wrst century looks to the future,
this dialogical fashion in which the narratives of Genesis 10 and 11 have shaped
the reality of change and variation and deWned the history of English con-
tinues to prevail in the world of social practices as well as in literature.
Language revivals exhibit the recognition, even celebration, of change and
variation as ordinary features of human language that Genesis 10 describes.
For certain languages, such as Cornish, Manx, and some American Indian
languages, the revival truly is just that—an attempt to make a living language
out of one that had reached a moment when no living native speakers
remained. In the case of Cornish, whose last native speaker is generally
thought to have died late in the eighteenth century, revival attempts began
almost immediately thereafter, drawing on the childhood memories of older
monoglot Anglophones, written records, and the comparative evidence of
genealogically aYliated languages like Breton. Less drastic situations have
involved languages that are still spoken but whose numbers of speakers and
domains have been severely reduced. Bosnian, Provençal, and Aramaic have
all been cultivated and reviviWed in the face of a culturally dominant language,
and in the ambit of English contact languages alone the list of revitalizations
(some still quite tenuous) includes Welsh, Irish, Yiddish, Maori, and various
Aboriginal languages. Whether a revival is complete or partial, all these cases
share something with each other besides the enabling notion that language
variation is a good, normal, and useful thing. And this is that the revival, like
the literary examples I considered above and like most acts of language
planning, responds to the eVects of largely extra-linguistic motivations in
politics, culture, ethnicity, and so forth. It was the eighteenth-century English
clearance of Highland farms, for example, that both displaced Gaelic speakers
and infused Anglophones to such an extent as to leave Gaelic today with a
small population of ageing speakers, just as it was English farming and
whaling in New Zealand that led to the reduction of Maori speakers by
approximately 75% during the nineteenth century.
By the same token, since revivals are predicated on the notion that language
is a vital and intrinsic part of culture, their ultimate goal is the emergence and
aYrmation of a culture and not simply a language associated with it. Revit-
alization of economically depressed Gaelic and Maori speech communities,
for instance, would redress a history of social oppression and reassert their
Wnancial and social viability. In this same vein, commenting on the cultiva-
tion of Bosnian as a distinct language beside Serbian and Croatian, the
director of education for the Bosnian National Council observed, ‘Language
128 Language Anxiety

deWnes the identity of a people. Having the Bosnian language brings recog-
nition to a people who have lived in Serbia and Montenegro for centuries.’87
It is this very reasoning that produced formal recognition of variation and
change—ratiWcation of Genesis 10, as it were—in the 1992 European Charter
for Regional or Minority Languages and the 1996 Universal Declaration of
Linguistic Rights, endorsed by PEN, UNESCO, and other linguistic organ-
izations. The former considers minority languages to be cultural wealth and
therefore as worthy of protection as are those who speak them, while the latter
begins with similar aYrmations of how the contexts of politics, geography,
and ideology contribute to a language’s identity. The Declaration goes on to
assert, among other things, the basic human right of individuals to retain
their language, the equality of all languages, regardless of their political status
in any one region, and the role of language in expressing individual cultural
views.88 As with the literary eVorts of Kelman and Welsh, such political
strategies and policies invest language change and variation with terriWc social
impact that, while self-aYrming to those who use a particular variety, is
necessarily meant as a challenge, even a threat, to those who do not.
The competing narrative of Genesis 11 underwrites its own competing
versions in narratives of the linguistic and social future. In some accounts, a
return to a single language of unmediated communication is every bit the
achievement that rendered Nimrod’s ambitions divine. English, more than
any other language, often bears the sobriquet ‘global language’ today, and
having already played this role in diplomacy, the language is increasingly
expanding into areas without large indigenous Anglophone populations.
While earlier generations in Mongolia cultivated Russian or Chinese as
second languages, for instance, English is today taking their place. ‘We see
English not only as a way of communicating’, observes Tsakhia Elbegdor,
the prime minister of Mongolia, ‘but as a way of opening windows on the
wider world.’89 This same empowering quality informs the embrace of English
as the dominant language of world telecommunication, now including the
Internet.90 Contesting narratives invoke not a putative communicative idyll
before Babel but rather the presumed oppression by which Nimrod made it
possible. As I indicated in the Wrst chapter, for many speakers worldwide,

87 Quoted in Nicholas Wood, ‘In the Old Dialect, a Balkan Region Regaining Identity’, New York
Times, 24 February 2005, A4.
88 See the ‘Universal Declaration of Linguistic Rights’ at [Link]
[Link].
89 Quoted in James Brooke, ‘For Mongolians, E is for English, F is for Future’, New York Times,
15 February 2005, A1.
90 See David Crystal, Language and the Internet (Cambridge: Cambridge University Press, 2001).
 Narratives of Change 129

particularly in Europe, the role of English in rebuilding Nimrod’s Tower is a

sinister one indeed. France, Germany, Russia, and Brazil have all taken steps
to curtail the reduction of linguistic variation through the spread of English,
whether these be formal, legislative eVorts to limit the inXux of English
vocabulary or informal ones to discourage its presence on shirts and in
advertising.
As with linguistic revivals, these various uses of the Babel model ultimately
serve non-linguistic ends. In advocating the diVusion of English as a second
language, thus, the president of Chile observed, ‘As a country, we want to be a
bridge and a platform for Xows of international trade and in the Asia-PaciWc
region’, while the rector of one of the country’s largest universities, in explain-
ing the introduction of a requirement for all students to study English, noted,
‘Our mission is to train professionals for an internationalized world, and
this is the only way for this country to develop the way it wants’.91 From the
very diVerent perspective of a native Anglophone, Arthur Schlesinger,
Jr., protests what he calls the ‘cult of ethnicity’ in modern America but
likewise uses the Babel model for non-linguistic purposes. Mixing literary
and biblical allusions with emotive metaphors, he bluntly warns of what
will happen if variation is allowed to proceed unchecked: ‘Will the center
hold? Or will the melting pot give way to the Tower of Babel?’92
In the end, narratives of change and variation function much like the found
facts of co-varying phonological shifts. Bringing together institutional and
conceptual impulses into meaningful patterns, the explanations they provide
become templates for organizing and interpreting linguistic phenomena.
Out of their competition—out of the disjunction between narratives that
grant the human origins of linguistic diversity and those that aYrm its
unearthly status—they make these phenomena real. They have also made
them, in the case of English, the conduit for anxieties over the speakers of
particular varieties, the economic and ethnic consequences of their presence
in a community, the character of a national population, and the direction of a
country’s future. When we tell stories about change and variation, we talk
about something far more consequential than language alone: we talk about
ourselves.

91 Quoted in Larry Rohter, ‘Learn English, Says Chile, Thinking Upwardly Global’, New York
Times, 29 December 2004, A4.
92 Schlesinger, The Disuniting of America: ReXections on a Multicultural Society, rev. edn. (New York:
W. W. Norton, 1998), 22.
 4

Policy and Politics

Victimized by sound change

In the Wrst chapter, when discussing theories of language change, I quoted
Roger Lass’s characteristically witty and insightful observation that linguistic
change ‘occurs over ‘‘geological’’ time, beyond the capacity of humans to act,
since no actor can see the consequences of his actions. A speaker engaged in a
change is not an agent, but a victim.’1 Lass’s perspective is that of a historical
linguist interested in structural changes like consonant shifts and syntactic
elaboration, and from this perspective ‘geological time’ is indeed an apt
metaphor. Including the exceptions accounted for in Verner’s Law, for ex-
ample, the consonantal changes designated as the First Consonant Shift
probably required at least half a millennium for completion, while the
expansion of English periphrastic verb phrases, whereby modals like should
came to create progressive structures like should have been going, took place
over several centuries. Any one speaker at any one moment in either change
might well be aware of the existence of variation, with certain eighteenth-
century speakers already saying something like the house is being built, even
as others retained the historical the house is building or the house is on building.
Speakers might associate variants with particular social groups, or know
that they are socially stigmatized, or suspect, like H. G. Wells and many of
the verbal hygienists described by Deborah Cameron, that they can extrapo-
late the linguistic future, whether utopian or dystopian, from what they hear
around them. And after the fact, speakers certainly can look back, identify the
one variant that would prevail, and reconstruct its development as inevitable.
But in Lass’s critique, such linguistic foresight is illusory. All linguistic
history is written retrospectively, and it all therefore beneWts from the possi-
bility of identifying patterns and developments among phenomena that

1 Lass, Historical Linguistics and Language Change (Cambridge: Cambridge University Press,
1997), 367.
 Policy and Politics 131

transpired in a non-teleological fashion. Typically, I have maintained, lan-

guage change merely happens, since a speciWc use of language is only rarely
directed to produce an identiWed result, even if that result can later be
connected to it. And given competing linguistic forces (such as phonetic
assimilation and dissimilation) and the vagaries of population movement,
individuals living during a ‘geological’ sound change have no way of knowing
its outcome. In fact, it might well be diYcult for them even to recognize a
change as change, both because not all synchronic variation produces dia-
chronic change and because, as sociolinguistic interviews often show, speakers
can be poor, even unaware, witnesses of the language they use. Indeed, the
Northern Cities Vowel Shift in the United States illustrates how the ‘geo-
logical’ progress of change can sometimes render a change elusive to the very
speakers who propagate it. In this change, which has been documented for
several decades, lax vowels are participating in a small co-varying shift
whereby [O], for instance, has moved down to [a] and [æ] up to [E], so that
caught is pronounced cot ([kat]) and cat has the same vowel as kept ([kEt]).2
As long-standing and widespread as the change is, it seems to pass unnoticed
by the majority of northern United States speakers, including my students,
who regularly deny that the Northern Cities Vowel Shift exists, even as they
demonstrate it in their denials. And to the extent that speakers are unaware
of a change in progress, they obviously cannot intentionally further it.
For such speakers, the existence and progress of the Shift can be identiWed
only retrospectively.
Such after-the-fact identiWcation of a change’s direction is readily apparent
in structural issues, as when critics identify the origins of periphrastic verbs in
the gradual restriction of the verb will from a word meaning ‘to wish’ to a
grammatical marker of futurity and in the extrapolation from this restriction
to similar uses of modal verbs (e.g., shall and must). But the retrospective
deWnition of coherence and teleology can occur with sociolinguistic issues
as well. Recent discussions of Middle English, for instance, have used post-
medieval sociolinguistic developments in literature, nationalism, and tech-
nology in order to project into the Middle Ages notions of Standard English
and a resistant vernacular culture, which have then been understood to
constitute the pre-eminent and inevitable trend in the late-medieval English
linguistic repertoire. At the very moment Chaucer is sometimes credited with
creating and fostering a modern notion of English literature and language,
however, Latin still dominated ecclesiastical domains, and French legal and

2 See William Labov, Principles of Linguistic Change, i, Internal Factors (Oxford: Blackwell, 1994),
177–201.
132 Language Anxiety

aristocratic ones. Even within the repertoire of English alone, varieties besides
the southeast one Chaucer used (such as the northwest variety that Wgures in
many late-medieval alliterative poems, like Sir Gawain and the Green Knight)
furthered their own literary traditions and left it uncertain, at least prior to
the late-Wfteenth-century emergence of print, from which region a standard
might develop—if, in fact, any would.3
Noah Webster’s 1789 Dissertations on the English Language provides a more
recent example of how linguistic history can be appropriated—maybe even
created—to serve the sociolinguistic present. As I will argue later in this
chapter, Webster certainly had a well-formed sense of a strong central gov-
ernment reXected in a centralized and stable linguistic tradition, and in the
Dissertations he thus observes that a ‘national language is a band of national
union’. But while he gave voice to a forceful and developing argument in early
America, his was only one, competing view at the time. Other critics, such as
Benjamin Franklin, put a priority on restricting the number of German
speakers in the colonies, while still others, like John Adams, championed
the creation of an American language academy. Webster could not have
imagined two hundred years of American sociolinguistic history and how
events would advance his views over those of Franklin and Adams: the
extensive European emigration of a century later, the nationalizing eVects of
the First World War, the economic hardships leading up to the Second World
War, the subsequent rise in emigration from Latin America and Asia,
or, Wnally, the Civil Rights movement of the 1960s and its backlash. Not
knowing any of this, Webster could not have seen how race and religion
came to propel nativism in times of economic diYculty and how in this
context language legislation has developed as a way to control immigration.
Nor could Webster have seen himself (or his views) as an agent propelling the
United States to a moment when the 1981 Select Committee on Immigration
and Refugee Policy would quote his comments on a national language within
its proposal to retain command of English as a requirement for naturaliza-
tion.4 Webster’s speciWcally eighteenth-century views on the role of language
in a new republic had become evidence for historical continuity and modern
immigration policy.

3 See further Machan, ‘Politics and the Middle English Language’, Studies in the Age of Chaucer, 24
(2002), 317–24.
4 Webster, Dissertations on the English Language (1789; rpt. Menston: Scolar Press, 1967), 397; James
Crawford (ed.), Language Loyalties: A Source Book on the OYcial English Controversy (Chicago, IL:
University of Chicago Press, 1992), 101. On language legislation and early American immigration, see
John Higham, Strangers in the Land: Patterns of American Nativism, 1860–1925, 2nd edn. (New
Brunswick, NJ: Rutgers University Press, 1988).
 Policy and Politics 133

Relying on the retrospection of linguistic history, speakers not only con-

struct the linguistic past but also, like Cameron’s verbal hygienists, sometimes
believe that they can know the linguistic future and can take action to achieve
or prevent that future. But while speakers certainly can foster the development
of a future they already see as given, evidence against their ability to make
accurate linguistic predictions, particularly in sociolinguistic areas, is consid-
erable. In 1600, prior to the British settlement at Jamestown, Virginia, no
speaker would have had any reason to imagine the growth of the United
States—which wouldn’t even exist for over another century and a half—
much less the impact its use of English would have on the structure and global
dissemination of the language. At that same movement, well before the
Highland Clearances and the mining of coal to sustain the Industrial Revolu-
tion, neither would any speaker have reason to imagine the overwhelming shift
of monoglot Gaelic and Welsh speakers to English in the centuries to come.
From the opposite perspective, wildly errant predictions also demonstrate
the limitations on foresight into the linguistic future. In his Dissertations, for
example, Webster conWdently supposed that the birth of the United States,
coupled with the new associations of peoples and ideas that it entailed, would
‘produce, in a course of time, a language in North America, as diVerent from
the future language of England, as the modern Dutch, Danish and Swedish are
from the German, or from one another’.5 Even allowing for the vagueness of
‘in a course of time’, and Oscar Wilde notwithstanding, this has not happened.
Similarly erroneous and even more speciWc was Henry Sweet’s 1877 assertion
that within a century ‘England, America, and Australia will be speaking
mutually unintelligible languages, owing to their independent changes of
pronunciation’.6
The limitations of linguistic foresight and its implications for agency
appear in still smaller time frames. According to the 1940 United States census,
from a white population of 118,214,870, some 1,858,024 individuals (1.6%)
spoke Spanish as a mother tongue. By contrast, the 2000 census records
254,762,734 United States residents over the age of Wve, 26,771,035 (or 10.5%)
of whom speak Spanish to some degree.7 To foresee this increase in the
number of Spanish speakers and their percentage among the general American

5 Webster, Dissertations on the English Language, 22–3.

6 Sweet, A Handbook of Phonetics (Oxford: Clarendon Press, 1877), 196.
7 Campbell Gibson and Kay Jung, ‘Historical Census Statistics on Population Totals by Race, 1790
to 1990, and by Hispanic Origin, 1970 to 1990, for the United States, Regions, Divisions, and States’,
Population Division, US Census Bureau, September 2002, Working Paper Series No. 56, Table E-6;
United States Census 2000, Summary Tables on Language Use and English Ability. Information can be
accessed at [Link] Comparisons between censuses are made diYcult by the
fact that over time the categorization of factors like age and race have changed.
134 Language Anxiety

population, an analyst of 1940 would have needed to know, presciently, many

developments of the following decades. Foremost of these are the Civil Rights
movement, the elimination of immigration quotas in 1965, and rising emigra-
tion from Latin America since the 1960s. No analyst of 1940 could have used
only contemporary data like birth rates, housing patterns, or educational
backgrounds to predict a Wvefold increase in the percentage of Spanish
speakers by the end of the millennium. So, too, is it impossible today to predict
either the emergence of Spanish as co-equal to English in the United States
(as many do). Sociolinguistic patterns like those that contributed to the rise
of Spanish—and similarly not now operative—could well lead to that lan-
guage’s contraction or the expansion of another language, such as Chinese,
which in the 2000 census had 2,022,143 speakers over the age of Wve.
It is in these senses, then, that speakers serve as victims to the structural and
sociolinguistic changes they further. They articulate particular vowels, syntactic
structures, or, simply, their native language, but they do so for immediate
reasons: to buy lunch, express opinions, and explain procedures. Only in very
limited ways might speakers be seen—and see themselves—as agents who speak,
in the Wrst instance, to eVect diachronic change and not to accomplish limited
pragmatic goals. Such agency seems restricted to lexicon, and then to the
stigmatization or amelioration of individual words, such as Miss and gay. At
the levels of syntax and phonology, targeted change has not been a signiWcant
historical factor for English. And larger sociolinguistic changes of the kind I
described for the global expansion of English and the American expansion of
Spanish necessarily transcend the eVorts and control of any one individual.
Yet if speakers are victims of changes in the language they speak, the
language itself, both in structure and pragmatics, is certainly changed by
them. Structurally, it is of course speakers’ use of particular forms that
maintains the forms’ currency and renders them as historical substitutes for
competing forms. While no Wfteenth-century speakers may have viewed their
selection of the verbal inXection [s] rather than [E‚] as part of a diachronic
process that would culminate in replacement of the latter by the former, the
fact that they, their children, and their children’s children did use the [s]
inXection constituted that replacement. Their usage is why most Anglo-
phones now say walks rather than walketh. In general, pragmatic terms,
these speakers’ simple use of English, whether by conscious decision or not,
sustained the language’s viability, as even more so did other individuals’ use of
written English to conduct law courts, teach school, and transact all govern-
ment and Wnancial business. To be sure, while such usage enhances the long-
term status of a language in terms of its numbers of speakers and domains of
usage, other kinds can produce long-term constriction. An old schoolchild
 Policy and Politics 135

rhyme begins ‘Latin is a language, as dead as it can be’, and the reasons it is so
(to the extent that it is so) are that its speakers shifted to other languages
(including developments of Latin) and that its domains have become increas-
ingly limited, now primarily to the classroom. One can continue to tinker
with grammatical and discursive niceties for a language which is dead in
this sense, though such tinkering is remote from the processes of language
change that I described in the second chapter. Without speakers, there can
be no changes of this kind.
So speakers are changed by language, but it, too, is changed by them,
and this complexity in the agency of linguistic change appears with particular
force in religious as well as civic programs. Policies and legislation on doc-
trine, education, immigration, voting, and legal process can all shape the
opportunities to use one language or another and thus not only that lan-
guage’s viability but also, at least indirectly, its grammatical structures. At the
same time, grammatical and sociolinguistic issues often nominally motivate
such policies in the Wrst place. Factors like the number of speakers of
a minority language, their rate of immigration, the structural impact of
their native language on English, and their presence in schools and the
workforce can all lead to government action that can enhance or curtail
these same factors. Language change is thus an ambivalent phenomenon,
in the sense that even as a language changes, so, too, can speakers uninten-
tionally or indirectly change the language.
There is another kind of ambivalence operative here, however. That is, if
religious and civic policies guide the language use of individuals, the policies
themselves are ultimately the creation of individuals. To be sure, any one
speaker (or voter) may have little genuine inXuence on or sympathy with
speciWc acts of public policy, whether they are the declaration of an oYcial
language, the proscription of immigrant languages, or the mandate of court
interpreters. But short of a totalitarian state, and no Anglophone country
merits this label, policy oYcials do necessarily emerge from the groups they
govern and they do necessarily respond to the concerns of at least some
individuals from those groups. Stated more forcefully, all speakers are
invested in linguistic policy decisions, and all speakers have responsibility
for them. The alternative approach, advocated by many critics of global
English in particular, conceives a kind of agentless linguistic policy—or at
least a policy that is the agent of a limited, malfeasant group—that overrides
the wishes of both Anglophones and the non-Anglophones they encounter.
Robert Phillipson thus imagines English ‘as a kind of linguistic cuckoo, taking
over where other breeds of language have historically nested and acquired
territorial rights, and obliging non-native speakers of English to acquire the
136 Language Anxiety

behavioural habits and linguistic forms of English’. And Alastair Pennycook

has seen such predatory processes as inherent in the post-colonial expansion
of English, though again in ways that represent language as somehow a
wilful agent acting irrespectively of at least some of its speakers’ desires:
‘If one of the central aspects of colonial discourse has been to construct the
native Other as backward, dirty, primitive, depraved, childlike, feminine,
and so forth, the other side of this discourse has been the construction of
the colonizers, their language, culture and political structures as advanced,
superior, modern, civilized, masculine, mature and so on.’8 Much like the
embrace of global English that it opposes, this approach contributes to the
very anxiety that this book discusses, for it renders speakers the victims of
policy as well as sound change.
In this chapter I look selectively at how particular public policies have
aVected English change and variation, and, in turn, at how the use of English
and other languages has driven these same policies. Adhering to the approach
I have followed so far, my methods are neither exhaustive nor reductive. In its
1500-year history that now includes the entire globe, the English language and
any number of regional or national policies have mutually shaped one
another in signiWcant ways. By the same token, a good deal of public policy
has not aVected language change, even indirectly, and some kinds of linguistic
policy have produced or been associated with only minimal anxiety about
language change. As Braj Kachru suggests, it may be the case that once Lord
Macaulay’s educational minute of 2 February 1835 was passed, ‘the subcon-
tinent was not the same, linguistically and educationally’. But the same
cataclysmic signiWcance cannot be attributed to the imposition of an Eng-
lish-only educational policy in Puerto Rico following the Spanish-American
War; Spanish was reinstated in classrooms in 1952, and to this day over 95% of
the population over the age of Wve speak Spanish as their dominant language.9
Along with the mutual eVects language change and policy have on each other,
such varying contextual dimensions contribute to the diYculty of identifying
clear cause-and-eVects relationships between language change and various
kinds of religious, social, and political change. By Kachru’s analysis, for
example, Macaulay’s minute might indeed be argued to have produced social
change. But it also could merely have portended subsequent administrative

8 Phillipson, English-Only Europe? Challenging Language Policy (London: Routledge, 2003), 4;

Pennycook, English and the Discourses of Colonialism (London: Routledge, 1998), 129.
9 Kachru, ‘English in South Asia’ and John A. Holm, ‘English in the Caribbean’, in The Cambridge
History of the English Language, v, Robert BurchWeld (ed.), English in Britain and Overseas: Origins and
Development (Cambridge: Cambridge University Press, 1994), 501 and 354; Selected Social Character-
istics in Puerto Rico ([Link]
 Policy and Politics 137

activities of greater sociolinguistic consequence (such as the founding of

universities in Bombay, Calcutta, and Madras), or itself been the inevitable
consequence of the revitalization of missionary activities, including the in-
struction of English, that followed William Wilberforce’s 1813 declaration to
Parliament that it should exchange India’s ‘dark and bloody superstition for
the genial inXuence of Christian light and truth’.10 These kinds of causal
ambiguities contribute their own complications to discussions of language,
and it is in this context that I want to explore how religious and governmental
policies have fostered, channeled, and even created anxiety about language
change and variation in particular.

Language planning and the word of God

Like nearly every one of the roughly 500 Amerindian languages spoken in
North America before the arrival of European colonists, Passamaquoddy
suVered badly from contact with French, Dutch, Spanish, German, and
English. Today, it has only about 600 speakers in Maine and nearby parts
of Canada, and most of these, as is commonly the case with dying languages,
are elderly; they are all bilingual with English. In response to such situations,
tribes have adopted diVering policies to increase a language’s number of
speakers and domains. One group inventively pursued the strategy of having
the Disney cartoon Bambi dubbed in Arapaho, so that pre-school children
could be exposed to the language in an entertaining and playful context.11
Allen Sockabasin, of the Passamaquoddy tribe, has explored an equally
inventive policy. Drawing on connections between language, community,
and religion, he has translated English songs, poems, and prayers, including
the rosary, into Passamaquoddy. Sockabasin has also had his translations
professionally recorded and issued on CDs, which he distributes for free.
Whether such recordings are suYcient to revitalize the language may be
doubtful, but they do forcefully point to how language planning can make
use of the emotional ties speakers have for particular domains of language
use. ‘I know when I say ‘‘my creator’’ in my language’, Sockabasin observes,
‘there is no other deWnition. It’s who made me.’12

10 Quoted in Kachru, ‘English in South Asia’, 504.

11 Stephen Greymorning, ‘ReXections on the Arapaho Language Project, or When Bambi Spoke
Arapaho and Other Tales of Arapaho Language Revitalization EVorts’, in Leanne Hinton and Ken Hale
(eds.), The Green Book of Language Revitalization in Practice (San Diego, CA: Academic Press, 2001),
287–97.
12 Quoted in Katie Zezima, ‘One Man’s Goal: For a Tribe to Pray in Its Own Language’, New York
Times, 15 November 2003, A11.
138 Language Anxiety

Language planning has two general impulses: corpus planning, which

addresses speciWcs of usage and grammar and therefore typically takes the
form of grammar books, dictionaries, and rhetorics; and status planning,
which attempts to adjust the relative social functions of a given variety in
relation to other varieties and which subsumes Sockabasin’s eVorts on behalf
of Passamaquoddy. In either case, language planning characteristically has
objectives beyond simple communication or the reWnement of a linguistic
code. These objectives may be presumptively ethical ones, such as the protec-
tion of a minority group’s linguistic rights, and such protection typically
depends on state or institutional support of some kind. Besides having eleven
oYcial languages and translating its documents into several more, for ex-
ample, the European Union actively fosters the status of minority languages by
allocating funds towards the maintenance of linguistic and ethnic diversity.13
But it can also be possible to realize such objectives without the assistance of
a Wnancial and bureaucratic infrastructure. Thus, the constriction of sexist
usage in many Anglophone countries has occurred largely under a popular
initiative, as has the ameliorization of much sexist and racist terminology.
The kinds of social objectives that language planning characteristically
serves bespeak other initiatives than group or individual civil rights, however.
As Robert Cooper describes these initiatives, ‘Language planning is typically
carried out for the attainment of nonlinguistic ends such as consumer
protection, scientiWc exchange, national integration, political control, eco-
nomic development, the creation of new elites or the maintenance of old
ones, the paciWcation or cooption of minority groups, and mass mobilization
of national or political movements’.14 These are the initiatives of dominant,
ruling groups, be they political, economic, or religious, that manage areas
like ‘consumer protection’ and ‘scientiWc exchange’ and that are also the
groups likely to have the Wnancial and administrative means for enforcing
language policy. In the modern world, the most typical kind of language
planning, whether of corpus or status, involves government-funded initiatives

13 Phillipson, English-Only Europe?, 10. This policy has not been without controversy, on which see
John Tagliabue, ‘Soon, Europe Will Speak in 23 Tongues’, New York Times, 6 December 2006, A10. On
language planning generally, see Robert L. Cooper (ed.), Language Spread: Studies in DiVusion and Social
Change (Bloomington, IN: Indiana University Press, 1982); Joshua Fishman (ed.), Advances in Language
Planning (The Hague: Mouton, 1974); James W. Tollefson, Planning Language, Planning Inequality:
Language Policy in the Community (London: Longman, 1991); and Robert Hodge and Gunther Kress,
Language as Ideology, 2nd edn. (London: Routledge, 1993). On the ethics of language planning, see Juan
Cobarrubias, ‘Ethical Issues in Status Planning’, in Cobarrubias and Joshua A. Fishman (eds.), Progress in
Language Planning: International Perspectives (The Hague: Mouton, 1983), 41–85.
14 Robert L. Cooper, Language Planning and Social Change (Cambridge: Cambridge University
Press, 1989), 35.
 Policy and Politics 139

implemented through educational programs. As the prerogatives of an elite,

then, the initiatives and programs of language planning can reproduce ideol-
ogy as much as they realize putatively linguistic goals.
The teaching of English as a second language is a case in point. This is
currently an extremely lucrative, global activity for both the United States and
Great Britain that has a presumptively linguistic objective and that achieves
varying success in varying environments. To the extent that they render
language a purchasable commodity and testify for the socioeconomic super-
iority and desirability of their teachers’ homelands, however, ESL and TOEFL
initiatives are also sometimes understood to inculcate Western economic
values along with phonology, morphology, and syntax.15 Pointing to issues
such as the limited availability and success of ESL initiatives, some critics even
argue that inadequate command of English as a second language reXects
neither speakers’ abilities nor their motivations. Rather, these critics trace
inadequate language learning to policies that simultaneously cultivate a need
and a desire to learn English but also limit speakers’ opportunities to learn
eVectively. By manipulating the availability of English on the world linguistic
market, such cultivation is thought to sustain a sociopolitical hierarchy with
Anglophone nations at the top. When ESL programs emphasize the acquisi-
tion of language meant to apologize for immigrant language diYculties—
which in some instances are eVectively dialect diVerences—even the details of
the programs can appear as complicit in such sociolinguistic policy.16
Whether economic values in such programs are incidental or primary in
relation to grammatical ones is of course an arguable point, and many
individuals might well reject claims that second language acquisition by its
very nature reproduces a hierarchy that disadvantages those in the learning
groups. But it’s not necessary to make claims as broad as these in order to
recognize that language planning is implicated in social policy and, therefore,
that language has the potential to be a socially meaningful symbol and force—
like clothing or music or hairstyles—that can further the objectives of a ruling
or aspiring elite. This is the case, indeed, with the many examples I consider in
this chapter but also with a variety of non-Anglophone examples. Early in the
seventeenth century, for example, Cardinal Richelieu engaged in language
planning by establishing the Académie Française, whose eVorts to stabilize
grammar, rhetoric, and vocabulary would serve as the linguistic counterparts
to more general governmental eVorts to stabilize society.17 Social policy of an

15 Robert Phillipson, Linguistic Imperialism (Oxford: Oxford University Press, 1992); Alastair
Pennycook, The Cultural Politics of English as an International Language (London: Longman, 1994).
16 James W. Tollefson, Planning Language, Planning Inequality, 7, 104–35.
17 Cooper, Language Planning and Social Change, 3–11.
140 Language Anxiety

aspirant group likewise underwrote Ivar Aasen’s nineteenth-century eVorts

to deWne a new Norwegian, or Nynorsk, as in part a nationalistic testament to
cultural independence from Sweden, as well as Elias Lönnrot’s collection and
composition of the Kalevala poems in expression of the integrity of Finnish
culture even under Tsarist rule. In both of these examples, a linguistic code
mediated not simply social policy but national identity too, and this may well
be one of the pre-eminent characteristics of language planning in the modern
era.18 Still other examples show language furthering one social identity or
repressing another: Stalin’s imposition of the Cyrillic alphabet on the nascent
Soviet republics; the role of Tagalog in various Philippine nationalistic eVorts;
semantic debates over the character and integrity of Serbian, Croatian, and
Bosnian languages in the former Yugoslavia; and the status of English as a
tool of Western capitalism for groups in Iran, the Philippines, and China
who are seeking closer or looser aYliations with the West.19
As a response to change and variation, language planning for religion
and theology, in the broader history of English, has both facilitated shift to
English and forestalled it. For Passamaquoddy, religion actually motivated the
language’s contraction as well as its revival: it was the decision of Jesuit
missionaries to use French and Latin that initiated shift in the Passama-
quoddy community, just as it is Sockabasin’s decision to translate prayers
into Passamaquoddy that injects vitality into it. At essentially the same time as
Christian missionary work contributed to the demise of Passamaquoddy,
however, language planning in the service of religion helped to forestall shift
to English in various Celtic regions. The early identiWcation of Welsh and
Scots Gaelic with, respectively, Methodism and Presbyterianism did much to
sustain the languages for their speakers during early modern social and
economic upheaval, while, more pragmatically, the 1588 publication of a
vernacular Bible provided symbolic and material support for Welsh in par-
ticular. But if in these instances, religious language planning helped forestall
shift and its concomitant disruption of culture and tradition, when the shift
did occur, mostly in the nineteenth century, economics proved to be the
stronger determinant by producing among both Anglophones and non-
Anglophones the sense that variation from English was debilitating for the
latter. This conjunction between religion, language planning, and economic
expansion proved even stronger in Australia, south Asia, the antipodes, and
South America, where planning led to the abandonment of indigenous

18 Cf. Cooper, Language Planning and Social Change, 69, and Einar Haugen, Language ConXict and
Language Planning: The Case of Modern Norwegian (Cambridge, MA: Harvard University Press, 1966).
19 See Tollefson, Planning Language, Planning Inequality, and Tove Skutnabb-Kangas and Robert
Phillipson (eds.), Linguistic Human Rights: Overcoming Linguistic Discrimination (Berlin: Mouton de
Gruyter, 1994).
 Policy and Politics 141

languages by non-Anglophones and also Anglophones’ encouragement of a

shift to English. The same, basic language plan obtained in each case: to
convert indigenes, theologically if not also economically, by means of a
language shift accomplished through education and the translation of the
Bible and prayers into native languages. The thinking went that without such
a shift, divergence between Anglophones and indigenous cultures would
persist, to the putative detriment of both in various ways: Anglophones
would lose opportunities for the conversion of non-believers as well as
economic growth, and non-Anglophones would lose the promise of partici-
pation in the culture shared by the world’s dominant groups. Through shift,
however, linguistic, economic, and cultural variation would all (hypothetic-
ally) diminish together. As at Babel, language diversity thereby generated
anxiety, and this kind of anxiety, channeled through religious policy, has
been present in some form since some of the earliest days of English.
During the late ninth-century reign of King Alfred the Great—to begin
with an early example—England began to emerge from a half century of
depredations and colonizing eVorts by mostly Danish Vikings. Beyond what-
ever the weaknesses of defense and political organization may have done to
invite and enable the raids, responsibility for them was popularly understood
to lie in the moral failures of the Anglo-Saxons, just as Gildas had earlier
attributed the Angles’ and Saxons’ invasion to the similar failures of the
British and as later Norman historians would explain their own invasion as
God’s scourge on the laxity of Anglo-Saxon England. When Viking raids
resumed with even greater ferocity a century later, Wulfstan, bishop of
Worcester, voiced this characteristically Augustinian historiographic view-
point as a context for his explanation of the state of Anglo-Saxon England:
‘Beloved men, recognize what the truth is: this world is in haste and it is
drawing near the end, and therefore the longer it is the worse it will get in
the world. And it needs must thus become very much worse as a result of
the people’s sins prior to the advent of Antichrist; and then, indeed, it will be
terrible and cruel throughout the world.’20 It was Alfred’s reign that was
largely responsible for England’s earlier civic and moral emergence, and it
was he who had rallied the English forces in a bleak winter of 878–9 and
gradually reasserted Anglo-Saxon control over lands that the Danes had
occupied. By 886 he had achieved a kind of political balance through the
Treaty of Wedmore, distinguishing the lands that he controlled from those
in the Danelaw, which remained in Viking power. If England as a political
entity still lay well in the future, Alfred’s military and political eVorts had

20 Michael Swanton (ed. and trans.), Anglo-Saxon Prose (London: Dent, 1975), 116–17.
142 Language Anxiety

done much to foster a centralized government with its own courts, law
codes, and text production.
In this dynamic context of social disorder and organization, the king
shifted his focus from social to linguistic matters, speciWcally to the state of
learning in England, the demise of language use, and the possibilities that
translation aVorded for a return to stability. Alfred himself began this return
through his translation of Gregory the Great’s pastoral handbook Cura
Pastoralis, to which he prefaced a survey of English learning. There, thinking
back to his youth in a time before Viking raids intensiWed in the 850s, Alfred
remembers a world of secular and religious scholars, devout kings, righteous
service, and a love of wisdom. He looks around and sees that this world has
passed, replaced by one in which few people south of the Humber river and
even fewer north of it can understand the divine service in English or translate
a letter from Latin to English. He sees burnt and desolate churches—once
Wlled with books and treasures—and an ignorant populace, unable to read
the books that would aid them. In Alfred’s voicing, this populace shares his
trauma, his recognition that they have been severed from the past that would
give their present meaning: ‘Our forefathers who formerly held these places
loved knowledge, and through it they acquired wealth and left if to us. One
can see their footprints here still, but we cannot follow after them and
therefore we have now lost both the wealth and the knowledge because we
would not bend our mind to that course.’ History is here a path, and success,
intellectual as well as material, lies in retracing and continuing the steps of
tradition. More speciWcally, history is a path of books, now lost because the
books, written in Latin, are illegible. And as depraved as the English may have
been to invite the Viking depredations, the lapse in linguistic memory that
renders the past unreadable may be the greater failure. Having asked himself
why his predecessors did not translate books from Latin into English, Alfred
realizes just how far the present has fallen linguistically as well as morally:
‘They did not imagine that men should ever become so careless and learning
so decayed.’ The king’s remedy for this situation is predictably, even neces-
sarily, linguistic. Describing the passage of Holy Writ and exegesis from the
Hebrews to the Greeks and then to the Romans, he proposes that these same
books now be translated into English. To provide an institutional context
for the production and distribution of these translations, he further proposes
that free-born youth be taught to read English: ‘afterwards one may teach
further in the Latin language those whom one wishes to teach further and
wishes to promote to holy orders’.21

21 Swanton (ed. and trans.), Anglo-Saxon Prose, 31–2.

 Policy and Politics 143

There is far more than nostalgia here. Alfred’s policy collapses morality,
history, and language, with a failure in one mirroring lapses in the others.
While cause and eVect may be moot, since variations in all three necessarily
co-occur, language (he claims) provides the means to restore stability every-
where. Just as the loss of treasure and wisdom follows upon shifts away
from Latin and literacy, so Alfred’s program of translation and education
will serve as the means to and conWrmation of England’s political and moral
pre-eminence. Through his rationalization of translation, moreover, the king
situates England in the translatio studii and imperii—in a cultural tradition
that renders Winchester the Rome or Athens of the early Middle Ages.
Alfred’s is thus a nuanced approach to anxiety about language change and
variation. Such change may indeed be the origin and symbol of social
dissolution. But if the English people can reassert control over language,
eliminating variation and restoring England’s linguistic practice to an im-
mutable moment from the past, the resulting stability will advance the
country beyond where it had been. In this way, Alfred evokes not simply
the Tower of Babel but the pre-lapsarian Garden of Eden—metaphors of
a cultural achievement that, he implies, is achievable by means of language
management. If Alfred’s policies failed to reach this goal, they did much to
suspend the variation and change that the king laments and further linguistic
achievement in England, with the educational program he described appar-
ently extended in several foundational Christian works that joined the
Pastoral Care in English translation at about this time: Gregory’s Dialogues,
Orosius’s History of the World, Bede’s Ecclesiastical History, Boethius’s
Consolation of Philosophy, and St Augustine’s Soliloquies. Although the re-
sumption of Viking raids in the late tenth century negated many of Alfred’s
political and literary achievements, the corpus planning in which he engaged
was thus clearly successful in the short term. Representing language change
through a heuristic that was religious as well as political, Alfred’s policy
channeled together recognition of the devastation wrought by the Vikings—
the hand of God in earthly aVairs—and the power of language in order both
to reXect these conditions and to provide a means of redress. The circum-
stances that led to Alfred’s program may indeed have been desperate, but
there is nothing urgent in his calm assertion about the long-term stabilizing
potential of language, once it is removed from the diurnal world of change.
But the fact that King Alfred did not permanently restore English culture
through language planning on variation and change bears further consider-
ation. When Viking raids resumed, they did so after Alfred’s program of
translation and education, after the Benedictine Reform these helped under-
write, and after the emergence of Winchester as a seat of learning and center
144 Language Anxiety

for the production of the West Saxon koine in which many Old English
texts survive. All these achievements would seem to have accomplished
much of the restoration that Alfred envisioned, and so the success of the
raids undermines the very claim that language change can be either cause of
or solution to larger social disturbance. Indeed, a century later the Normans
and Angevins rehearsed this same thinking and achieved much the same
result. Through the Conquest, they reconWgured England’s linguistic reper-
toire by introducing French, cultivating diglossia, and at the very least
neglecting (if not repressing) production of manuscripts in English. Whatever
success the Normans had in stabilizing the country by stabilizing linguistic
variation was as short-lived as Alfred’s, for by the middle of the twelfth
century England witnessed the anarchy of King Stephen’s reign and by the
thirteenth the barons’ revolt. For Normans and Anglo-Saxons alike, it seems,
the real, divisive issues were not linguistic.
This was the case, too, at the height of the Lollard controversy. Inspired
by John Wyclif, a theologian and graduate of Balliol College, Lollardy Xour-
ished at the end of the fourteenth and beginning of the Wfteenth centuries. It
was less a movement, however, than an association of movements, some
learned and some popular, some avowedly political and some not. Lollard
beliefs, likewise, ran a spectrum from the mildly heterodox to the overtly
heretical, such as Wyclif ’s own denial of the doctrine of transubstantiation.
For such views, Wyclif was condemned, in succession, by Pope Gregory XI in
1377, the Blackfriars Council in 1382, and Archbishop Courtenay of Canter-
bury in 1382. One of the prominent issues in Lollard rhetoric, both for
and against, was their language planning over the role of the vernacular in
Church worship and practice.
Portions of the Gospel had been translated into Old English, which had
also been used for numerous sermons and homilies, and a pre-Lollard Anglo-
Norman Bible translation is also extant. But since the Norman Conquest and
within the dynamics of diglossia, Latin dominated all written and spoken
religious activity in late-medieval England. This domination proceeded not
simply from St Jerome’s Vulgate Bible as the standard text of God’s word
but also, with the twelfth-century rise of scholasticism, from theological
discussion at Oxford and Cambridge. For their part, the Lollards tended to
represent this dominance of Latin as part of a larger obfuscatory process that
concealed from believers the speciWcs of Scripture and their faith, the gaps
between Scripture and received opinion, and the moral failures of those who
managed Church practice. Making theology available in the vernacular was
thus viewed as part of a process of integrity and virtue that could contribute
to the conviction of English faith. According to one commentator, the refusal
 Policy and Politics 145

to render God’s word in English amounted to a wilful and cursed attempt to

hinder the salvation of the populace: ‘For, if a master of skole knoweþ a sotilte
to make his children clerkis, and to spede hem in here lernynge, he, hidynge
þis lore from hem þat ben able þerto, is cause of here vnkynnynge. So, if
writynge of þe gospel in Englische and of good doctrine þerto be a sotiltee and
a mene to þe comoun pepel to knowe þe ryZt and redi weye to þe blisse of
heuene, who loueþ lasse Crist, who is more cursed of God þan he þat lettiþ
þis oon knowynge?’22 In the prologue to the second WycliYte Bible transla-
tion, the righteousness of a demand for vernacular translation produces
desperation among believers: ‘For, þouZ couetouse clerkis ben wode bi symo-
nie, eresie and manie oþere synnes, and dispisen and stoppen holi writ as
myche as þei moun, Zit þe lewid puple crieþ aftir holi writ to kunne it and
kepe it wiþ greet cost and peril of here lif.’23 The Lollards’ opponents recog-
nized full well that translation was in eVect a wedge issue that opened up
social and ecclesiastical concerns far beyond simply the language in which
one read the Bible. To the chronicler Henry Knighton, in translating the Bible
‘from Latin into the language not of angels but of Englishmen’, Wyclif ‘made
that common and open to the laity, and to women who were able to read,
which used to be for literate and perceptive clerks, and spread the Evangelists’
pearls to be trampeled by swine. And thus that which was dear to the clergy
and the laity alike became as it were a jest common to both, and the clerks’
jewels became the playthings of laymen, that the laity might enjoy now forever
what had once been the clergy’s talent from on high.’24 Through translation
some of the fundamental distinctions of late-medieval society—literate from
illiterate, clergy from laity, men from women—were at risk, and with them
the more encompassing distinction between metaphorical jewels and jests.25
Language change and variation Wgured provocatively and complexly in this
late-medieval controversy. Given the prominence of these linguistic topics in
their policies and their willingness to defend them literally to the death, the
Lollards certainly embraced change. And yet their motivation, many of them
argued, was not the declaration of some new truth but the recovery of

22 Anne Hudson (ed.), Selections from English WycliYte Writings (Cambridge: Cambridge Univer-
sity Press, 1978), 107.
23 Hudson (ed.), Selections from English WycliYte Writings, 67.
24 Knighton, Knighton’s Chronicle, 1337–1396, ed. and trans. G. H. Martin (Oxford: Clarendon Press,
1995), 243, 245.
25 For an account of the Lollard movement, see Anne Hudson, The Premature Reformation:
WycliYte Texts and Lollard History (Oxford: Clarendon, 1988). For a concise account of the relations
between Lollardy and English, which has lately been the focus of much critical commentary, see
Margaret Aston, ‘Wyclif and the Vernacular’, in Anne Hudson and Michael Wilks (eds.), From Ockham
to Wyclif, Studies in Church History, Subsidia 5 (Oxford: Blackwell, 1987), 281–330.
146 Language Anxiety

transcendent, immutable truths that had become obscure through Church

policy over time. Regarding fundamental Christian law, one writer notes that
‘Crist in þe houre of his assencioun comaundid to hise diciplis to preche it to
alle pepelis—but, we be siker, neiþer only in Frensch ne in Latyn, but in þat
langage þat þe pepel vsed to speke, for þus he tauZt hymself’.26 If made
intelligible to all people, the Bible would appear with its original sense,
stripped of excrescence and therefore less liable to misunderstanding.
This attention to translation as not disruption but a linguistic and moral
stabilizing process appears in a variety of ways. The prologue to the Bible
translation, for example, follows St Augustine’s injunctions in De Doctrina
Christiana and takes pains to advocate the collation of Latin texts in order to
arrive at correct readings. It even describes a process that carefully distin-
guishes variants: ‘And where þe Ebru bi witnesse of Ierom, of Lire and oþere
expositouris discordiþ fro oure Latyn biblis, I haue set in þe margyn bi maner
of a glose what þe Ebru haþ, and hou it is vndurstondun in sum place.’ Having
speciWed how to render in English syntactic structures like Latin ablative
absolutes and words like autem, the writer even suggests the potential English
has for being more accurate than the Latin of received tradition: ‘I purposide
wiþ Goddis helpe to make þe sentence as trewe and open in English as it is in
Latyn, eiþer more trewe and more open þan it is in Latyn.’27 In the same vein,
the prologue to one of the WycliYte Glossed Gospels oVers a detailed account
of how gloss and text are distinguished: ‘þe text of þe gospel is set Wrst bi itsilf,
an hool sentence togider, and þanne sueþ [follows] þe exposicioun in þis
maner: Wrst a sentence of a doctour declaringe þe text is set aftir þe text, and in
þe ende of þat sentence, þe name of þe doctour seiynge it is set, þat men wite
certeynli hou feer [far] þat doctour goiþ.’28 A similar use of translation to
channel textual (and social) stability runs collectively through the texts of the
various WycliYte translations themselves. Although physically the two hun-
dred or so extant copies of the WycliYte Bible vary from New Testament
extracts to simple quartos to lavishly illuminated folios, linguistically
WycliYte books in general display consistency that suggests not simply the
coordinated reproduction of texts but the cultivation of a sociolect as well.29

26 Hudson (ed.), Selections from English WycliYte Writings, 108.

27 Hudson (ed.), Selections from English WycliYte Writings, 69, 68.
28 This prologue occurs in Cambridge, Trinity College MS B.1.38 and is quoted in Kantik Ghosh,
‘Manuscripts of Nicholas Love’s The Mirror of the Blessed Life of Jesus Christ and WycliYte Notions of
‘‘Authority’’ ’, in Felicity Riddy (ed.), Prestige, Authority, and Power in Late Medieval Manuscripts and
Texts (Woodbridge: D. S. Brewer, 2000), 20.
29 On the physical format of WycliYte Bibles see Henry Hargreaves, ‘The WycliYte Versions’, in
G. W. H. Lampe (ed.), The Cambridge History of the Bible, ii, The West from the Fathers to the
 Policy and Politics 147

In eVect, while critics like Knighton saw translation into English within
the context of Lollard social disruption, much Lollard practice conceived
language change as a necessary disruption that would restore linguistic
stability and reduce the possibility of further change. As with King Alfred’s
revival of learning, the linguistic goal was to transcend change and variation.
It was Church orthodoxy, in fact, that could be charged with pernicious
change, since, as Wyclif argued, theological terms like transsubtanciacio
(‘transubstantiation’) were all recent inventions in the schools and not
words—or concepts—with a historical or scriptural foundation.30 Con-
versely, even if the Latin language of St Jerome’s Vulgate may have remained
constant, in this case linguistic constancy, due to variability in interpretation
and lapses in human conduct, had helped obscure the original immutability
of the word of God. Such language policies served a more general Lollard goal
of not so much remaking or revolutionizing the Church as returning it to the
condition from which it had deviated, and so again like Alfred’s policies—
and, indeed, like much of the language anxiety I examine in this book—they
involved fundamental paradoxes. Such policies advocated language change
in order to restore language immutability, just as they cultivated stabilizing
linguistic practices in order to eVect change. And critics of Lollardy had
paradoxes of their own. The Wfteenth-century prelate Reginald Pecock, for
example, maintained that since much of Lollardy’s popular, heretical success
could be attributed to its use of English, English should also be used by those
seeking to refute the heretical arguments.31
That language in itself represented a signiWcant point of contention in the
debates over Lollardy is beyond doubt. William Thorpe, in an account of his
1407 ‘testimony’ before Thomas Arundel, archbishop of Canterbury, states a
version of perhaps the broadest theoretical justiWcation for translation into
English. To Thorpe, the key points are that the essence of the Gospel does not
rest in the words of any particular version and therefore that an English
rendering can do as well as a Latin one: ‘Ser, by autorite of seint Ierom, þe
gospel is not þe gospel for redying of þe lettre, but for þe blieue þat men haue

Reformation (Cambridge: Cambridge University Press, 1969), 387–415. For the Lollards’ linguistic
habits see generally Hudson, The Premature Reformation, 174–227, and M. L. Samuels, ‘Some Appli-
cations of Middle English Dialectology’, in Margaret Laing (ed.), Middle English Dialectology: Essays on
Some Principles and Problems (Aberdeen: Aberdeen University Press, 1989), 64–80.
30 Aston, ‘Wyclif and the Vernacular’, 300.
31 Pecock, The Repressor of Over Much Blaming of the Clergy, ed. Churchill Babington, Rolls Series
19, 2 vols. (London: Longman, 1860), i, 128; The Reule of Crysten Religioun, ed. William Cabell Greet,
EETS os 171 (London: Oxford University Press, 1927), 17–22.
148 Language Anxiety

in þe word of Crist—þat is þe gospel þat we bileue, not þe lettre þat we reden.

Forþi þe lettre þat is touchid wiþ mannes honde is not þe gospel, but þe
sentence þat is verily bileued in mannes herte þat is þe gospel.’32 The issue
appears prominently in a number of treatises from various viewpoints, and
Lollard books were eventually burnt in a very public fashion. The status of
English also Wgured signiWcantly in oYcial responses to Lollardy. Archbishop
Arundel’s own 1409 Constitutions proscribed both translation of the Bible into
English and the possession of Bibles in English that dated to Wyclif’s lifetime,
while Bishop William Alnwick’s 1429 investigations of heresy in Norwich
maintained that knowledge of the Creed, Pater Noster, or Ave Maria in
English could be regarded as proof of heresy.
But from the broadest sociolinguistic perspective and not that of Lollardy
alone, English and translation were not the sole or primary motivating factors
here, since at this very moment other domains advanced the vernacular
without comment or controversy. Indeed, as exempliWed by Chaucer’s
Boece, a rendering of the Consolation of Philosophy, academic standards of
deWning and glossing meaning could be successfully used in the production of
vernacular translations of school texts. In rendering this work in English, that
is, Chaucer incorporated into his translation extensive expansions and ex-
planations from scholastic commentary traditions on the Latin Consolation.
Chaucer thereby not only clariWed his own work’s sense but acted himself as a
commentator, and in the process he implicitly advanced the claim that
English could be used in ways historically restricted to Latin. If the ten extant
medieval manuscripts (or fragments) of the Boece do not indicate a ground
swell of interest among readers, neither do they give any sign that the work
was resisted or suppressed because of Chaucer’s innovative use of English.
Even more pointed and potentially far-reaching indications that the status of
English could be advanced without controversy come from the reign of Henry
V. During this comparatively brief period (1413–22), the king served as at
least an inspiration for some early Wfteenth-century English poetry, employed
the language in his correspondence, and perhaps fostered the cultivation of an
English sociolect in the Chancery and other oYces of government text
production.33 And a still greater sign of the late-medieval acceptance (if not

32 Anne Hudson (ed.), Two WycliYte Texts, EETS 301 (Oxford: Oxford University Press, 1993), 78.
33 On Chaucer’s translation, see Machan (ed.), Chaucer’s Boece, Middle English Texts 38 (Heidel-
berg: Carl Winter, 2008), xi–xx. Derek Pearsall has suggested that Henry’s support of English may be in
part a response to Lollardy’s use of the vernacular. See Pearsall, ‘Hoccleve’s Regement of Princes: The
Poetics of Royal Self-Representation’, Speculum, 69 (1994), 386–410.
 Policy and Politics 149

necessarily embrace) of English is the almost decade-by-decade increase in

the rate of manuscripts produced in English during the fourteenth and
Wfteenth centuries.
It is yet another paradox of late-medieval English sociolinguistics, then,
that its practices both suppressed and advanced the expanded use of the
vernacular and thus channeled non-linguistic concerns through language in
very much a situational fashion. As a rhetorical Wgure, paradox necessarily
points to something behind itself—some higher truth or conXict or realiza-
tion—that would in this case aYrm that these linguistic policies are not
simply whimsy or sleight of hand. The resolution of this linguistic paradox
again lies in the fact that the issues underlying it are not essentially linguistic.
The Lollard controversy foregrounded change and variation and thereby
fostered concerns about them, but it used these concerns to channel anxiety
over narrowly religious policy and what has been called vernacular theology.
Lollardy’s linguistic ambitions were actually rather limited. It did not assay
status language planning that would have transformed the role of English
within the linguistic repertoire of late-medieval England; it did not seek to
insert English into the domains of law, court culture, and business; and it did
not seek to institutionalize English grammars, rhetorics, and dictionaries. By
the same token, neither did opponents like Knighton indicate that they hoped
to suppress the use of written English in general, which, as I have said, was
expanding signiWcantly at this very moment. For all sides of the argument, the
diachronic change and synchronic variation of language could be used to
focus the truly consequential issues of theology and Church government, and
they thereby oVered reason to argue and believe that the contemporary
alteration of language would produce transformation of these same issues.
If language change mirrored moral and institutional failure, it could also
putatively lead to their eventual correction, and in this vein religion and
culture—not language—remained as much the primary issues for the Lol-
lards as they had been for King Alfred and would be for American nativist
movements or the revival of Passamaquoddy. The limited linguistic conse-
quences of Lollardy can thus be seen to follow from the misdirection that
deWned its representations of change and variation. When the status of
English within England’s linguistic repertoire did Wnally change, accompanied
by changes in the corpus and its management, it did so not because of
religious policy but through the early modern institutionalization of English
in government and legal proceedings, education, and business, made possible
by growths in printing and literacy and enabling the nationalism and ideology
of linguistic standardization that sustained this change.
150 Language Anxiety

The government of language

A distinction between secular and religious policies is not easily drawn in the
examples of Alfred’s translation program and the representation of language
change in Lollard activities. Alfred was of course a king, but his policy focused
on the translation of key ecclesiastical texts and the preparation of men for
Holy Orders, while Lollard activities, though concentrated on theological
matters, ultimately were associated with the social unrest of the 1381 Peasants’
Revolt and other civil disturbances of the period. These kinds of overlap
perhaps inevitably emerged from the investiture that the medieval English
crown and Church had in one another. With the increasing separation of
Church and state in predominantly Anglophone regions since the early
modern period, however, the former’s role in language planning has dimin-
ished. In England, perhaps the last and most signiWcant acts of religious
linguistic policy were self-erasing ones that indirectly helped to exclude the
Church from language planning by furthering the use of vernaculars and
thereby relaxing any centralized control over sacred language. I refer here to
Reformation proclamations like the 1549 First Act of Uniformity, which
required the English Book of Common Prayer in public religious observance
and restricted other languages—Hebrew, Greek, Latin—to private worship.34
But also, more generally, I have in mind the Roman Catholic Council of
Vatican II (1962–5), which mandated a shift from Latin to local vernaculars for
religious services, since this shift had the similar eVect of further constraining
Church opportunities for language planning.
Institutional as well as ideological reasons situate contemporary language
planning of all kinds, primarily, within the provenance of government rather
than the Church. Beyond whatever impact the rise of scientiWc methods and
religious skepticism may have had on individual perceptions of language, it is
government—local, regional, or national—that characteristically has the
funding and infrastructure to manage language planning. Taxes provide the
means to maintain institutions like schools and courts, which in turn serve as
the avenues for the articulation and enforcement of policies about language
use and change. Indeed, language planning as it exists today, ultimately
impinging on all aspects of daily life, was in many ways made possible only
by the development of political states, which have joined together institu-
tional infrastructure, technological developments in printing, mandatory
education, the discipline of modern linguistics, and notions of nationalism.

34 G. R. Elton, The Tudor Constitutions: Documents and Commentary, 2nd edn. (Cambridge:
Cambridge University Press, 1982), 402–5.
 Policy and Politics 151

If individual schools provide instruction in English or other languages, for

example, they do so using modern technology and according to modern
understandings of language acquisition, within government-regulated educa-
tional policies that are maintained by schoolboards and courts, and against
a background of nationalist discourse about the relations between language
and society.
Educational policy reXects and channels social anxieties through language
change and variation in several distinct ways. Perhaps most simply, policies
can create shift and transform social identity by suppressing the use of non-
English languages in classrooms, as has been the practice in nearly all contact
situations until fairly recently. Irish, Welsh, and Scots Gaelic were thus actively
discouraged and sometimes even proscribed in nineteenth-century Irish,
Welsh, and Scottish classrooms, as were Aboriginal languages in Australia,
Spanish in Puerto Rico, and Maori in New Zealand. In Wales, the 1861
educational code even tied government funding to elementary students’
proWciency in English and arithmetic, leaving schools little reason to teach
Welsh and every reason to suppress it. And whatever reasons there may have
been to teach Welsh became moot by the British 1870 Education Act, which
simply prohibited the use of the language in schools.35 In all these regions save
Puerto Rico, the schoolroom proscription of an indigenous language con-
tributed to a signiWcant shift to English both directly and indirectly. Recent
studies have shown, that is, that the use of a dominant language like English in
the classroom, irrespective of the educational impact of what happens in that
classroom, furthers shift in other ways. When children return home from
school, they introduce the dominant language into domestic domains, bring-
ing with it not only bilingualism but also the social values associated with that
dominant language. This same conclusion appeared as long ago as an 1839
report by the Edinburgh Gaelic School Society, which indicates that Scottish
educational policies aVected Gaelic-speaking parents as well as their children:
‘it is diYcult to convince [the parents] that it can be any beneWt to their
children to learn Gaelic, though they are all anxious, if they could, to have
them taught English.’36 And in this way, as in the United States, just three
generations can lead from monoglot, non-Anglophone parents, to bilingual
children, to monoglot, Anglophone grandchildren. By itself, this kind of shift

35 Mari C. Jones, Language Obsolescence and Revitalization: Linguistic Change in Two Contrasting
Welsh Communities (New York: Oxford University Press, 1998), 10; Stephen May, Language and
Minority Rights: Ethnicity, Nationalism and the Politics of Language (London: Longman, 2001), 170.
36 Quoted in Daniel Nettle and Suzanne Romaine, Vanishing Voices: The Extinction of the World’s
Languages (Oxford: Oxford University Press, 2000), 138.
152 Language Anxiety

can also lead to personal and familial trauma over a speaker’s history, identity,
and ethnicity, as famously described by Richard Rodriguez.37
In instances where such shift has been compelled, the brutality of its
methods can channel additional trauma through language. Welsh children
discovered speaking their native language at school had the ‘Welsh knot’ (or
‘not’), a knotted rope, placed around their neck, and wore it until another
child’s linguistic transgression earned the knot. At day’s end, the last child
wearing the rope was punished, sometimes by being beaten with the rope. Tally
sticks that recorded instances of using Irish or Gaelic served similar purposes
in Irish and Scottish classrooms.38 Even more brutally, speakers of Aboriginal
or Amerindian languages suVered not only direct corporal punishment for
speaking a language other than English but also forced removal to boarding
schools. According to one historical account of the Navajo reservations,
In the fall the government stockmen, farmers, and other employees go out into the
back country with trucks and bring in the children to school. Many apparently came
willingly and gladly; but the wild Navajos, far back in the mountains, hide their
children at the sound of a truck. So stockmen, Indian police, and other mounted men
are sent ahead to round them up. The children are caught, often roped like cattle, and
taken away from their parents, many times never to return. They are transferred from
school to school, given white people’s names, forbidden to speak their own tongue,
and when sent to distant schools are not taken home for three years.39

Once the Indians arrived at the boarding schools, additional eVorts were
made to manage their linguistic as well as social variance from the Anglo
population at large. At the Carlisle Indian Industrial School in Pennsylvania,
founded in 1879, new arrivals had their braids cut and clothes replaced. In
what amounted to a re-enactment of Babel, not only were they forbidden to
speak their own languages but also speakers of diVerent languages were
scattered among each other in the barracks, so as to minimize the possibilities
of maintaining tribal groups or languages.40 The United States Bureau of
Indian AVairs abandoned all such policies by 1934, but the long-term failure
and detrimental eVects of a compelled shift to English have become ingrained
in Native American culture.41

37 Rodriguez, Hunger of Memory: The Education of Richard Rodriguez: An Autobiography (Boston,

MA: D. R. Godine, 1982).
38 Jones, Language Obsolescence and Revitalization, 10; Viv Edwards, Multilingualism in the English-
Speaking World: Pedigree of Nations (Malden, MA: Blackwell, 2004), 97–9.
39 Quoted in David H. DeJong, Promises of the Past: A History of Indian Education in the United
States (Golden, CO: North American Press, 1993), 118.
40 Sally Jenkins, The Real All Americans: The Team that Changed a Game, a People, a Nation (New
York: Doubleday, 2007), 78.
41 See DeJong, Promises of the Past, and Crawford (ed.), Language Loyalties, 41–51.
 Policy and Politics 153

It is important to recognize, I think, that not all shift was accomplished in

brutal ways for pusillanimous reasons. A kind of historically myopic self-
righteousness makes it possible to dismiss as simply examples of the Western
avowal of the ‘white man’s burden’ sociolinguistic activity that was in some
cases sincerely meant and sincerely appreciated. In 1663, for example, John
Eliot issued a translation of the Bible into the Natick dialect of Massachusetts
on the conviction that it would serve as a vehicle for the establishment of
a nation founded on Christian principles—which was precisely what the
Massachusetts Bay Colony had hoped to accomplish—and would thus allow
Natick speakers to transcend European limitations as embodied in European
languages. Less theologically but perhaps just as benevolently, other early
European visitors to North America sought to record American Indian
languages as an expression of a Romantic desire to improve the human
condition and preserve cultural origins that were sometimes perceived to be
as ancient as Greece’s.42 Even Carlisle was founded with the best of intentions
by Captain Henry Pratt on the Wrm belief that assimilation to Anglo
culture oVered Native Americans the best chance for prosperity and success
in an America largely predisposed to contain if not exterminate them. In an
editorial in the Wrst issue of the school newspaper, he deWned Carlisle’s
purpose in this way: ‘Instead of educating soldiers to go to the western plains
to destroy with powder and ball, it is proposed . . . now to train at this
institution a corps of practical, educated, and Christian teachers, who will
by precept and practice induce their tribes on the plains to adopt the peaceful
pursuits of Christian people.’43
It is also important to acknowledge, though, that Eliot and Pratt were
exceptions to the often unambiguously cynical manipulation of language
change for non-linguistic reasons, of which the experience of the Hawaiian
islands is paradigmatic. An indigenous population of approximately 800,000
at Captain James Cook’s 1778 landing had declined by 1893 to perhaps 40,000
individuals with varying degrees of competence in the Hawaiian language. As
in other PaciWc regions (such as New Zealand), partly the decline of speakers
owed to the introduction of diseases for which the indigenous people lacked
immunities, and partly it owed to the paradoxical Western attitude that saw
Hawaiians as having both the virtue of noble savages and the simplicity of
children—and for both reasons as a people in need of and unable to prevent
westernizing intervention. Laura Fish Judd, an early Anglo resident, thus
observed in 1880, ‘If the Italian is the language of the gods, the French of
42 Edward G. Gray, New World Babel: Languages and Nations in Early America (Princeton, NJ:
Princeton University Press, 1999), 56–84, 112–38.
43 Quoted in Jenkins, The Real All Americans, 81.
154 Language Anxiety

diplomacy, and the English of business men, we may add that the Polynesian
is the dialect of little children. It is easier to say ‘‘hele mai’’, than ‘‘come here’’,
and ‘‘I wai’’, than ‘‘give me water’’.’44 While many early schools conducted
education in Hawaiian, they also worked expressly to assimilate indigenes to
Anglo culture by requiring students, as part of their oral exercises, to re-
nounce wooden gods, identify Western countries, and recite passages from the
Bible. Nonetheless, despite the decline in the population and the educational
inXuence of the West, a nativist sentiment persisted and experienced a
resurgence in the 1890s with the cultivation of the hula dance and the
emergence of a class of kahuna (‘priests’) focused on traditional activities
and beliefs.45 Though Hawaiian was in probably irreversible decline by 1893,
then, it still had some life. What changed matters was the resolve of Anglo
businesses to bring closure to their expanding inXuence over the islands. Over
the course of a century they had wrested control of plantation lands and
workers from local chiefs, in order to guarantee the continued, duty-free
export of sugar into the United States.46 In 1893, accordingly, the Hawaiian
monarchy was overthrown, making Hawaii an independent (though Ameri-
can controlled) republic. Another step in this quiet acquisition of the islands
occurred in 1896, when Hawaiian was proscribed in schools and English
required. And in 1900 the Organic Act extended the English-only policy to
the conduct of government business.
For the purposes of this book, the crucial point here is that, unlike at
Carlisle, this suppression of language variation furthered not the presumptive
good of the Hawaiians but, quite nakedly, the monetary gain of Wnanciers.
The eVectiveness of these policies has been nearly complete. In 1880, 150
schools used Hawaiian to teach 4,078 pupils; 60 used English for 3,086
students. In 1902, no Hawaiian schools remained, while 203 schools used
English for 18,382 students.47 By 1983, when not even 50 children could
speak Hawaiian, a language revival program was established, reintroducing
students to Hawaiian culture as well as language. As of 1978 Hawaiian has
been the co-oYcial language of what is now the state of Hawaii, though long-
term prospects for the language’s maintenance remain dim.48

44 Quoted in Albert J. Schütz, The Voices of Eden: A History of Hawaiian Language Studies
(Honolulu, HI: University of Hawaii Press, 1994), 26.
45 Schütz, The Voices of Eden, 347–50.
46 Nettle and Romaine, Vanishing Voices, 94–7, and Sam L. No’eau Warner, ‘The Movement to
Revitalize Hawaiian Language and Culture’, in Hinton and Hale (eds.), The Green Book of Language
Revitalization in Practice, 133–44.
47 Schütz, The Voices of Eden, 352.
48 ‘Major EVort Is Under Way to Revive and Preserve Hawaii’s Native Tongue’, New York Times, 15
April 2007, A18.
 Policy and Politics 155

The motivations behind the Bureau of American Indian AVairs’ policy of

removing Native American children from reservations to English-only board-
ing schools displaced not economics but essentially genocidal ambitions to
language change and variation. Nominally, removals took place out of con-
cern for the Native Americans. John H. Oberly, the superintendent for Indian
education, thus observed in 1885, ‘It is an understood fact that in making large
appropriations for Indian school purposes, the aim of the Government is the
ultimate complete civilization of the Indian. When this shall have been
accomplished the Indian will have ceased to be a beneWciary of the Govern-
ment, and will have attained the ability to take care of himself.’ Without
removal to boarding schools, Native Americans, it was alleged and feared,
would become extinct. Yet ‘civilization’ is of course a loaded term in this
context, and few bureaucrats shared the compassion of Captain Pratt, the
founder of Carlisle. An 1868 commission composed of government oYcials
and army generals oVered one gloss of ‘civilization’ and its relevance to
language variation: ‘Through sameness of language is produced sameness of
sentiment, and thought; customs and habits are moulded and assimilated in
the same way, and thus in process of time the diVerences producing trouble
would have been gradually obliterated. By civilizing one tribe others would
have followed . . . Schools should be established, which children should be
required to attend; their barbarous dialects should be blotted out and the
English language substituted.’ Twenty-one years later the federal commis-
sioner of Indian aVairs, J. D. C. Atkins, oVered much the same gloss and
relevance: ‘The Wrst step to be taken toward civilization, toward teaching the
Indians the mischief and folly of continuing in their barbarous practices, is to
teach them the English language. The impracticability, if not impossibility, of
civilizing the Indians of this country in any other tongue than our own would
seem to be obvious, especially in view of the fact the number of Indian
vernaculars is even greater than the number of tribes.’49
In New Zealand (or Aotearoa), contact between English and varieties of the
indigenous Maori language illustrates still other ways by which government
policies have fostered, in sometimes brutal fashions, anxiety about language
change and variation. A Polynesian people, the Maori Wrst arrived at the
uninhabited North and South Islands late in the Wrst millennium, probably
coming from Tahiti and before that the Marquesas Islands. An inXux of
additional Polynesian immigrants in the fourteenth century fortiWed the
descendants of the original group, and by 1642, when Abel Tasman captained
the Wrst European ship to dock in New Zealand, Maori language and culture,

49 Quoted in Crawford (ed.), Language Loyalties, 48, 51.

156 Language Anxiety

the latter known for its carving and weaving in particular, were well and
diversely established. Barely a century later, however, the sociolinguistic
character of the area began to change swiftly and radically with the 1769
arrival of Captain Cook. European seal hunters landed in 1792, and by 1807
so, too, had whalers from the United States, Norway, Spain, and the East
India Company. At this time, perhaps nearly 200,000 predominantly mono-
glot Maoris resided in New Zealand. By 1840 this population had decreased
by 40%, and by 1890 it totaled about 45,000 individuals, a decline of roughly
75% over the course of the nineteenth century.50
The reasons for this decline are of course varied. As in Hawaii, North
America, Africa, and Asia, Europeans introduced potentially fatal illnesses
and diseases like measles, for which the indigenous populations had no
antibodies, and also venereal diseases, which resulted in sterility. The trading
of weapons for goods and the cultivation of strategic alliances that pitted
individual groups against one another further contributed to the eradication
of the Maoris. For the study of the anxiety attendant on language change,
however, the most signiWcant factors are the sociolinguistic ones. In 1838, only
about 2,000 Europeans resided in New Zealand, but within four years, due to
the attraction of farming and whaling opportunities, that number reached
10,000. In the latter half of 1861, after the discovery of gold, the European
population of the South Island’s Otago alone leapt from 13,000 to more than
30,000.51 In this way, already in 1858 the population of pakehas (basically,
‘Europeans’) reached 59,000, thereby outnumbering that of Maoris less than a
century after the appearance of the Wrst English ship.52 In addition to the
advantages attendant on their control of technology and institutions such as
the court system, Anglophones thus soon had simple numbers on their side.
Even more sociolinguistically signiWcant than this, the school system, as in
India, contributed to a dramatic shift in the area’s linguistic repertoire.
The earliest mission schools used Maori as a medium of education, with a
Maori translation of the Bible appearing in 1827.53 As I noted in Chapter 1,
recent critics have emphasized the fact that language instruction is never
value-free, since it always carries with it some of the ideals and principles of
the culture it sustains, and in this light it can be argued that, ironically, this
early instruction in Maori became a policy that may ultimately have helped to

50 Ranginui Walker, Ka Whawhai Tonu Matou: Struggle without End (Auckland: Penguin Books,
1990), 80–1.
51 Laurie Bauer, ‘English in New Zealand’, in BurchWeld (ed.), English in Britain and Overseas:
Origins and Development, 383.
52 Walker, Ka Whawhai Tonu Matou, 113.
53 Walker, Ka Whawhai Tonu Matou, 85.
 Policy and Politics 157

undermine the language. By fostering the Western technology of literacy in

a fundamentally oral society, missionaries implanted Western culture and
created a means for the acquisition of land in accordance with Western
conceptions of property. Further, by privileging some dialects as languages
to be used for instruction in schools, missionary policy helped to simplify
Maori linguistic diversity and establish belief in a monoglot culture analogous
to the model of England and other dominant European powers.54 The initia-
tive to Xatten out variation within Maori or between it and English appears
with particular force in the Wrst signiWcant grammar and dictionary of Maori,
published by Thomas Kendall and Samuel Lee in 1820. In order to teach the
language to Europeans, who were likely educated in the highly synthetic
languages of Latin and Greek, the authors willfully misrepresented Maori so
as to make its grammar seem familiar. According to their preface, a ‘particular
object of the work, is the instruction of the European Missionary in the
Language of New Zealand . . . and for this end it was that Examples in declen-
sion and conjugation have been given, after the manner of European Gram-
mars; when, in fact, there exists no such thing in the language in question’.55
Supplemented by the continued denigration of Maori as the inferior,
debased language of an inferior, debased people, Maori literacy in eVect
worked to accelerate a linguistic shift to English that English literacy only
furthered. The Native Schools Act of 1867, thus, directed that village schools
be conducted in English, while the Native School Code of 1880 mandated that
teachers have some familiarity with Maori, not to maintain the language but
to ease the transition of students in the earliest grades and to facilitate the
teaching of English. By 1905 the Inspector of Native Schools had promoted the
proscription of Maori on playgrounds, which, as with Native American
languages in the United States, Aboriginal languages in Australia, and Celtic
languages in Ireland, Wales, and Scotland, came to mean the sometimes
violent proscription of Maori in general.
The statistical concomitants of such policies are again striking: in 1900, over
90% of Maori primary students spoke Maori as their Wrst language; by 1960,
only 26% did; and by 1979 there was genuine concern that soon Maori would
follow the pattern of Cornish and Manx and have all its speakers shift to
English. The 1982 establishment of early-childhood Maori immersion pro-
grams through Te Kōhanga Rao, the 1986 creation of the Waitangi Tribunal to
hear Maori grievances—an implicit recognition of a desire to protect the

54 Peter Mühlhäusler, Linguistic Ecology: Language Change and Linguistic Imperialism in the PaciWc
Region (London: Routledge, 1996), 104–240. This book contains a wealth of historical quotations about
the alleged barbarism of the Maori language and people.
55 Quoted in Schütz, The Voices of Eden, 252.
158 Language Anxiety

language—and the 1987 Maori Language Act, which made Maori and English
co-oYcial languages that could both be used in courts of law, have led to some
resurgence in the numbers of speakers and domains of usage. But even so,
with Maori speakers overrepresented in low-paying jobs, unemployment, and
low levels of education, the vitality of the language remains tenuous.56
According to the 2006 National Census of New Zealand, 565,329 individuals
were ethnically Maori; this represents about 14% of a national population
currently estimated to be over four million. While Maori does hold co-oYcial
status with English, perhaps only 157,000 people are Xuent with the language,
most of whom are in fact Maori by descent.57 The Maori population may have
rebounded tenfold in one hundred years, then, but like the Celtic-speaking
populations of Ireland, Wales, Cornwall, and Scotland it has shifted over-
whelmingly to English.
Emblematic of the sociolinguistic issues implicit in the fairly dramatic shift
of Maori speakers to English is the Treaty of Waitangi, signed on 6 February
1840, at a time when Maori speakers still outnumbered Anglophones. The Wrst
occasion on which the word Maori was used as a collective proper noun for all
the indigenous peoples of New Zealand—who discriminated among them-
selves by tribal names—the treaty was written in both English and Maori,
which had been a written language for about twenty years and which therefore
sustained an oral, not a textual, culture. Most of the roughly Wve hundred
Maori who ‘signed’ the treaty, indeed, did so only by making some kind of a
mark. In English, the Wrst article of the treaty gave ‘sovereignty’ to the Queen
and her representatives over the whole of the region, but in the Maori version
the word that corresponds to ‘sovereignty’ is ‘kāwanatanga’, a word that better
translates as ‘governorship’. The Maoris thus likely believed that they were
ceding not their autonomy but rather the governance of imported British
institutions. And this misunderstanding was reinforced by the second article,
which in the Maori version used ‘rangatiratanga’, which does mean something
like ‘absolute chieftainship’, to describe the Maoris’ relationship to their lands,
homes, and possessions. Here, interlanguage variation helped to erase itself by
fostering a situation that accelerated the Maoris’ shift to English; if Anglo-
phones had sovereignty, that is, Maori held no interest for them or advantage

56 Walker, Ka Whawhai Tonu Matou, 146–8, 238–40; Mühlhausler, Linguistic Ecology, 246–7;
Edwards, Multilingualism in the English-Speaking World 59, 106, 117; May, Language and Minority
Rights, 285–308; Richard A. Benton, ‘Language Policy in New Zealand: DeWning the IneVable’, in
Michael Herriman and Barbara Burnaby (ed.), Language Policies in English-Dominant Countries: Six
Case Studies (Clevedon: Multilingual Matters, 1996), 62–98; Jeanette King, ‘Te Kōhanga Reo: Māori
Language Revitalization’, in Hinton and Hale (eds.), The Green Book of Language Revitalization in
Practice, 119–28.
57 [Link]
 Policy and Politics 159

for the Maoris themselves. And from this linguistic confusion (if not misdir-
ection) also followed many of the political, social, and economic diYculties of
the next century and a half.58
For indigenous Hawaiians, Native Americans, Maoris, and, indeed, most of
the linguistic minority groups that English contacted inside and outside of the
British Isles, the intensity of policies designed to aVect shift and eliminate
variation seems surprising. Economically, technologically, and educationally
disadvantaged to begin with, such groups would seem to have posed little
threat to the institutional and commercial advantages of Anglophones, who
in most of these regions (as in New Zealand) attained signiWcant numerical
superiority as well in a fairly short time. Even less clear is how a minority
language not viable in business or government—and often lacking a written
form—could in and of itself pose a threat to English and its speakers. What
makes these policies clear (and familiar) is recognition of how they channel
non-linguistic concerns into variation and change—into variation from
English and the change of non-Anglophones to its use.
Maori itself was never a threat to English; its speakers were, potentially, to
the farming, whaling, and mining industries. And if the variation of their
language was erased—if they substantially shifted to English—at least some
cultural traditions would be suppressed if not erased as well. In their place
would be a sociolinguistic world that would disproportionately advance
the non-linguistic concerns of some speakers (Anglophones) over others
(Maoris). Whether in such situations non-native speakers—or, for that mat-
ter, speakers of a non-standard variety—can ever acquire precisely the same
opportunities and advantages as native speakers of Standard English merely
by changing their language seems doubtful. Even as the Maori overwhelm-
ingly shifted to English, indeed, a kind of stage Maori was cultivated in early
twentieth-century Anglophone writing as the representation of an English
dialect putatively spoken by Maoris alone. This wasn’t the case. The variety,
in some ways like Dickens’s cockney, reXected stereotypes as much as reality,
and what was linguistically real was shared with lower-class pakehas in what
amounted to a class, not an ethnic, sociolect: ‘I been come all te way from Noo
Zeelan to Wght te Sherman soldier in Parani and make him clear outer t’this
country. When I come here some feller been tell me all about t’that dirty trick
all te Sherman been up tin Parani an Peljimi. No good you say t’that all
gammon, it te true talk all right.’59 If language had been the only issue,

58 Walker, Ka Whawhai Tonu Matou, 90–7; Bauer, ‘English in New Zealand’, 383; May, Language and
Minority Rights, 286–7; and Donald McKenzie, Oral Culture, Literacy and Print in Early New Zealand:
The Treaty of Waitangi (Wellington: Victoria University Press, 1985).
59 Quoted in Shaun F. D. Hughes, ‘Was There Ever a ‘‘Māori English’’?’, World Englishes, 23 (2004), 576.
160 Language Anxiety

then once the Maoris shifted languages, there would have been no reason
to fashion Maori English. In terms of the conceptualization of the Maoris,
the dialect served to reinforce their isolation as an ethnic group, even though
they now spoke English. And in this way, as a Labovian stereotype Maori
English points even more strongly to the way language variation can stand in
euphemistically for non-linguistic issues. To me, however, the most striking
issue here is that such use of variation is by no means unique to New Zealand
or, indeed, to the modern post-colonial world: it is characteristic of English in
its recorded history. This same kind of supererogatory anxiety animates the
Norman mythologizing of language contact between English and French,
which I considered in the previous chapter, and it also runs through responses
to the continued use of non-Anglophone languages in multilingual settings.
Bilingual education, whether its goal is maintenance of more than one
language or transition to English alone, has been a particularly vitriolic issue
in the United States. The Bilingual Education Act of 1968—oYcially, Title VII
of the Elementary and Secondary Education Act—established federal guide-
lines, and subsequent revisions of the act and Supreme Court decisions
have kept the issue a prominent and controversial part of federal policy.
In the 1974 case Lau v. Nichols, for example, the United States Supreme
Court ruled on behalf of a group of Chinese-American parents whose non-
Anglophone children had received no special aid for their language skills and
who were consequently (the parents argued) performing poorly in school.
While the Court of Appeals had ruled in favor of the school district, observing
that ‘every student brings to the starting line of his education career diVerent
advantages and disadvantages caused in part by social, economic, and cultural
background’, the Supreme Court maintained that ‘there is no equality of
treatment merely by providing students with the same facilities, text-books,
teachers, and curriculum; for students who do not understand English
are eVectively foreclosed from any meaningful education’.60 Some of the
controversy about bilingual education arises from the fact that unimpeach-
able conclusions about long-term eYcacy are diYcult to draw in light of the
variation among programs, their assessment methods, the sociocultural con-
texts of individual schools, and the pressures of political groups. As with
diachronic structural change, it’s impossible either to control for all variables
but one or to test conclusions by reproducing a given set of conditions.
Perhaps not surprisingly, then, competing studies have produced data both
supporting and challenging the notion that bilingual education provides the

60 Quoted in Crawford (ed.), Language Loyalties, 252, 253.

 Policy and Politics 161

most eVective transition to English competence for students who do not

speak the language at home.61
Yet the intensity of current American responses to linguistic variation as
Wltered through bilingual education can also be tied chronologically to rising
rates of immigration, increasing tax burdens, and backlash to civil rights
legislation in general. From the 1980s onward these have produced an increas-
ingly politicized environment for bilingual education, rendering it an outlet
for anxiety about patriotism and employability. President Ronald Reagan
once commented that it ‘is absolutely wrong and against American concepts
to have a bilingual education program that is now openly, admittedly dedi-
cated to preserving their [i.e., non-Anglophone students’] native language
and never getting them adequate in English so they can go out into the job
market and participate’. Reagan’s Secretary of Education, Terrel Bell, justiWed
the diversion of funds from bilingual education by noting, ‘We will protect
the rights of children who do not speak English well, but we will do so by
permitting school districts to use any way that has proven to be successful’.62
Funding, indeed, has proved one of the most divisive issues. Since ultimately
the same tax base that underwrites education also underwrites social services,
the military, and everything from federal parks to border control, a decision to
support bilingual education has many implications for government activity and
national identity. Such a decision generates controversy not only over taxes in
general but in its prioritizing of one cause over another through the allocation
of funds. A clear indication of this aspect of the controversy is the fact that, even
as critics have lambasted the expense, government funding has never provided
bilingual education for more than a fraction of the Limited English ProWciency
students who express an interest in transitional programs to English.63 When
the Bilingual Education Act was allowed to lapse in 2002, then, it did so in
response to anxieties not simply or primarily about language or education
but also about taxes, government regulation, immigration, race, and national
priorities. And by doing so, it fueled anxieties in all these same areas.

61 James Crawford, Bilingual Education: History, Politics, Theory, and Practice, 4th edn. (Los
Angeles, CA: Bilingual Educational Services, 1999), 102–56.
62 Quoted in Crawford, Bilingual Education, 53.
63 Ronald Schmidt, Sr., Language Policy and Identity Politics in the United States (Philadelphia, PA:
Temple University Press, 2000), 17. Also see Crawford, At War with Diversity: US Language Policy in an
Age of Anxiety (Clevedon: Multilingual Matters, 2000), 84–103; Sandra del Valle, Language Rights and
the Law in the United States: Finding Our Voices (Clevedon: Multilingual Matters, 2003), 217–74; and
Roseann Dueñas González and Ildikó Melis (eds.), Language Ideologies: Critical Perspectives on the
OYcial English Movement, i, Education and the Social Implications of OYcial Language (Urbana, IL:
NCTE, 2000).
162 Language Anxiety

Like education, immigration alone has a signiWcant impact on every gov-

ernment’s spending, whether directed at customs agents, border guards, social
services for immigrants, court translators, workplace guidelines, or voting
rights. Also like education, immigration shapes individuals’ views of their
country: its values, economic stature, race, ethnicity, and level of civic respon-
sibility. And in any of these areas language planning can facilitate government
policy by fostering and channeling anxiety about language change and vari-
ation. Australia, New Zealand, Canada, and the United Kingdom have all
regulated immigration in part through evaluations based on an individual’s
command of English, for instance. The relatively small regional variation in
Canadian English, further, is partly a response to political anxieties, following
the 1867 Confederation, over the possibility of American expansion into
western territories. The government at that time facilitated Canada’s own
westward expansion of the railroads, which brought with them a great
many teachers, bankers, and government oYcials from Ontario. In this way,
political anxiety likewise served to expand Anglophone regions and contain
the Francophone population largely in Quebec. Given the relative uniformity
of Canadian English, variation and change retained—and continue to retain—
the power to channel non-linguistic anxieties. While Anglophones may have
come to dominate western territories, new immigrants continued to arrive in
the nineteenth century, leading the Calgary Herald, in 1899, to call for restric-
tions and quotas by evoking the familiar narrative of language change: ‘This
policy of building a nation on the lines of the Tower of Babel, where the
Lord confounded the language so that the people might not understand
one another’s speech, is hardly applicable to the present century.’64
In the United States, which experienced its own increase in immigration
from southern and eastern Europe at this same time, an inability to read or
speak English Xuently could factor into judgements of immigrants as mentally
deWcient and thus lead to the disqualiWcation of many individuals from these
regions. In 1912, Henry Goddard found this to be the case with 83% of the
immigrant Jews he studied, 80% of the Hungarians, 79% of the Italians, and
87% of the Russians. Language shift, race, and social engineering likewise
coalesced in the Australian policy of the 1920s and 1930s that directed the
forced seizure of mixed Aborigine-European children, who were to be raised
among Anglophones and married to others with increasingly less Aboriginal

64 Quoted in Edwards, Multilingualism in the English-Speaking World, 28. See further R. K.

Chambers, ‘ ‘‘Lawless and Vulgar Innovations’’: Victorian Views of Canadian English’, in Sandra Clarke
(ed.), Focus on Canada (Amsterdam: John Benjamins, 1993), 1–26; and Laurel Brinton and Margery
Fee, ‘Canadian English’, in The Cambridge History of the English Language, vi, John Algeo (ed.), English
in North America (Cambridge: Cambridge University Press, 2001), 422–40.
 Policy and Politics 163

blood until the entire group would genetically and linguistically merge with
the full-blooded Europeans.65
The channeling of extra-linguistic concerns through this kind of language
planning becomes particularly salient in moments of non-linguistic social
crisis. While the United States had never had any formal federal language
policy, for example, the events of the First World War produced state and local
restrictions on language as well as immigration. When the country entered the
war, concerns about treason and national loyalty led a number of states to
enact laws that restricted the use of non-English languages (particularly
German) in public, at schools, and even on the telephone. Perhaps most
famously, the 1919 Siman Act in Nebraska proscribed the teaching of any
foreign language until students were in the eighth grade, leading to the arrest
of Robert Meyer for teaching a Bible story in German to a group of younger
children. In upholding Meyer’s conviction for wrongfully teaching a foreign
language, the Nebraska Supreme Court emphasized not language but safety
and citizenship:
The salutary purpose of the statute is clear. The Legislature had seen the baneful eVects
of permitting foreigners, who had taken residence in this county, to rear and educate
their children in the language of their native land. The result of that condition was
found to be inimical to our own safety. To allow the children of foreigners, who had
emigrated here, to be taught from early childhood the language of the country of their
parents was . . . to educate them so that they must always think in that language, and,
as a consequence, naturally inculcate in them the ideas and sentiments foreign to the
best interests of this country.

For its part, in overturning the state court’s ruling in Meyer v. Nebraska, the
United States Supreme Court depended just as heavily on the non-linguistic
eVects and implications of language planning, speciWcally on the rights of
Meyer to pursue his profession and of parents to educate their children in
ways they see Wt. ‘Practically,’ the Court reasoned, ‘education of the young is
only possible in schools conducted by especially qualiWed persons who devote
themselves thereto . . . Mere knowledge of the German language cannot rea-
sonably be regarded as harmful . . . PlaintiV in error taught this language in
school as part of his occupation. His right thus to teach and the right of
parents to engage him so to instruct their children, we think, are within
the liberty of the amendment.’66 In terms of how language can stand in for

65 Edwards, Multilingualism in the English-Speaking World, 27–31.

66 Crawford (ed.), Language Loyalties, 235–6. See also del Valle, Language Rights and The Law in the
United States, 30–9; and Dennis Baron, The English-Only Question: An OYcial Language for Americans?
(New Haven, CT: Yale University Press, 1990), 144–50.
164 Language Anxiety

non-linguistic issues, the similarity of the courts’ arguments, despite the fact
that they use them to reach opposite conclusions, is here striking. Like the
lower court whose ruling it overturned, that is, the Supreme Court invests
language change and variation—neutral linguistic phenomena—with the
power to mediate individual freedom and national identity.
Immigration and social diversity intersect most generally with language
planning and change in laws regarding the oYcial status of English and other
languages. In the late twentieth century these connections were especially
prominent in the United States, which witnessed movements to make English
the oYcial language of not simply the country but also individual states,
counties, and even municipalities.67 One indication of the non-linguistic
motivation of these movements is the history of US English, one of the pre-
eminent organizations. Founded by John Tanton and S. I. Hayakawa in 1983,
US English began as an oVshoot of another organization that Tanton
had founded in the previous decade: Federation for American Immigration
Reform. Serving as they do to channel extra-linguistic social concerns about
immigration, economic recession, increased taxation, and the formation of
national culture, these movements sometimes strain to maintain even a
nominal focus on language policy.68 In April of 2005, for instance, Joe Man-
chin, the governor of West Virginia, vetoed a bill that would have made English
the state’s oYcial language, not because he did not support the idea, and not
because non-Anglophones present particular problems—or, for that matter,
even make up a constituency—in a state wherein only 2.7% of the population
does not speak English at home. He did so, rather, for the legalistic reason that
the legislation had violated the state’s constitution by being tacked onto other
legislation, speciWcally onto the funding of municipal recreation boards.69
The putative need for such legislation thereby remained unchallenged.
Such strains and concerns are neither uniquely American nor uniquely
modern, however. The 1536 Act of Union of England and Wales oVers the
earliest expression of a broad government policy for containing social and
linguistic change and variation through the declaration of languages’ relative
status. After stating that English should be the language used in all Welsh
courts, the Act adds ‘from hensforth no personne or personnes that use the

67 See Baron, The English-Only Question; Larry Rohter, ‘Repeal Is Likely for ‘‘English Only’’ Policy
in Miami’, New York Times, 14 May 1993, A7; Jodi Wilgoren, ‘Divided by a Call for a Common
Language: As Immigration Rises, a Wisconsin County Makes English OYcial’, New York Times, 19
July 2002, A8.
68 Carol L. Schmid, The Politics of Language: ConXict, Identity and Cultural Pluralism in Compara-
tive Perspective (Oxford: Oxford University Press, 2001), 44–5, 32–56.
69 ‘English-Only Bill Is Vetoed’, New York Times, 18 April 2005, A17.
 Policy and Politics 165

Welsshe speche or langage shall have or enjoy any maner oYce or fees within
the Realme of Englonde Wales or other the Kinges dominions upon peyn of
forfaiting the same oYce or fees onles he or they use and exercise the speche
or langage of Englisshe’.70 The 1537 Irish ‘Act for the English Order, Habite,
and Language’ similarly mandated English for all those who wished to be
recognized as ‘his Highness true and faithfull subjects’. And in the 1616 Act
for the Settling of Parochial Schools, James I mandated the same situation
for Scottish schools. Now that ‘the vulgar English tongue be universally
planted’, the Act observes, ‘the Irish language, which is one of the chief and
principal causes of the continuance of barbarity and incivility amongst the
inhabitants of the Isles and Highlands, may be abolished and removed’.71 The
striking thing about such language legislation in the history of English is that
while it might lead to policies on education and business, by itself it has in fact
had little impact on language use and shift. The shift of Welsh speakers to
English, for example, depended more on the economic opportunities associ-
ated with the incursion of Anglophones into Wales than it did on the legal
restrictions on the status of Welsh. Indeed, the limitation of such laws is clear
from the fact that to this day, even though English is the oYcial language—
a language so declared by law—in neither the United Kingdom nor the
United States, Anglophones of both countries enjoy numerical superiority
as well as control over powerful domains of economics, education, and
government. For all these reasons, they also have the prerogative of channel-
ing non-linguistic issues through linguistic ones.
In predominantly Anglophone countries attempts to legislate the status of
non-English languages have also been more common of late and have
betrayed anxieties of their own about change and variation. If the English
Only movement’s eVorts to pass a Constitutional amendment declaring
English the oYcial language of the United States point to anxiety over
immigration and taxation, thus, the 1921 declaration of Irish as co-oYcial
with English in the Irish Free State uses language to signal a desire to eVect
political and cultural independence from England. This use of language
planning to channel concerns over national identity appears likewise in
Canada’s 1988 OYcial Languages Act, which helped to protect the oYcial
rights of Francophones and Francophone communities; the 1993 Welsh Lan-
guage Act, which did much the same for the status of Welsh in Wales; the 1992

70 Quoted in Glanville Price, The Languages of Britain (London: Edward Arnold, 1984), 106–7.
71 The Irish Act is quoted from Patricia Palmer, Language and Conquest in Early Modern Ireland:
English Renaissance Literature and Elizabethan Imperial Expansion (Cambridge: Cambridge University
Press, 2001), 137; and the Scots Act quoted is from Nettle and Romaine, Vanishing Voices, 140.
166 Language Anxiety

Native American Languages Act, which sought to promote and revitalize

Amerindian languages; and the 1996 South African Constitution, which
declared English one of that country’s eleven oYcial languages.
In such circumstances, English can be assigned so much symbolic, cultural
weight that it can serve as a reason for a non-native group to maintain its
linguistic variation from the Anglophones around them, even though the
primary point of contention remains non-linguistic. From the middle of the
nineteenth century, for instance, Chinese speakers raised concerns that edu-
cation in English, particularly through missionary schools, would be a vehicle
for introducing Western culture and values to China; and despite other
profound ideological diVerences, their communist successors raised these
very same concerns to support their own eVorts to retard the spread of English.
Contemporary speakers of indigenous African languages have expressed simi-
lar anxieties, singling out the World Bank and the International Monetary
Fund as institutions that advocate language policies to further their own non-
linguistic economic agenda. ‘The hidden push for English, in particular,’
Alamin Mazrui claims, ‘can be seen as part of a right wing agenda intended
to bring the world nearer to the ‘‘end of history’’ and to ensure the Wnal
victory of capitalism on a global scale.’72 In every one of these cases, the oYcial
sanctioning or proscription of variation between English and other languages
has served socially symbolic as well as—perhaps more than—communicative
ends, with language becoming salient conWrmation of an ethnic (or national)
group’s independent identity. While Anglophones have agitated about
the persistence of indigenous languages and indigenous speakers about the
introduction of English, for both groups language variation produces an
anxiety that centers, ultimately, on social issues.
Policies like these recall the religious policies I discussed above: they seek to
use change as a way of fostering linguistic and cultural stability that is
resistant to further change, and they invest language with the power not
simply to reXect culture but to transform it. In this way, such policies also
project just how much language can be made to mediate non-linguistic issues.
Whether or not one personally accepts the argument that the spread of
English is part of a vast, global, capitalist conspiracy, the very belief that it is
again testiWes for the anxiety associated with language change and variation.
Status planning on behalf of minority languages attests to this same anxiety.
By themselves, legislation and business activity can do little to accomplish

72 Mazrui, English in Africa after the Cold War (Clevedon: Multilingual Matters, 2004), 54. See also
Kingsley Bolton, Chinese Englishes: A Sociolinguistic History (Cambridge: Cambridge University Press,
2003), 226–58.
 Policy and Politics 167

or retard the revival of a dying language. Language revival must also be

backed by government or private funding for educational and economic
programs, and it must speak to ideas that are already broadly shared rather
than attempt to initiate them. In Cooper’s analysis of language planning
and national languages, ‘Symbols are created not by legislation but by history.
Irish and Kiswahili are not national symbols because the Irish and Tanzanian
constitutions proclaim them as national languages. They are national
symbols because of their association and identiWcation with their national
liberation movements and with their citizens’ shared memory.’73 While such
association and identiWcation can not be compelled by Wat on speakers, they
recall arguments about a global Anglophone conspiracy, in that they reXect a
strong belief in the power of change and variation to shape non-linguistic
concerns.
Shared memory, of course, is an elusive quality, often situationally deWned
and contested. Sociolinguistic developments in Australia oVer a Wnal case in
point. As immigration patterns changed following the Second World War,
showing less dependence on Great Britain, so, too, did internal conceptions of
the nation, which came to accept and advance an Australian variety of English
as well as a broader, multicultural identity. In advocating the maintenance
of Aboriginal and Torres Strait languages as well as the acquisition of a non-
English language by all Anglophones, the 1987 National Policy on Languages
formalized a memory of Australia as a multilingual place in which change and
variation were natural occurrences. This same memory was itself contested
four years later, in the 1991 Australian Language and Literacy White Paper,
which focused less on languages as cultural resources and more on English
literacy and other educational concerns. As Michael Clyne summarizes this
dynamic reformation of Australian national and linguistic identity, ‘Through-
out the history of Australia and the six British colonies that preceded the
federated nation, there has been an open-ended tension between English
monolingualism as a symbol of a British tradition, English monolingualism
as a marker of Australia’s independent national identity, and multilingualism
as a reXection of a social and demographic reality and of an ideology of an
independent multicultural and outreaching Australian nation’.74

73 Cooper, Language Planning and Social Change, 103.

74 Clyne, Dynamics of Language Contact: English and Immigrant Languages (Cambridge: Cambridge
University Press, 2003), 9. See also Uldis Ozolins, The Politics of Language in Australia (Cambridge:
Cambridge University Press, 1993); Michael Herriman, ‘Language Policy in Australia’, in Herriman and
Burnaby (eds.), Language Policies in English-Dominant Countries, 35–61; and Edwards, Multilingualism
in the English-Speaking World, 116–17.
168 Language Anxiety

Language, nation, identity

In modern Anglophonic states, government funded and regulated language
planning is predominant (if not pervasive), whether its focus is the language
used in classrooms, business, or courtrooms; the rights of immigrants and
the non-Anglophone languages they might speak; or the status of non-
Anglophone languages in states dominated by English, as is the case in
most (but not all) countries in which English is an indigenous language.
I have suggested ways in which this planning, while directed at variation and
change, has an ultimate goal both more general and non-linguistic, and in this
way channels anxiety about race, taxation, and other social issues towards and
through anxiety about language. To this end, American legislatures and law
courts have consistently avoided pronouncing the existence of something like
linguistic rights, such as the United Nations’ Universal Declaration of Lin-
guistic Rights and the European Charter for Regional or Minority Languages.
They have relied instead on the United States Constitution’s fourteenth
amendment—speciWcally, its guarantees of legal due process and the right
to pursue life and liberty—as the justiWcations for bilingual education and the
like. In this way, what are popularly envisioned to be linguistic issues are
redressed more generally as issues of civil rights. To understand these better,
I want now to place them within the context of how language Wgures in
the construction of nation and national identity. It is within this context as
well that I will return to the notion of individuals’ responsibility for their
government’s language planning on their behalf.
The role of language in national identity depends signiWcantly, of course,
on the conception of nationhood at a given time, and for England, glimmers
of both nationalism and linguistic nationalism appear throughout the pre-
modern period. In the prologue to his Wfteenth-century Troy Book, for
instance, John Lydgate claims that his patron Henry V wanted the Troy
story told in English so that all inhabitants of England might read it:
By-cause he wolde that to hyge and lowe
The noble story openly wer knowe
In oure tonge, aboute in every age,
And y-writen as wel in oure langage
As in latyn or in frensche it is;
That of the story the trouthe we nat mys
No more than doth eche other nacioun:
This was the fyn of his entencioun.75

75 Lydgate, Troy Book, ed. Henry Bergen, 4 vols., EETS es 97, 103, 106, 126 (London: Kegan Paul,
1906–35), Prologue, lines 111–18.
 Policy and Politics 169

There is a suggestion, here, of the importance of language in the construction

of a nation, but it is only that: a suggestion. Lacking the social structures
and ideology that make possible modern nationalism and a standard lan-
guage, neither Henry nor any of his immediate successors could do much to
cultivate a sense of linguistic nationalism: the population was predominantly
illiterate, the culture would continue for over a century to conduct much of
its education and oYcial business in Latin and French, and English itself,
while increasingly consistent in orthography, lacked codiWcation and the
formal support of education or standardization. All these characteristics
began to change in the seventeenth and eighteenth centuries, and the advent
of industrialization and its impact on social organization (e.g., increased
urbanization and bureaucratic infrastructure) accelerated the change still
more. It is in this period that, in Benedict Anderson’s famous analysis,
England, other European countries, and the Thirteen Colonies acquired the
intellectual and institutional means to imagine themselves as a community—
to conceive a group with shared interests and culture from among a collection
of individuals largely unknown to one another.76
Philosophically, the clearest early statements on the relation between lan-
guage and national identity are those of the eighteenth- and nineteenth-
century German Romantics Johann Herder, Wilhelm von Humboldt, and
Johann Gottlieb Fichte, who in various ways articulated connections between
blood, soil, and language. In Herder’s prize-winning essay ‘On the Origin of
Speech’, presented to the Berlin Academy of Sciences in 1770, he asked: ‘Has a
nationality anything dearer than the speech of its fathers? In its speech resides
its whole thought domain, its tradition, history, religion and basis of life, all
its heart and soul. To deprive a people of its speech is to deprive it of its one
eternal good . . . With language is created the heart of a people.’77 And in his
posthumously published but widely inXuential On Language, von Humboldt
provided a detailed argument of how language constitutes the deWnitive trait
of a nation and its people: ‘The mental individuality of a people and the shape
of its language are so intimately fused with one another, that if one were given,
the other would have to be completely derivable from it . . . Language is, as it
were, the outer appearance of the spirit of a people; the language is their

76 Anderson, Imagined Communities: ReXections on the Origin and Spread of Nationalism, rev. edn.
(London: Verso, 1991). For further consideration of the limitations of English linguistic nationalism
before the modern period, see Machan, English in the Middle Ages (Oxford: Oxford University Press,
2003), 161–78.
77 Quoted in Ronald Wardhaugh, Languages in Competition: Dominance, Diversity, and Decline
(Oxford: Blackwell, 1987), 54. On the role of language in nationalism, both historically and today, see
further May, Language and Minority Rights, 52–90.
170 Language Anxiety

spirit and the spirit their language; we can never think of them suYciently
as identical.’78 By these analyses, whether language causes nationhood or
reXects it may be moot, but without a common language there can be no
nation of shared culture and tradition, and without a nation, there can be
no formal, political state.
This view of the linguistic integrity of the nation-state has had considerable
impact on the anxiety associated with language change and variation. Indeed,
the legacy of the imagined communities that Anderson sees uniting modern
nations—a legacy reinforced by the Babel model of change and variation—is
that to be imagined, nations require a single dominant language and that,
therefore, any true nation perforce has such a language. The modern Western
conception of nationhood is often so dependent on the predominance of one
language, that by a kind of syllogism, multilingualism can be seen to under-
mine nationhood: if nations by deWnition are monoglot, then a multilingual
society cannot become a nation.79 And unless they somehow occurred for
entirely internal reasons, even language shift or radical structural change can
easily be seen to challenge the cultural and political integrity of a people and
their view of themselves. Given this line of reasoning, indeed, Grimm’s
explanation of the First Consonant Shift as an expression of Germanic
pride and impetuosity makes perfect sense: it could have occurred without
violating the social and linguistic integrity of the Germanic people only if its
occurrence was a sign of that integrity. And to Grimm it was.
Since Herder’s formulation of language as ‘the heart of the people’, of
course, notions of nationhood and statehood—and of ethnicity within
them—have become increasingly less stable and more controversial. One
common argument contends that it is ethnicity that is pre-modern, even
primordial, and that national or state groupings are imposed upon it, while
another essentially reverses these positions, maintaining that the existence of
modern nations leads to the situational fabrication of ethnicity. In suppress-
ing the separatist inclinations of minority language groups in the various
republics, for instance, the Soviet Union (it could be argued) did much to

78 Von Humboldt, On Language: the Diversity of Human Language-Structure and Its InXuence on
the Mental Development of Mankind, trans. Peter Heath (Cambridge: Cambridge University Press,
1988), 46.
79 See further Robert McColl Millar, Language, Nation and Power: An Introduction (New York:
Palgrave, 2005). As Sue Wright concisely puts the issue, ‘recognition of pluralism was at odds with
nationalist claims to be language communities with the distinctive and unique ways of thinking that
Herder had suggested. An early objective in the nationalist project was thus to achieve linguistic
convergence within the group and to diVerentiate the national language from all allied dialects on the
continuum’ (Language Policy and Language Planning: From Nationalism to Globalisation [Basingstoke:
Palgrave Macmillan, 2004], 35).
 Policy and Politics 171

foster those same inclinations, just as, more generally, ethnicity arises and
persists in modern, industrial society through what has been called an internal
colonial model and its division of labor along cultural rather than strictly class
lines.80 And yet another common argument describes ethnicity as neither
primordially nor politically determined, but instead as a process. In this view,
rather than some kind of Platonic category, ethnicity is a construction—and
not necessarily an intentional one—that over time interprets and associates
cultural activity in ways that allow ethnic identities to take shape in relation to
one another.81
All these arguments need to accommodate the complications posed by
linguistic change and variation. If nationality or ethnicity are in some sense
essentialist, for instance, there must be a way to account for the preservation
of identity despite alterations across time and space in a language’s grammat-
ical structure as well as in speakers’ mental conceptualizations of it. And if
they are not, if they are situationally deWned, we need a way to identify when
sociolinguistic phenomena achieve a kind of integrity that includes some
practices or individuals as variants and excludes others as categorically diVer-
ent. Recent attempts to get around these analytical complexities through a
category of hybridity, which focuses on overlaps and liminal areas in social
categories, solve some diYculties but create others, for the very category of
hybridity is itself open to the accusation that it simply oVers another version
of the ethnicity it attempts to evade.82
These are not airy, academic debates. Romantic notions of the immanence
of language in statehood remain inXuential to this day; a common motto of
Welsh language movements, for example, has been ‘cenedl heb iaith, cenedl
heb galon’—‘a nation without a language is a nation without a heart’.83 And
the foundational premise of the United States English Only movement is that
the ability to speak English is part of the deWnition of the country. Such
debates also have real Wnancial and social consequences for the conceptual-
ization of language change and variation and, in turn, for how this concep-
tualization relates to more general social concerns. Indeed, however ethnicity
has been formed in the modern era, minority languages have consistently

80 Andrew Dalby, Language in Danger: The Loss of Linguistic Diversity and the Threat to our Future
(New York: Columbia University Press, 2003), 124, and Michael Hechter, Internal Colonialism: The
Celtic Fringe in British National Development (New Brunswick, NJ: Transaction Publishers, 1999).
81 A good discussion of post-Conquest England from this perspective is Hugh M. Thomas, The
English and the Normans: Ethnic Hostility, Assimilation, and Identity 1066-c1220 (Oxford: Oxford
University Press, 2003).
82 See further May, Language and Minority Rights, 19–51, and Wardhaugh, Languages in Competi-
tion, 39–63.
83 Wardhaugh, Languages in Competition, 85.
172 Language Anxiety

helped to deWne and enforce the notion of a nation-state. If ethnicity has

been one way to maintain identity and social leverage in conglomerate,
immigrant societies, for instance, it has also been part of the cultural residue
that a nation, depending on circumstances, sees itself as repressing, transcend-
ing, or assimilating. In this way, a dominant group can style itself as the
nation partly through its use of minority groups and languages. According
to Tom Nauerby, then, ‘The national languages were actually the exact
opposites of what nationalist mythology supposed them to be . . . they were
not the primordial foundation of nations, but the more or less artiWcial
products of the nation-building process itself ’.84
By this kind of analysis, one ethnic group in eVect appropriates the concept
of nation, using its language as a way to deWne it. At one end of a spectrum for
implementing such a policy, one could place the Soviet Union or revolution-
ary France, both of which fairly aggressively, even brutally, advanced the cause
of the nation by suppressing minority languages and utilizing Russian and
French, respectively, as ways to stabilize political action. At the other end,
Norway and its treatment of Nynorsk has been oVered as a more benign
example of nation-building through language. Fashioned by Aasen in the
nineteenth century out of the Norwegian rural dialects that showed the least
linguistic inXuence of the Dano-Norwegian elite that had governed Norway
since the late fourteenth century, Nynorsk has been cultivated since then as a
landsmål, a speech deeply rooted in the land. Yet if in this way Nynorsk can be
said to have underwritten a concept of nation, even it has also served,
according to some critics, to maintain the privilege of upper-rank Norwegian
groups; in this scenario, an elite that does not speak Nynorsk justiWes its
status, in part, though a subordinated, imagined joint class of rural farmers
and urban laborers.85 Somewhere in between these extremes are most Anglo-
phone countries. In all of them, any group’s appropriation of statehood has
been and will remain aggressively contested, precisely because none of them
has one ethnic group so predominant that the country might be judged
monoethnic (like Iceland), wherein language, nation, ethnicity, and state
truly do coalesce. For both the United Kingdom and the United States, this
situation certainly hasn’t stopped individual groups from claiming cultural
primacy, but the mere fact that such claims have been so aggressively con-
tested of late undermines any individual claim.

84 Nauerby, No Nation Is an Island: Language, Culture, and National Identity in the Faroe Islands
(Aarhus: Aarhus University Press, 1996), 8.
85 See Gregg Bucken-Knapp, Elites, Language, and the Politics of Identity: The Norwegian Case in
Comparative Perspective (Albany, NY: State University of New York Press, 2003). On Nynorsk more
generally, see Haugen, Language ConXict and Language Planning.
 Policy and Politics 173

In such political models language variation and change are of great conse-
quence as correlates of ethnicity and nationhood. Since the latter are typically
identiWed in the ways they manifest themselves, as in clothing or ritual or,
indeed, language, arguments about these correlates often stand in for arguments
about social and cultural identity. Managing change and variation becomes a
way of managing social identity, then, whatever relation these phenomena
actually bear to it and however complexly that identity is fashioned. In
nineteenth-century Wales, for instance, contact between Welsh speakers and
Anglophones played out in a nexus of social issues of which language was only
one part. In the slate industry widespread in southern Wales, those who spoke
Welsh also tended to be renters who were politically liberal and religiously
non-conformist. By contrast, H. Paul Manning points out,
Lord Penchyn, the most inXuential of the owners, was combination rentier and capitalist,
owning the largest slate quarry (with a 40 percent market share) as well as having an
enormous landed estate of some 50,000 acres, which meant that he stood as both
employer and landlord to many of the workers, and was a Tory MP and an Anglican
to boot, at a time when Welsh Nonconformist Liberalism stood not only for home rule
for Ireland and possibly Wales, but also for Disestablishment of the church and for
the tenant-farmer (gwerin) against the passive landed wealth of the squirearchy.86

Talking about and controlling Welsh thus served indirectly to talk about and
control—mediate—a host of non-linguistic concerns.
Similar issues have appeared in contact between English and the roughly
250 indigenous languages in the Philippines, though there with a result quite
diVerent from that in Wales. In 1901, following the Spanish-American War,
1,000 Anglophone teachers came to the region as part of a concerted policy to
counter and constrain linguistic variation. By 1925 this policy did limit
variation by producing some shift to English, but it did little to improve the
educational achievement of native Filipinos, for a study at that time showed
that they lagged behind their American counterparts in reading and writing.
It was partly in response to this situation that when the Commonwealth of
the Philippines was established in 1935, English and Spanish were made co-
oYcial. Following the Second World War and Philippine independence in
1946, reaction against English as a colonial legacy increased and came to focus
on one indigenous language—Tagalog—as a symbol and vehicle of national
unity. Today, English is still widely spoken but mostly in a native variety

86 Manning, ‘The Rock Does not Understand English: Welsh and the Division of Labor in
Nineteenth-Century Gwynedd Slate Quarries’, in Brian D. Joseph et al. (eds.), When Languages Collide:
Perspectives on Language ConXict, Language Competition, and Language Coexistence (Columbus, OH:
Ohio State University Press, 2003), 48.
174 Language Anxiety

(Taglish).87 If the example of Wales presents a fairly consistent dichotomy

between powerful interests mediated by English and less powerful ones
mediated by Welsh, the example of the Philippines illustrates a continually
shifting sense of national identity reXected in the shifting relations of Spanish,
English, and Tagalog.
South Africa oVers a still more complex example of how managing change
and variation can stand in (often ineVectively) for managing issues of eco-
nomics, politics, and race. The Wrst white settlers appeared on the Cape in
1652, and by 1795 British Anglophones had taken control of the region. By the
early part of the nineteenth century, even though speakers of Dutch out-
numbered Anglophones by a ratio of Wve to one, attempts were made to
deWne the area through English. In 1811, for instance, knowledge of English
was speciWed as a pre-requisite for civil service. Dutch was declared co-oYcial
with English in 1882, but twenty years later this designation was restricted to
English alone. The dynamic between these languages—and Afrikaans, which
joined English and Dutch as co-oYcial in 1925—bespeaks an attempt to foster
national culture through language and, more signiWcantly, stands startlingly
out of synch with the linguistic and social realities of a region in which the
vast majority of the population spoke and speaks only one or more indigen-
ous African languages like Zulu. While these realities do Wgure largely in the
1996 constitution of the new South Africa, which declared eleven oYcial
languages and thus gestured strongly towards a multilingual, multiethnic
conception of nationhood, language management remains a limited means
for generating social identity. According to the 1996 census, English was only
the Wfth most commonly used language, with about 3.4 million speakers;
Zulu was Wrst, with 9.2 million, and Afrikaans third, with 5.8 million. When
use of a second language was calculated, the statistics changed in a reXection
of perceived social utility. Zulu remained Wrst, with 25 million speakers, but
English moved up to third, with 17.6 million speakers, and Afrikaans dropped
to fourth, with 16 million speakers. Within the additional contexts of strong
ethnic divisiveness and the institutionalization of a continuum of varieties of
English, the South African linguistic repertoire provides little support for an
argument that managing change and variation can be at all eVective as a way
of creating social identity as well: if given the opportunity, its speakers choose
languages not as national political statements but for the immediate social
utility they are understood to provide.88

87 See further Roger M. Thompson, Filipino English and Taglish: Language Switching from Multiple
Perspectives (Amsterdam: John Benjamins, 2003).
88 On the history of language policy in South Africa, see Nkhelebeni Phaswana, ‘Contradiction or
Assimilation? The South African Language Policy, and the South African National Government’, in
 Policy and Politics 175

For Wales, the Philippines, and South Africa, it might be said that language
planning has been relatively modest and that whatever failures it experienced
could thus be attributed to its own ineVectuality. The Soviet Union provides a
non-English illustration that is anything but tepid but in the end just as
inconclusive. By imposing a shift to Russian, Stalin could in many ways
accurately claim that he controlled the identity of the individual and collective
Soviet Socialist Republics. Yet the break-up of the Soviet Union and the
subsequent rejection by many former republics of Russian and the Cyrillic
alphabet provides a useful corrective to such claims. In its entirety the Soviet
example suggests that connections between language and social identity can
be as much by Wat as through Herder’s ‘heart of the people’, that language
policies meant to manage language change succeed or fail in sometimes
unpredictable ways, and that speakers always, Wguratively as well as literally,
have something to say about the language they use. They also suggest that
for all this, there is something deadly serious about individual and state
investment in the signiWcance of change and variation.
For the Soviet Union, deadly serious meant the deaths of millions of
innocent people. For the United States, it has meant a struggle since the
founding of the republic over the relations between English and nationhood
and all the ideas and practices that depend on these relations. Within this
context, anxieties about language change and variation have contributed
signiWcantly to the formation of national and ethnic identity. Originating
as it did in the eighteenth century, the United States emerged against a
distinctive linguistic background: the nationalism of Herder and the German
Romantics, the virtual completion of the codiWcation of written English, the
increased expansion of print, and a concomitant rise in literacy. Coupled with
the period’s proverbial emphasis on human thought and reasoning, such
characteristics lead John Howe to follow others in describing the founding
of America as ‘in a fundamental sense, a linguistic act . . . Embarked on an
unprecedented experiment in republican independence, members of the
revolutionary generation employed written language as an essential instru-
ment of political action through which they articulated political ideologies,
negotiated political conXict, recorded political accomplishments, and charted

Sinfree Makoni et al. (eds.), Black Linguistics: Language, Society, and Politics in Africa and the Americas
(London: Routledge, 2003), 117–31. On South African varieties of English, see Daniel Gough, ‘Black
English in South Africa’, in Vivian de Klerk (ed.), Focus on South Africa (Amsterdam: John Benjamins,
1996), 53–77; Susan Watermeyer, ‘Afrikaans English’, in de Klerk (ed.), Focus on South Africa, 99–124;
and Roger Lass, ‘South African English’, in Rajend Mesthrie (ed.), Language in South Africa (Cam-
bridge: Cambridge University Press, 2002), 104–26. More generally see Vic Webb, Language in South
Africa: The Role of Language in National Transformation, Reconstruction, and Development (Amster-
dam: John Benjamins, 2002); 68–88 oVer statistics on language use.
176 Language Anxiety

the political future.’89 Indeed, the formative years of the republic, between
the Revolution and the Civil War, witnessed an overriding concern with
language and what it meant to speak and write as an American. And this
concern was expressed not just by imaginative writers (e.g., Cooper and
Whitman) but also in educational and political circles, making it one of the
new country’s deWning discursive traditions.90 Howe characterizes opposing
poles of thought about language and politics in the period that once
more recall the Bible’s competing explanations of change and variation.
More signiWcantly—and familiarly—for the concerns of this chapter, both
views invest change and variation with terriWc non-linguistic signiWcance and
thereby transfer to language what are properly the issues of nationhood
and government organization.
At one extreme were writers who ‘utilized language as if it constituted a
Wxed and unvarying medium of expression existing apart from the changing
contexts of history, a medium stable in its grammar and vocabulary, certain in
its meanings, and unambiguous in its capacity to express universal truth’. In
a pre-Babel mode, such language would be resistant to change precisely
because of its powers of transparent expression; and by extension, linguistic
change and variation would imply the failure of human behavior as well as,
more simply, the instability of a political process. At another extreme, em-
bracing thinking more akin to the account of language change in Genesis 10,
‘other political writers understood that language, far from constituting an
autonomous realm of universal meaning separated from the Xux of history,
was inextricably embedded in human experience’. Language here serves as ‘an
instrument of political exploration and creativity to be deployed in the
construction of a continuously unfolding political future’.91 More than just
the inevitable byproducts of natural language, change and variation from this
perspective play crucial roles in the continual, situational adjustment of both
language and politics. Their absence would indicate, indeed, the unnatural
failure of political thought and action to keep pace with social development.
In the colonial period and early days of the republic, up through the
Constitutional Convention of 1787, belief in the constancy of language (spe-
ciWcally political language) prevailed. In 1776, the year of the Declaration of

89 Howe, Language and Political Meaning in Revolutionary America (Amherst, MA: University of
Massachusetts Press, 2004), 2.
90 See further Michael P. Kramer, Imagining Language in America: From the Revolution to the Civil
War (Princeton, NJ: Princeton University Press, 1992). On the role that a feeling of discontinuity from
the past played in this discourse, see John Algeo, ‘External History’, in Algeo (ed.), English in North
America, 1–58.
91 Howe, Language and Political Meaning in Revolutionary America, 5.
 Policy and Politics 177

Independence, John Jay even described a homology between political unity,

the new country, Providential design, and English: ‘Providence has been
pleased to give this one connected country, to one united people; a people
descended from the same ancestors, speaking the same language, professing
the same religion.’92 It is this belief that underwrote John Adams’s advocacy of
an American language academy. Comparing the United States to Athens
and Rome and positing as necessary a connection between democracy and
language, he once located motivation for such an academy in English’s role as
the world’s pre-eminent tongue: ‘As eloquence is cultivated with more care in
free republics than in other governments, it has been found by constant
experience that such republics have produced the greatest purity, copiousness,
and perfection of language. It is not to be disputed that the form of govern-
ment has an inXuence upon language, and language in its turn inXuences
not only the form of government, but the temper, the sentiments, and
manners of the people.’93 Neither the conviction in the Wxedness of language
nor the commitment to an American academy died with Adams. In 1889,
indeed, an essay in the North American Review still called for the creation
of the latter: ‘No other means will so eVectively secure unity, prevent section-
alism, and abolish dialects.’94
When colonial writers like Charles Brockden Brown advocated only the
gradual change of language, and not its revolution and reconstruction, dis-
approval and avoidance of neologisms, semantic extension, metaphor, and
the like acted not merely as idle stylistic gestures. They expressed, rather,
a sense that the social and political processes to which these tropes pointed
should themselves be slowed.95 This approach to language, politics, and
change explains why an 1857 newspaper editorial would see the primary
contribution of Noah Webster, the founding father in any history of American
English, to have been the preservation of putatively absolute constancy and
regularity in American speech: ‘Here, Wve thousand miles change not the
sound of a word. Around every Wreside, and from every tribune, in every Weld
of labor, and every factory of toil, is heard the same tongue. We owe it to
Webster.’96 Precisely this same point about the immanence of linguistic

92 Quoted in May, Language and Minority Rights, 209.

93 ‘A Letter to the President of the Congress’ (5 September 1780), quoted in Crawford (ed.),
Language Loyalties, 31. On the early emphasis on the constancy of language, see Howe, Language
and Political Meaning in Revolutionary America, 38–62.
94 Quoted in David Simpson, The Politics of American English, 1776–1850 (Oxford: Oxford Univer-
sity Press, 1986), 13.
95 Simpson, The Politics of American English, 1776–1850, 46–7.
96 Quoted in Harlow Giles Unger, Noah Webster: The Life and Times of an American Patriot (New
York: John Wiley and Sons, 1998), xii.
178 Language Anxiety

constancy in political stability—and, more importantly, about how instability

in one produces instability in the other—was made when the regularity
attributed to Webster’s inXuence was denied to American English, and with
it its implications for any American social achievement. In 1864, for example,
the dean of Canterbury collapsed linguistic and moral failure in his assess-
ment of the American Civil War: ‘Look at the process of deterioration which
our Queen’s English has undergone at the hands of Americans . . . and then
compare the character and history of the nation—its blunted sense of moral
obligation and duty to man . . . and its reckless and fruitless maintenance of
the most cruel and unprincipled war in the history of the world.’97 In a more
benign version, the use of change and variation to reXect marginal, inconstant
behavior—and thus by implication to associate linguistic and political regu-
larity with each other—emerges from the tendency of the early novelists
like James Fenimore Cooper, very much unlike contemporary writers like
Kelman and Welsh, to use non-standard and dialect forms primarily with
minor characters and never with his narrators.98
Already during the War of Independence this belief in the constancy of
political language was challenged. If such language were stable, Wxed, and
transparent, after all, it becomes diYcult to account for the breach of trust
and conWdence that led the colonists to distrust the king and to foment what
they viewed as justiWed revolution. For colonial America, stable language had
resulted in political failure, which in turn implied the importance of adapt-
able, Xexible language that was suited to speciWc situations. Buoyed by the
emergence of federalism and its emphasis on a new, centralized government,
Howe’s other extreme of linguistic thinking gradually asserted itself. ‘Feder-
alists employed language in ways that were politically inventive’, he notes, ‘as a
creative instrument essential to the ongoing task of exploring America’s
evolving republican experiment. The triumph of Federalism in 1787–88
brought the legitimation, not just of a ‘‘new science of politics’’, but of a
new political language as well.’99 If Adams and his advocacy of an American

97 Quoted in GeoVrey Nunberg, Going Nucular: Language, Politics, and Culture in Confrontational
Times (New York: PublicAVairs, 2004), 150.
98 Simpson, The Politics of American English, 1776–1850, 149–201. Simpson points out, however, that
Natty Bumpo, Cooper’s protagonist in the ‘Leatherstocking’ series, is a special case as a ‘dialect speaker
who is not simply subordinated to the polite usages and values of the socially superior characters, and
who seems to bear about him a deWnite degree of authorial approval and conviction’ (168). On the role
of English in the literary formation of early American identity, also see Christopher Looby, Voicing
America: Language, Literary Form, and the Origins of the United States (Chicago, IL: University of
Chicago Press, 1996).
99 Howe, Language and Political Meaning in Revolutionary America, 224–5. On the early view that
standard languages were a way for European elites to maintain social privilege and hierarchy, see Gray,
New World Babel, 159–63.
 Policy and Politics 179

language academy epitomize arguments for linguistic stability, Thomas

JeVerson was one of the most prominent of the early voices for a new
approach to language change and variation. A polymath who studied Ameri-
can Indian languages and advocated the study of Old English as a requirement
at the newly founded University of Virginia, JeVerson championed what
he called ‘neology’, which encompassed semantic expansions as well as neolo-
gisms. In a letter to Adams, indeed, he maintained that without neology
‘we should still be held to the vocabulary of Alfred or Ulphilas; and held to
their state of science also’. In an 1813 letter to John Waldo, JeVerson laid open
the political implications of this view of change and variation. Language in
England might remain as constant and immutable as the country and political
process it supported, but as for Americans, ‘we shall . . . enlarge our employment
of it, until its new character may separate it in name as well as in power, from the
mother-tongue’.100 For the American experience to have been reasonable and
politically meaningful, language had to adapt. Yet if language change thereby
produced anxiety in its absence rather than its presence, the anxiety itself
remained just as palpable in the implication of linguistic change and variation
in what are essentially non-linguistic issues of nation formation.
As Alexis de Tocqueville would later champion it, by this view America
would be a place where language, like society, would be given free rein,
develop communally, and show a consistency devoid of sociolinguistic hier-
archies. In his widely read and inXuential Democracy in America (1835), de
Tocqueville, very much under the inXuence of Herder and the German
Romantics, not only posited ‘democracy’ and ‘language’ as essentialist cat-
egories but also presumed an essentialist connection between them. In aris-
tocracies, such as those that still prevailed in Europe, ‘language must naturally
partake of that state of repose in which everything remains. Few new words
are coined because few new things are made.’ But in democracies like the
United States, there is a ‘competition of minds’ and ‘many new ideas are
formed’. Change, linguistic as well as political, is what deWnes democracies
and advances them beyond aristocracies: ‘democratic nations love change for
its own sake, and this is seen in their language as much as in their politics.
Even when they have no need to change words, they sometimes have the
desire.’ And given de Tocqueville’s sense that democracies are fundamentally
populist political systems, it follows that their many new words—and the
ideas they express—will not be intellectual or philosophical (as in aristocra-
cies) but mostly words taken ‘to express the wants of business, the passions
of party, or the details of the public administration’. The language of a

100 Quoted in Howe, Language and Political Meaning in Revolutionary America, 80, 82.
180 Language Anxiety

democracy will thus continually grow in these areas, ‘while it will gradually
lose ground in metaphysics and theology’. By extension, if democracies char-
acteristically lack the social divisions of aristocracies, because in democracies
individuals of all social ranks intermingle, so, too, must the language of
democracies like America lack the social and regional variation found in
Europe. De Tocqueville’s argument is thus almost hermetically sealed, express-
ing the same anxiety over change and variation that JeVerson had: whatever
changes de Tocqueville did observe in American English—and English was not
a language he himself spoke—had to reXect only the simple fact that the
United States was a democracy.101
The most articulate and frequent advocate of this position, of course, was
Webster. Unlike JeVerson, Webster was a federalist, but the two shared a
commitment to the controlled change of language as a necessary aspect of
the new country’s political identity. In fact, for Webster the transition from
federalism to linguistics was slight. He believed strongly, for example, in a
standard, written language that was in many ways the grammatical equivalent
of federalism’s centralized government, and it is from this political perspective
that he wanted English to change so as to reXect America. And even though
the advent of historical linguistics and the work of comparativists like
Sir William Jones had generally displaced belief in Babel and its account of
a single language scattered by divine judgement, Webster held tightly to the
monogenesis of language. Indeed, he espoused beliefs in the evolution of
all modern languages from the Chaldean spoken by Noah and in the descent
of European languages from Japheth’s lineage in particular.102
As has been characteristic of the language anxiety that I have explored in
this book, contradictions run through Webster’s positions on language, lan-
guage change, and the relation of the two to American character. On one
hand, he saw written and spoken English as the expression of the people in
general, which in the United States still meant a primarily agricultural popu-
lation. Just as in de Tocqueville’s analysis, this agricultural base would work
against divisions of labor and the creation of a social elite, whose language use
a dictionary like Johnson’s could be seen to support.103 The practical means

101 De Tocqueville, Democracy in America, ed. Phillips Bradley, 2 vols. (New York: Vintage, 1945),
ii, 68–74. De Tocqueville’s notion that as a putatively classless society, the United States necessarily
lacked regional and social variation was earlier expressed by the American Timothy Dwight. See
Frederic G. Cassidy, ‘Geographical Variations of English in the United States’, in Richard W. Bailey and
Manfred Görlach (eds.), English as a World Language (Ann Arbor, MI: University of Michigan Press,
1982), 187.
102 Unger, Noah Webster, 286.
103 Simpson, The Politics of American English, 1776–1850, 59, 70–1.
 Policy and Politics 181

for expressing this egalitarian sociolinguistic identity of American English

took shape in his grammar, spelling-book, and dictionary. For Webster these
linguistic aids, appearing at a time when literacy was increasing and public
education becoming more generally available, if still rather limited, not
only diVerentiated and regulated American English but also served cultural,
political, and moral purposes:
We have therefore the fairest opportunity of establishing a national language, and of
giving it uniformity and perspicuity, in North America, that ever presented itself to
mankind. Now is the time to begin the plan. The minds of the Americans are roused
by the events of a revolution; the necessity of organizing the political body and of
forming constitutions of government that shall secure freedom and property, has
called all the faculties of the mind into exertion; and the danger of losing the beneWts
of independence, has disposed every man to embrace any scheme that shall tend, in
its future operation, to reconcile the people of America to each other, and weaken
the prejudices which oppose a cordial union.104

But on the other, as much as Webster aYrmed the importance of language

change to the identity of the American republic, he regarded this change as
best restricted in scope and directed towards the end not of a continually
evolving language but of one evolved to the point that, like British English in
JeVerson’s view, it might remain a Wxed reXection of a Wxed country. In this he
recalls not only JeVerson but also the Lollards, King Alfred, and modern
language legislation. As early as 1785, while stressing the importance of
creating a national language that would demand respect at home and abroad,
Webster observed: ‘Nothing but the establishment of schools and some
uniformity in the use of books can annihilate diVerences of speaking and
preserve the purity of the American tongue.’105 He likewise felt that just as
the egalitarian, agricultural basis of American society worked against social
and regional variation in language, so, too, this basis would lead to the
atrophy and elimination of dialects.106 In his Dissertations, while identifying
and excoriating various regional pronunciations and linguistic habits, Web-
ster indicates that such diVerences ultimately owe to carelessness and are
thus correctable: ‘Great eVorts should be made by teachers of schools, to make
their pupils open the teeth, and give a full clear sound to every syllable.
The beauty of speaking consists in giving each letter and syllable its due
proportion of sound, with a prompt articulation.’107 It is for such reasons,

104 Webster, Dissertation on the English Language, 36.

105 Quoted in Unger, Noah Webster, 96.
106 Simpson, The Politics of American English, 1776–1850, 77.
107 Webster, Dissertation on the English Language, 109.
182 Language Anxiety

too, that he omitted dialect words—crucial and salient markers of variation—

from his dictionary.
The popular traces of Webster’s commitment to linguistic change are
largely limited to orthographic variants like theater, honor, and jail in the
United States as opposed to theatre, honour, and gaol in the United Kingdom
and much of the former Empire. Far more signiWcant has been Webster’s
resistance to change and variation. In this vein, it’s worth recalling that
his grammar and spelling-book were often packaged and sold with a copy
of the Bible, thereby linking together physically as well as conceptually
regularity in language and morality. Moreover, like early modern vulgaria,
which inculcated Latin in schoolboys through sentences about the privileges
and intelligence of the wealthy, Webster’s speller, used in virtually every
nineteenth-century American school, drove home its orthography alongside
adages on nation, virtue, and the common good. As an illustration of love,
Webster oVered ‘The love of God is the Wrst duty of man’, and as an illustration
of patriotism, ‘Patriotism is the characteristic of a good citizen, the noblest
passion that animates a man in the character of a citizen’.108 As the 1981 Select
Committee on Immigration citation of him suggests, Webster the social
conservative has remained most memorable among thinkers who have in-
voked language as a necessary medium of forming the republic and main-
taining its unity amidst the social, religious, and ethnic diversity of the United
States. It is an irony of many kinds that in 1859, on the eve of the war that
would fracture the country, JeVerson Davis, the future president of the
Confederacy, should likewise have seen Webster and his linguistic ministra-
tions as symbols of the states’ unity: ‘Above all other people, we are one,
and above all books which have united us in the bond of common language,
I place the good old Spelling-Book of Noah Webster. We have a unity
of language no other people possesses, and we owe this unity, above all else,
to Noah Webster’s Yankee Spelling-Book.’109
Webster can serve as such an ambivalent symbol of linguistic constancy and
change, and of the dangers and beneWts commonly attributed to the latter,
precisely because the issues for which he has been appropriated are linguis-
tically unresolvable. They are not, in fact, primarily linguistic at all but rather
social concerns Wltered through language change and variation. In United
States language planning at least, they dynamically and necessarily supple-
ment each other as modern-day evocations of the conXicting accounts of
change and variation in Genesis 10 and 11. What drives this dynamic in the

108 Quoted in Unger, Noah Webster, 306; also see 344.

109 Quoted in Unger, Noah Webster, 343.
 Policy and Politics 183

United States and elsewhere are the connections between language policy and
ethnic—and then national—identity. Language planning, and policies on
change and variation in particular, serves as a heuristic for explaining history
and social interaction. To a modern pluralist, Ronald Schmidt points out, the
dominance of English is ‘the unjust result of a very unequal competition
between diVerent language groups’.110 To assimilationists, this same domin-
ance is not only natural and appropriate but besieged by non-Anglophone
individuals and the policies that allow and encourage them to remain so. Like
Norman chroniclers, assimilationists can fabricate a monolingual American
past, just as they attribute variation and change to external, disruptive forces.
For pluralists and assimilationists alike, the euphemistic role of language in
such discussions becomes especially prominent at moments of social crisis.
A few months after the United States entered World War I, for instance, then-
former president Theodore Roosevelt authored a jingoistic document entitled
‘The Children of the Crucible’ that emotively collapsed patriotism, history,
and national identity into the English language: ‘We must also have but
one language. That must be the language of the Declaration of Independence,
of Washington’s Farewell address, of Lincoln’s Gettysburg speech and second
inaugural. We cannot tolerate any attempt to oppose or supplant the language
and culture that has come down to us from the builders of the Republic
with the language and culture of any European country.’111 Thirty-nine
distinguished citizens (including Roosevelt) signed ‘The Crucible’.
While there may be much to choose from here in terms of the sentiments of
such arguments, there is in fact little diVerence between their strategies and
consequences. The English Only movement’s push for a constitutional
amendment designating English the oYcial language of the United States
may well be demagogic and even hysterical. But so, too, are arguments that
make blithe, unsupportable generalizations about historical shift and that
describe a vast Anglophone conspiracy insidiously spreading capitalism and
genocidally erasing indigenous cultures by means of the Internet, instruction
in English as a foreign language, and global economics.112 Such extremes
depend on a disregard for linguistic history, imagine language change as a
problem that government language policy creates and can correct, credit

110 Schmidt, Language Policy and Identity Politics in the United States, 77.
111 Quoted in Crawford (ed.), Language Loyalties, 85.
112 See, for example, Thomas Ricento (ed.), Ideology, Politics and Language Policies: Focus on English
(Amsterdam: John Benjamins, 2000); Phillipson, Linguistic Imperialism; Skutnabb-Kangas and Phil-
lipson (eds.), Linguistic Human Rights; Norman Fairclough, Language and Power (London: Longman,
1989); Robert Hodge and Gunther Kress, Language as Ideology, 2nd edn. (London: Routledge, 1993);
Tollefson, Planning Language; and Pennycook, The Cultural Politics of English as an International
Language.
184 Language Anxiety

language planning with far more initiative and inXuence than has historically
been the case—how much real impact has the Académie Française, for
example, had on the direction of spoken French?—and, ultimately, transfer
responsibility for how change and variation are conceived from individuals to
institutions.
Throughout this book I have suggested that as linguistic phenomena,
change and variation have no intrinsic meaning. They are, in an absolute
sense, as value-free as a phoneme, a piece of inXectional morphology, or even
a word as the construction of phonology and morphology. By the same
token, like all these phenomena change and variation can be invested with
great meaning that in turn has great social consequence. If words like gay or
nigger become socially charged or taboo, they do so not because of anything
inherent in their phonological or morphological shape. Even their semantic
reference alone fails to account for their eVect, since meaning changes dia-
chronically as well as varies synchronically and since there are other socially
acceptable words (such as homosexual) that have the same general referents.
It is their pragmatics—the ways in which speakers use them—that renders
words powerful, sacred, benign, obscene, racist, and so forth. Should a word
like nigger suddenly disappear, racism would not perforce disappear with it,
any more than sexism disappeared with the spread of Ms at the expense of Mrs
and Miss. Natural language is an inherently creative system, and humans
are inherently creative beings; they always have found and always will Wnd
ways to express their thoughts. So long as speakers have a social or intellectual
need for them, new words can always be found to bear the pragmatic weight
speakers desire. Arguments over whether change is natural or unnatural,
whether political language is stable or adaptive, and whether Webster was a
linguistic innovator or preservationist remain unresolvable because they fail
to account for the fact that in these cases language points to something
else. And the something else, ultimately, is where opinions truly divide.
Language policies, like literary eVects, are ways to generate pragmatic
weight and invest social meaning. SpeciWcally, in their treatment of language
change and variation they have been ways for speakers to construct ethnicity,
nationality, and statehood. As all the examples of this chapter suggest,
they have played extremely important roles in this regard, underwriting
both the intellectual achievement of King Alfred’s revival of learning and
the moral degradation of the forced removal and language shift of non-
Anglophone children in Australia and the United States. When speakers
argue about the signiWcance of change and variation, as in the early days of
the United States, they argue about fundamental issues of history, identity,
and the future. At the same time, since the meaning of change and variation is
 Policy and Politics 185

what speakers invest in it, such arguments never address what generates this
meaning. Almost perversely, the mere fact that language can be judged to
have so much non-linguistic signiWcance proves not this signiWcance but
that the issues are not linguistic. And just as the elimination of racist words
would not eliminate racism, so the elimination of language change and
variation would not eliminate the disputes over identity that they channel.
It is again worth recalling that God visited His punishment upon Nimrod at
a time when there was no change and variation, when communication was
so complete as to allow for the building of the Tower.
Neither the hysteria of the English Only movement nor that of the augurs
of linguistic imperialism does much to alleviate the anxiety associated with
change and variation. It does much, in fact, to augment it. It obscures the
fact that pragmatics is an individual as well as social way to generate meaning
and that from this perspective individuals are as much the producers of
language as its victims. In every aspect of language use—phonemic split,
syntactic expansion, lexical diVusion—individual speakers provide the
usage that constitutes the data of a language, whether one’s interests are the
data themselves or the data as means towards an abstraction like the langue or
an ideal speaker-listener. In the case of language planning, the inXuence
of religious and government policies on how individuals might view change
and variation is considerable. And popular entertainment as well as court
decisions and school policies can lead speakers to become complicit in
the subordination of their own variety, eager to shift from Maori, Welsh,
or Spanish, for example, to English as a language whose communicative,
economic, and even ethical superiority they readily accept.113
But an important distinction is to be drawn here. In eVect, a minority
speaker’s shift to English may well be coerced by circumstances and economic
necessity; this was certainly the case with speakers of Hindi and Native
American languages. Decisions to invest meaning in change and variation,
to adhere to language planning as a way of redirecting other concerns about
ethnic and national identity, and to claim prescience about the sociolinguistic
future are not coerced in this fashion. They are matters of choice for individ-
uals as well as policy-makers. And prior to seeing change and variation as
conWrmations of a preconceived national character or as symptoms of decline
and dissolution, speakers have yet another choice. And that is the choice not to
channel non-linguistic anxieties through language and thereby displace the
genuine issues of concern.

113 For an analysis of how linguistic subordination can work, see Rosina Lippi-Green, English with
an Accent: Language, Ideology, and Discrimination in the United States (London: Routledge, 1997).
 5

Say the Right Thing

When a dictionary’s not enough

‘The rock-bottom practical truth’, Wilson Follett once wrote, ‘is that the
lexicographer cannot abrogate his authority if he wants to. He may think of
himself as a detached scientist reporting the facts of language, declining
to recommend use of anything or abstention from anything; but the myriad
consultants of his work are not going to see him so . . . the work itself by
virtue of its inclusions and exclusions, its mere existence, is a whole universe
of judgments, received by millions as the Word from on high.’1 By this
reasoning, in orthography, sense, and usage, English is what the dictionaries
say it is, and while thus general and theoretical in scope, Follett’s observation
actually responded to a very speciWc event: the September 1961 publication
of Webster’s Third New International Dictionary. Deriving ultimately from
Noah Webster’s 1828 American Dictionary of the English Language, which in
turn derived from his 1806 Compendious Dictionary of the English Language,
Webster’s Third (as it was and is called) represented a tradition of Wnancial,
linguistic, and even cultural achievement. For many Americans—and, for
that matter, readers worldwide—the Merriam-Webster company stands next
only to Oxford University Press as the pre-eminent maker of dictionaries,
and the cry ‘Look it up in Webster’s’ serves as an emblem of the authority
invested in it and in dictionaries in general.
Webster’s Third was a long time in the making. A second edition had
appeared in 1934, and work on the third was well under way when Philip
Gove assumed editorial responsibility in 1951. A scholar of eighteenth-century
literature by training, Gove was very much a traditionalist, committed to

1 Quoted in Herbert C. Morton, The Story of ‘Webster’s Third’: Philip Gove’s Controversial Diction-
ary and Its Critics (Cambridge: Cambridge University Press, 1994), 189–90. The following account
draws on Morton’s splendid book, as well as on James Sledd and Wilma R. Ebbitt (eds.), Dictionaries
and THAT Dictionary: A Casebook on the Aims of Lexicographers and the Targets of Reviewers (Chicago,
IL: Scott, Foresman, 1962).
 Say the Right Thing 187

preserving qualities that had made Merriam-Webster a household name, but

also to altering the form and content of the dictionary in ways that reXected
how English had changed in the quarter century since the previous edition.
He thus modiWed the phonetic transcriptions and the procedures for cross-
referencing and the like, eliminated proper nouns, and added 100,000 new
words (as well as many new senses) to what had appeared in 1934. He likewise
worked to eliminate impressionistic racial or ethnic judgements from the
deWnitions (Webster’s Second had deWned the Maori, in part, as ‘vigorous and
athletic, tall in stature, and pleasing in features, and brave and warlike’) and,
in general, to substitute concise and logical phrasal deWnitions for the some-
times loosely structured and expansive deWnitions of the previous edition. By
these means, Webster’s Third appeared as a formidable symbol of its own
authority: 450,000 entries contained in 2,662 pages (not including a 72-page
preface outlining the book’s methods and conventions) that weighed 13½ lbs
and was listed at $47.50. Even for a book issued today, such statistics would be
impressive; for one issued nearly a half-century ago, they are extraordinary.
For all this, much of the volume’s early reception bordered on the hostile,
attributing to Gove (and Webster’s Third) not merely inaccuracy but malfea-
sance and outright, willful ineptitude. Elsewhere in his review, indeed, Follett
charged the dictionary with jettisoning ‘a century and a third of illustrious
history’ and claimed that ‘it plumes itself on its faults and parades assiduously
cultivated sins as virtues without precedent’. Webster’s Third, he asserted, ‘is
out to destroy . . . every obstinate vestige of linguistic punctilio, every surviv-
ing inXuence that makes for the upholding of standards, every criterion for
distinguishing between better usages and worse’. Even more sardonically and
personally, David Glixon charged, ‘It would seem that permissiveness, now on
the wane in child-rearing, has caught up with the dictionary makers. Having
descended from God’s throne of supreme authority, the Merriam folks are
now seated around the city desk, recording like mad.’2 According to an
editorial in the New York Times, ‘Webster’s has, it is apparent, surrendered
to the permissive school that has been busily extending its beachhead on
English instruction in the schools. This development is disastrous because,
intentionally or unintentionally, it serves to reinforce the notion that good
English is whatever is popular.’3 Not all reviews were so critical or vitupera-
tive. Many academic journals, particularly those in Britain, praised the book
for its scholarly care, imagination, and erudition. But the negative reviews,
often in prominent forums like the New York Times or Life magazine, reached

2 Quoted in Morton, The Story of ‘Webster’s Third’, 187–8, 172.

3 Quoted in Sledd and Ebbitt (eds.), Dictionaries and THAT Dictionary, 78.
188 Language Anxiety

the widest audience and largely set the agenda for any subsequent, positive
reviews, by placing them in the position of Wrst having to defend the diction-
ary from the accusations made against it.
There are several oddities here. For one thing, the most critical and widely
read reviews were often the most factually inaccurate. For example, Dwight
Macdonald in the New Yorker blistered the dictionary as failing to follow the
model of its predecessors, slavish to fashionable linguistics, and biased against
historical sources in its citations. Yet as often cited as this witty review was,
it was also profoundly ignorant about lexicography, the methods and contents
of Webster’s Second, and even the design and contents of Webster’s Third itself.
Macdonald simply misunderstood the book’s citation and cross-referencing
procedures, and he failed to appreciate that a very traditional source like
Shakespeare was the most commonly cited authority—cited three times
more frequently than the next most common authority, which happened
to be the even more traditional Bible. For another thing, at issue were not
simply Webster’s Dictionary but also the Merriam-Webster company, both of
them venerable in American letters and therefore, one would think, unlikely
institutions to elicit the lexicographical equivalent of road rage. And for a
third, the cause of the New York Times’s indignation and its readers’ schaden-
freude was neither social policy nor legislation but, ultimately, just a book—
a dictionary—which reviews credited with the ability to cause not only a
language but a society to rise and fall.
Macdonald, Follett, and others clearly touched a nerve in their assertion
that readers saw dictionaries as unimpeachable and absolute authorities
on language—in Follett’s phrase, as ‘a whole universe of judgments, received
by millions as the Word from on high’. Whether dictionaries ever have had
or even claimed this authority is perhaps immaterial, for it is readers
who have invested them with it. And this disconnect between lexicographers’
and readers’ expectations surely produced at least some of the latter’s frus-
tration and outrage. Webster’s Third did indeed decline to make some of the
judgements of which Follett speaks; the label ‘colloquial’ was omitted, for
instance, while ‘slang’ was greatly restricted on the theory that the distinction
it implies—between slang and non-slang—isn’t as Wxed or universal as
one might imagine. The dictionary’s deWnition of ain’t is a case in point. In
the sense ‘am not’, the deWnition stated, ain’t was ‘disapproved by many and
more common in less educated speech, used orally in most parts of the U. S.
by many cultivated speakers esp. in the phrase ain’t I’. The dictionary labeled
the sense ‘have not’ as substandard. This could well be regarded as a model of
restrained, non-judgmental description of linguistic reality—many people
really do say ain’t—and yet many reviewers interpreted the deWnition as
 Say the Right Thing 189

a prescriptive grammar claim that ain’t, a slang word, was now perfectly
acceptable, non-stigmatized language. They saw the dictionary as in eVect
advocating the word. And in this curious way, the deWnition became a symbol
of everything that was putatively wrong with Webster’s Third as a book that
refused to make any value judgements, that never distinguished prestige from
non-prestige forms, and that in general furthered permissiveness in society
as well as language. Merriam-Webster itself may also have been responsible
for some of these misreadings and reactions. Its publicity program for Web-
ster’s Third presented the book, inaccurately, as not simply revolutionary in
methods but also directed to families as an arbiter of language—the sort of
book to be kept beside the dining-room table for reference and word games.
The company thereby encouraged fundamental confusion already present
among readers (and Merriam-Webster) who failed to distinguish prescriptive
grammar, which was what Follett and others sought, from descriptive gram-
mar, which is what Merriam-Webster and Gove sought and in fact gave.
But even this confusion would seem to fail to account for the controversy
and moral indignation that Webster’s Third incited. The battle metaphor of a
‘beachhead’ in the New York Times, read by a generation who remembered
and may well have participated in the landings at Normandy and in the
PaciWc, seems designed to foster just this response. Glixon’s rhetoric perhaps
goes even further, emotively conXating the dictionary with bad parents and
fallen angels. In reference to such responses, Herbert Morton points out that
‘the central issue was not merely the dictionary itself, though it was that
primarily; it was also what the critics thought the dictionary symbolized. At
stake, so it was made to appear, was the preservation of the English language
and the survival of deeply rooted cultural traditions.’4
Dictionaries have perhaps always had unusual authority and consequence
in the United States, where they might be seen as substitutes for the language
academy for which John Adams argued unsuccessfully.5 Yet dictionaries in
general necessarily play prominent roles in shaping attitudes towards lan-
guage not only by specifying words’ meanings but also, more generally, by
including or excluding particular words and by specifying status and usage
conventions for those that are included. Drawing on Foucault’s notions of
discursive conventions and how they construct the identity of the objects they

4 Morton, The Story of ‘Webster’s Third’, 2.

5 Allen Walker Read, ‘The Allegiance to Dictionaries in American Linguistic Attitudes’, in Mile-
stones in the History of English in America, ed. Richard W. Bailey (Durham, NC: Duke University Press,
2002), 110–20. There was some hope, too, that the Oxford English Dictionary would serve in place of a
language academy in the United Kingdom. See Lynda Mugglestone, Lost for Words: The Hidden History
of the Oxford English Dictionary (New Haven, CT: Yale University Press, 2005), 143.
190 Language Anxiety

describe, John Willinsky thus observes, ‘At heart, the dictionary is a school-
book; it contributes to the disciplining of language, as easily as to its study’.6
It is language’s ability to discipline social experience, in turn, that constitutes
its social power and seems to underlie the anxieties generated by Webster’s
Third. By eVectively deWning language—disciplining it to exist in a certain
way, and thereby turning it into a discipline—dictionaries can serve as tools
for discriminating among educated and non-educated speakers and, by ex-
tension, social classes and concomitant economic and political inXuence.
Put another way, practical decisions about contents, the rhetoric of deWni-
tions, labeling, and even pronunciation guides have implications for concep-
tions of morality, propriety, and social class.7 By extension, to know what is in
an English dictionary—better still, to control what is in a dictionary—is to
know what English is, which increasingly since the early modern period has
itself been knowledge of terriWc social consequence. In this vein, the dispute
over the alleged permissiveness of Webster’s Third seems very much an
expression of post-war America’s anxiety over ‘the survival of deeply rooted
cultural traditions’, whether they be traditional gender roles, child-rearing,
the value of education, high culture, or capitalism. At the very moment when
the Soviet Union seemed to be winning the space race—Yuri Gagarin had
become the Wrst man in space in April 1961, four months before the dictionary
was published—the New York Times in fact mockingly described Webster’s
Third as ‘Bolshevik’.8
The largest issue here is metalinguistics, or linguistics that concerns itself
with linguistics. Like any discipline, linguistics, alongside pursuing its inquiry
of language phenomena, is in part focused on deWning its methods of inquiry
as well as those phenomena, and such deWnitions have consequences for
how we understand the achievements of language. How we deWne language,
for instance, will aVect how we understand interlanguages like pidgins
and creoles. And whether we accept language as a Wxed, determinable code
will shape attitudes toward social and regional variations. It is this nexus
of academic debate and social consequence that this chapter explores. In
particular, I want to consider how, through deWnitions of natural language
in general as well as English in particular, the discipline of linguistics has
helped cultivate anxiety about the natural attributes of language, including
change and variation. As with literary narratives and legislative policy, I will

6 Willinsky, Empire of Words: The Reign of the OED (Princeton, NJ: Princeton University Press,
1994), 44.
7 For consideration of the ideological dimension of some of these practical decisions, see Henri
Béjoint, Tradition and Innovation in Modern English Dictionaries (Oxford: Clarendon Press, 1994), 107–39.
8 Quoted in Morton, The Story of ‘Webster’s Third’, 182.
 Say the Right Thing 191

suggest, linguistics has itself served as a way to channel extra-linguistic

concerns through language.

Language standards
The variation and change that natural languages experience always leave them
susceptible to transformations that hinder communication. Dante acknow-
ledged as much in his De Vulgari Eloquentia, perhaps the medieval period’s
most detailed and sophisticated account of language and grammar, when
he traced the development of vernacular languages across Europe and noted
in general that ‘no human language can be lasting and continuous, but must
needs vary like other properties of ours, as for instance our manners and our
dress, according to distance of time and place’. For Dante, grammar was
invented precisely as a way of resisting the changeability that lies at the
heart of natural language and that leads to failures in communication:
Hence were set in motion the inventors of the art of grammar, which is nothing else
but a kind of unchangeable identity of speech in diVerent times and places. This,
having been settled by the common consent of many peoples, seems exposed to the
arbitrary will of none in particular, and consequently cannot be variable. They
therefore invented grammar in order that we might not, on account of the variation
of speech Xuctuating at the will of individuals, either fail altogether in attaining, or
at least attain but a partial knowledge of the opinions and exploits of the ancients,
or of those whom diVerence of places causes to diVer from us.9

Grammar thereby enables humans to transcend the limitations of language

and connect themselves not only to their peers but also to the ancients from
whom, in evolving humanist thought, they had declined in so many ways. It is
this stabilizing quality that characterizes and distinguishes written, presti-
gious languages from non-regularized spoken vernaculars and that therefore
holds out to the latter a means to achieve the status of the former.
In 1490 William Caxton, who must have been around seventy at the time,
expressed a similar sentiment, perhaps less grandly, by commenting on how
much English had changed in his own lifetime: ‘our langage now vsed varyeth
ferre from that. which was vsed and spoken whan I was borne’.10 Inasmuch
as Caxton refers to what was ‘spoken’ when he was young, he evokes, from
a diachronic perspective, the very qualities of variation—speciWcally, its

9 Dante, De vulgari eloquentia, in Alex Preminger et al. (ed. and trans.), Classical and Medieval
Literary Criticism: Translations and Interpretations (New York: Frederick Ungar, 1974), 420–1.
10 Caxton, The Prologues and Epilogues of William Caxton, ed. W. J. B. Crotch, EETS os 176
(1928; rpt. New York: Burt Franklin, 1971), 108.
192 Language Anxiety

tendency to obscure communication—that Dante sees grammar as resisting.

And, synchronically, this same sense of unintelligibility, of not using the same
language, can confront contemporary tourists speaking structurally distinct
varieties, such as New Zealand and African American Vernacular English,
or Geordie and South African English. It is, however, the written channel,
the focus of Dante’s concerns, that presents perhaps the clearest examples of
how time and space work against constancy in language and communication.
To compare passages of English from diVerent moments in the language’s
history, as I did in Chapter 2, is to confront the fact that, diachronically,
English has indeed changed in ways that render varieties from diVerent
epochs mutually unintelligible.
To counteract this potential for mutual intelligibility and thereby increase
the communicative and social inXuence of a language, users of every major
Western European language (including English) since the early modern period
have followed Dante’s lead, with varying degrees of success, and worked to
cultivate spoken and written varieties with two primary qualities. First, these
cultivated varieties display minimum variation in form. In spoken language,
this means consistency in pronunciation or stress patterns, while in written
language it means consistency in orthography, punctuation, and discursive
practices. Second, these varieties manifest maximum variation in function,
which means that whether spoken or written they can be used in a number of
domains for a number of purposes. Together, these qualities foster a standard
as a variety with wide intelligibility among speakers separated by geography
or time, and they also encourage and even require its use in inXuential cir-
cumstances. In eVect, they allow languages to transcend variation and change
as a way of improving (if not guaranteeing) communication.11
Standardization is thus at once a historical and a contemporary process.
Historically, it involves the selection and codiWcation of a particular variety in
grammar books, rhetoric handbooks, and, of course, dictionaries, all of which
distinguish right forms from wrong ones. The contemporary element of
standardization derives from the fact that to be a standard a variety cannot
simply be deWned as such. It must be maintained and cultivated in sequential
synchronic moments across time, through its acceptance as a standard by
the language’s users and through an elaboration of its functions, largely by
means of its explicit instruction in schools. Its signiWcance thus continuously
evolves, building on its historical inception, subsequent interpretations of
this inception, and subsequent interpretations of its current use.
11 This key formulation of a standard as a variety displaying minimal variation in form and
maximum variation in function originates with Einar Haugen, ‘Language, Dialect, Nation’, in The
Ecology of Language, ed. Anwar S. Dil (Stanford, CA: Stanford University Press, 1972), 237–54.
 Say the Right Thing 193

For Western European vernaculars (again as Dante suggested), Latin, with

its long history of codiWcation and institutional prominence in schools and
law courts, provided much of the model and inspiration for the immobiliza-
tion of language that underlies standardization. To be sure, in his 1528
De Recta Latini Graecique Sermonis Pronuntiatione Erasmus lamented the
variable and sometimes mutually unintelligible pronunciations of this lan-
guage, then utilized by none as a Wrst language, learned primarily through
rigorous classroom instruction, and written and spoken in only select do-
mains. And in England, John Hart’s 1569 Orthographie likewise documents
that English pronunciations of Latin diVer from those used by speakers of
other vernaculars.12 But throughout much of its written history, and even
taking into consideration the signiWcant syntactic and lexical diVerences
between the classical and medieval versions, Latin for the most part did
remain considerably more Wxed than any vernacular, particular in scholarly
conceptions. The most ancient, most inXuential works, of Cicero or Virgil,
had long since been completed by the early modern period, and textual
criticism, the critical tool that made Humanism possible, promised to deter-
mine and deliver their correct texts once and for all.
Medieval Christian exegesis had infused this demonstrable grammatical
regularity with moral and even eschatological qualities. It was St Augustine,
for example, who drew a distinction between vox as external sound and
verbum as internal concept in a way that paralleled a distinction between
the fallen world and the eternal logos of Christ. Within this framework, the
emphasis of linguistic study needed to be the cultivation of semantics and
regular grammatical construction, since a simple description of ordinary
change and variation could only conWrm the fallen nature of the world—
both materially and linguistically.13 The contemporary Xuctuations that dis-
turbed Erasmus did so not simply because of grammatical solecism, then, but
because of their ethical implications. And it is for reasons like these that
schoolmasters throughout early modern Europe strove toward consistency
and regularity in the instruction and uses of Latin, employing a pedagogy that
depended on rote memorization of morphology and vocabulary and also, as
importantly, on the translation into and out of the vernacular sentences that
modeled acceptable grammar and sentiments (vulgaria). Joined to print
technology, which provided wide access to (in many cases) virtually identical

12 Hart, An Orthographie (1569; rpt. Menston: Scolar Press, 1969), Biir .

13 See further Vivien Law, The History of Linguistics in Europe from Plato to 1600 (Cambridge:
Cambridge University Press, 2003), 94–111; and John Fyler, Language and the Declining world in
Chaucer, Dante, and Jean de Meun (Cambridge: Cambridge University Press, 2007), 1–59.
194 Language Anxiety

texts, Latin demonstrated that prestigious, eloquent, moral, and learned

language was that which eluded change and variation.
Early codiWers of English focused on the absence of just this constancy in
describing their own language. So long as the language deviated from Latin
and Greek in its regularity, its potential to displace their cultural authority
could remain only that. Hart, indeed, saw his composition of the Orthogra-
phie as a selXess gesture for the beneWt of the language and his countrymen, all
diminished by this deviation: ‘no man ought to trauell in this life onely or
chieXy for himselfe and his next bloud, to the hindrance of others, but for the
common welth of his country, though with daunger of life, or the price
thereof in deede. Whoso may proWte his country in any condicion, and
especiallye wyth small cost and no daunger, he were vnnaturall to be a
niggarde thereof.’ And English, it seems, sorely needs Hart’s ministrations:
But in the modern & present manner of writing (as well of certaine other languages as
of our English) there is such confusion and disorder, as it may be accounted rather a
kind of ciphring, or such a darcke kind of writing, as the best and readiest wit that
euer hath bene, could, or that is or shalbe, can or may, by the only gift of reason,
attaine to the ready and perWte reading thereof, without a long and tedious labour, for
that it is vnWt and wrong shapen for the proportion of the voice.14

Accordingly, he addressed his work to reforming English spelling, including

the identiWcation of permissible sound-values for graphs, the invention of
necessary new graphs, and the determination of spellings that correctly reXect
speech.
About a decade later, Richard Mulcaster’s Elementarie was even more
speciWc than the Orthographie about the communicative discord produced
by variation in English. To Mulcaster, this instability extended particularly to
the relations between graphs and sounds: ‘euerie letter almost being deputed
to manie, and seuerall, naie to manie and well nigh contrarie sounds and vses,
euerie word almost either wanting letters, for his necessarie sound, or hauing
some more than necessitie requireth, [I] began to despare in the midst of such
a confusio [sic], euer to Wnd out anie sure direction, whereon to ground Art,
and to set it certain’.15 And for William Bullokar, a few more years later in his
Amendment of Orthographie, the need to identify variation as a problem was a
strong enough given in metalinguistic discourse that he could articulate it
even as he also aYrmed the greatness of the current form of English: ‘So that
for lacke of true ortography our writing in Inglish hath altered in euery age,

14 Hart, An Orthographie, Aiiiir---v , Aiiir .

15 Mulcaster, The First Part of the Elementarie, (1582; rpt. Menston: Scolar Press, 1970), 77.
 Say the Right Thing 195

yea since printing began . . . and if now be a time of the most perfect vse of
the same, which must be confessed for the great learning dispensed in this
land at this day . . . thinke it the great gift of God, if a perfectnesse be now
surely planted, not to be rooted out as long as letters endure.’16 English is at
once perfect and degenerate—one of the many contradictions animating
metalinguistic discussion then and still today.
While Hart, Mulcaster, and Bullokar saw orthographic instability as the
source of English’s most pressing problems, Richard Verstegan, in his 1605 A
Restitution of Decayed Intelligence, foregrounded controversy over English’s
unstable vocabulary and its propensity to borrow words from other lan-
guages. Some critics ‘think our toung thereby much bettred’ through such
borrowing, according to Verstegan, but others say ‘that it is of it self no
language at all, but the scum of many languages’ and still others ‘that it is
most barren, and that we are dayly faine to borrow woords for it . . . out
of other languages to patche it vp withall’.17 With a thirty-three-page alpha-
betized list of ‘our moste ancient English woords’ and their modern synonyms
and orthographic equivalents, Verstegan’s Restitution sided with those who
thought lexical stability should come internally, from English’s native word
stock. But other critics, such as Richard Carew in his ‘Excellency of the
English Tongue’, advocated that an infusion of learned borrowings—deri-
sively labeled ‘inkhorn terms’ by their opponents—was the best way for
English to achieve lexical balance. And still other critics, like Thomas Wilson,
saw inkhorn terms as a problem precisely because they represented variance
from whatever constancy English already had. In his 1553 Arte of Rhetorique,
he couches this argument in the kind of emotive associations that critics
of Webster’s Third would use four centuries later:
Among al other lessons, this should Wrst be learned, that we neuer aVect any straunge
ynkehorne termes, but so speake as is commonly received: neither sekyng to be ouer
Wne, nor yet liuyng ouer carelesse, vsyng our speache as most men do, & ordryng our
wittes, as the fewest haue doen. Some seke so farre for outlandishe Englishe, that thei
forget altogether their mothers language. And I dare swere this, if some of their
mothers were aliue, thei were not able to tell, what thei say, & yet these Wne Englishe
clerkes, wil saie thei speake in their mother tounge, if a man should charge them
for counterfeityng the kynges English.18

16 Bullokar, The Amendment of Orthographie for English Speech (1580; rpt. Amsterdam: Theatrum
Orbis Terrarum, 1968), 2.
17 Verstegan, A Restitution of Decayed Intelligence, (1605; rpt. Ilkley: Scolar Press, 1976), 204.
18 Wilson, The Arte of Rhetorique (1553; rpt. Gainesville, FL: Scholars’ Facsimiles and Reprints,
1962), 86r .
196 Language Anxiety

It certainly wasn’t the case, then, that all early grammarians agreed on what
should be stabilized or how this should take place. What they did share was a
sense that if English was to attain the expressibility and status of Latin and
Greek, its variation needed to be controlled and channeled. At the same
time, by designating vocabulary as acceptable or unacceptable and making
these designations accessible only to readers who had access to handbooks like
the Orthographie, the Elementarie, and the Arte of Rhetorique, these early
metalinguistic discussions helped to fashion language into a marker of social
rank. Only the kind of noble who once could use Latin as a concomitant of
education and status—someone like Kynaston—would be able to learn which
English words to use for the same purpose.19 In another few generations
Daniel Defoe, drawing on the inspiration of the Académie Française and its
dictionary, infused this grammatical and social discussion with political
overtones by making grammar a point of national pride and suggesting
that ‘the English Tongue is a Subject not at all less worthy the Labour of
such a Society than the French, and capable of much greater Perfection. The
Learned among the French will own, That the Comprehensiveness of Expres-
sion is a Glory in which the English Tongue not only Equals but excels
it Neighbors.’20 English, to Defoe, needed to be purged of irregularities and
puriWed in style, and these perceived needs led him, and later Jonathan Swift,
to call for the creation of England’s own language academy, a call that is
still sometimes repeated today.21
The burst of codiWcation in the seventeenth and primarily eighteenth
centuries indicates just how strong a hold this valuation of linguistic stability
took on the minds of Anglophones. The Wrst extant English grammar of
English was Bullokar’s 1586 Pamphlet for Grammar, and by 1600 only one
other such grammar had appeared. By 1700 there were just an additional
thirty. During the course of the eighteenth century, however, 236 more
English grammars were published.22 Dictionaries proliferated in this period
in a similar fashion. The earliest English dictionary is Robert Cawdrey’s 1604
A Table Alphabeticall, but between it and Johnson’s 1755 Dictionary there

19 See further Paula Blank, ‘The Babel of Renaissance English’, in Lynda Mugglestone (ed.), The
Oxford History of English (Oxford: Oxford University Press, 2006), 212–39.
20 Defoe, Essays upon Several Projects (London: Thomas Ballard, 1702), 229–30.
21 John Honey, Language Is Power: The Story of Standard English and its Enemies (London: Faber
and Faber, 1997), 164.
22 See Ian Michael, English Grammatical Categories and the Tradition to 1800 (Cambridge: Cam-
bridge University Press, 1970); 549–87 enumerate the English grammars, and 588–94 order them
chronologically. Another Wve English grammars appeared only in manuscript during this period.
More generally, see G. A. Padley, Grammatical Theory in Western Europe, 1500–1700: Trends in
Vernacular Grammar, 2 vols. (Cambridge: Cambridge University Press, 1985 and 1988).
 Say the Right Thing 197

appeared at least another twenty dictionaries, some of them, such as Nathan

Bailey’s 1721 An Universal Etymological English Dictionary, appearing in over
twenty editions.23 Grammars and dictionaries alike drew heavily upon one
another, so that the amount of original work this codiWcation represents
is considerably less than these publication Wgures suggest. Nonetheless, codiW-
cation thereby emerges as one of the pre-eminent linguistic concerns of the
early modern period, cultivating stability in orthography, punctuation, rhet-
oric, and vocabulary and deWning principles of language change built on
analogy, etymology, history, logic, and usage. The standardization of spoken
language has lagged behind that of written, beginning in earnest only with
Walker’s 1791 Critical Pronouncing Dictionary, though this delay was quickly
remedied. In England the following century witnessed, for example, both the
expansion of public schools as boarding institutions that cultivated (among
other things) constancy in language, and also the distribution, to an increas-
ingly literate society, of handbooks designed to help speakers eliminate
stigmatized forms like ‘dropped h’. Daniel Jones’s 1917 An English Pronouncing
Dictionary proved particularly inXuential in inculcating the ideology of a
spoken standard language as well as the speciWc form of what was regarded
as correct English pronunciation. Its value judgements of good and bad
pronunciation have framed discussion of varieties of English literally far
aWeld—in New Zealand, for example—from what Jones himself spoke and
described.24
It is the breadth of such cultivation—its geographic, chronological, and
social range—that has had the greatest consequence in the creation of a
standard and its implications for linguistic change and variation. By itself,
minimalization of forms—regularity in orthography and so forth—carries
little signiWcance. It may be the one structural characteristic of standards that
diVerentiates them from other varieties, but it alone cannot transform a
variety into a standard, and the forms this minimalization deWnes as stand-
ard—a pronunciation of have with an initial [h] ([hæv]) rather than one

23 DeWitt T. Starnes and Gertrude E. Noyes, The English Dictionary from Cawdrey to Johnson,
1604–1755 (Chapel Hill, NC: University of North Carolina Press, 1946).
24 For a survey of standardizing works in the early period, see Manfred Görlach, Explorations in
English Historical Linguistics (Heidelberg: C. Winter, 2002), 137–212. For general discussion of the kinds
of issues such works examined, see Sterling Andrus Leonard, The Doctrine of Correctness in English
Usage, 1700–1800 (1929; rpt. New York: Russell and Russell, 1962); Richard W. Bailey, Images of English:
A Cultural History of the Language (Ann Arbor, MI: University of Michigan Press, 1991), 179–213; and
Linda C. Mitchell, Grammar Wars: Language as Cultural BattleWeld in 17th and 18th Century England
(Aldershot: Ashgate, 2001). On the details of spoken standardization, see Lynda Mugglestone, ‘Talking
Proper’: The Rise of Accent as Social Symbol, 2nd edn. (Oxford: Oxford University Press, 2003). On the
use of Jones’s standard in discussion of New Zealand English, see Elizabeth Gordon et al., New Zealand
English: Its Origins and Evolution (Cambridge: Cambridge University Press, 2004), 12.
198 Language Anxiety

without it ([^v]) or a spelling like which rather than hwich—have no intrinsic

quality that makes them standard. Already in the Wfteenth century, in fact,
printed English in general displayed considerably less variation than did
written English of the previous century, and one variety in particular, Chan-
cery English, had consistency even in the shapes of graphs, to go along with
its orthographic, morphological, and lexical consistency. Developed within
the government oYce of the Chancery and inXuential on other government
writing and on English writing and printing in general, Chancery experi-
enced, at least implicitly, some of the codiWcation associated with standards.25
And a century later, as in Hart’s Orthographie, written English can already
sometimes appear virtually identical to what lexicographers like Johnson or
grammarians like Bishop Lowth would come much later to advocate: ‘As of
this part of diVerence, I shall write more at large after I haue brieXy shewed
you the two other vices of vsurpation of powers and misplacing of letters.’26
Here, inasmuch as ‘shewed’ remains a recognized if obsolete variant of
‘showed’, the only formal diVerences with modern Standard English involve
the distribution of u and v. Such similarities have increasingly led some critics
to describe English as having been standardized already in the late-medieval
period. John Fisher has argued, for instance, that Chancery English is the
speciWc variety from which modern Standard English arose, while others
have spoken more generally and casually of ‘the standardization of much
written English in the early Wfteenth century’.27 And still other critics have
even identiWed a standard Old English in the tenth-century West Saxon koine
in which most Old English writing survives.
But there is a fundamental and profound diVerence between a minimaliza-
tion of forms of the sort encouraged by centralized copying houses or the
development of print and standardization as an ideological and institutional
gesture. The one might be called standardized language, the other standard
language. If printed English of the sixteenth and even Wfteenth centuries
shared many of the formal characteristics of Modern English, it was framed
by none of the social structures that render a variety a standard: selection,

25 On the development and characteristics of Chancery English, see John H. Fisher, The Emergence
of Standard English (Lexington, KY: University of Kentucky Press, 1996), 65–83; and also Fisher,
Malcolm Richardson, and Jane L. Fisher (eds.), An Anthology of Chancery English (Knoxville,
TN: University of Tennessee Press, 1984).
26 Hart, An Orthographie, Eiiiiv .
27 Fisher makes this argument in The Emergence of Standard English, as does Thomas Cable in
‘The Rise of Written Standard and English’, in Aldo Scaglione (ed.), The Emergence of National
Languages (Ravenna: Longo Editore, 1984), 75–94. The quotation comes from Jocelyn Wogan-Browne
et al. (eds.), The Idea of the Vernacular: An Anthology of Middle English Literary Theory, 1280–1520
(University Park, PA: Pennsylvania State University Press, 1999), 340.
 Say the Right Thing 199

codiWcation, and broad-based elaboration of function. All these are post-

medieval, even post-early-modern, developments. Indeed, as I have noted,
grammar books and dictionaries developed over the course of the seventeenth
century and mostly in the eighteenth, and in at least the early part of this
period Latin continued to be the language of instruction in grammar schools
and at Oxford and Cambridge. It was in this period, too, that English came
to be invested with political signiWcance in the formation of England as a
nation. These are the forces that produce what has been called the ideology of
a standard—a way of thinking about language that invests cultural and
linguistic prestige in one variety and that accordingly conceptualizes a hier-
archy of other varieties and the sociolinguistic consequences of their use.28
It is this ideological component that ultimately creates a standard language
from standardized language by transforming what might be simply a mini-
malization of variation into the practical and theoretical suppression of
variation and change. Standard languages above all depend on the investing
of meaning, on the ascription of social or cultural signiWcance to linguistic
forms that are otherwise simply tokens in a grammatical system. Indeed, a
vowel, it might be said, is only a piece of acoustic phenomenon until speakers
assign social value to it as a good or bad pronunciation. More concisely,
attitude as much as form makes a standard what it is. Of even greater
consequence as an ideological gesture, the very concept of standardization
can make standards seem natural, inevitable varieties and in the process shape
our perception of all linguistic activity. Indeed, to see a standard vernacular in
pre-modern England, where it could not possibly occur, is to testify for the
standard ideology’s naturalization of linguistic regularity.
For all these reasons I would say that if 1066 and the Battle of Hastings serve
as the convenient and conventional transition date from Old to Middle
English, and 1476 and Caxton’s establishment of the Wrst English printing
press as that from Middle to Early Modern English, then 1755 and the
publication of Dr Johnson’s Dictionary well symbolize the undisputed exist-
ence of Standard English. And by ‘Standard English’ I mean not a historical
stage in the language’s development but a speciWc, cultivated variety of the
language: primarily, a written variety constant in orthography, punctuation,

28 For a critique of attempts to identify a standard English in the Middle Ages, see Jeremy J. Smith,
‘Standard Language in Early Middle English?’, in Irma Taavitsainen et al. (eds.), Placing Middle English
in Context (Berlin: Mouton, 2000), 125–39. On the ideology of a standard more generally, see James
Milroy and Lesley Milroy, Authority in Standard Language: Investigating Standard English, 3rd edn.
(London: Routledge, 1999). They deWne standard ideology as ‘a public consciousness of the standard.
People believe that there is a ‘‘right’’ way of using English, although they do not necessarily use the
‘‘correct’’ form in their own speech’ (25). They attribute the decisive establishment of this ideology to
the eighteenth century (29).
200 Language Anxiety

morphology, syntax, and usage, which, while it may have some regional
variation (e.g., Standard British English as opposed to Standard United States
English) is everywhere also the language and object of classroom instruction,
can be referenced through several forms of codiWcation, and is recognized
as the standard language in powerful domains of business, education, and
government. And the timing of just when Standard English can be said
to come into existence is signiWcant, because as a response to variation
and change, a standard language is also a response to the cultural context of
its creation.
When grammatical discussions progressed in the seventeenth and eight-
eenth centuries, Standard English emerged not simply in codiWcation but
through ways of conceptualizing language. As a medium of ideas that also
stood metaphorically for cultural and national stability, a relatively invariable
form of English—and as importantly the very notion of such a form—came to
stand for the language as well. In this way change and variation could be both
linguistically and socially deviant and as such open to further qualiWcations of
their deviance. By that I mean that once change and variation have been
generally marked as channels for extra-linguistic anxieties—once it has been
demonstrated that they can assume extra-linguistic signiWcation—additional
non-linguistic concerns can easily be mediated through them. Indeed, in the
past few centuries, as formal features of both spoken and written language
have become increasingly codiWed, the attribution of social meanings
to variation and change has been an increasingly prominent part of the
standardization process.
In the case of Standard English, as Defoe’s interest in the Académie
Française demonstrates, early grammatical discussions drew much of this
meaning from presumptive associations between language and national iden-
tity. The opening sentence of Alexander Gill’s 1621 Logonomia Anglica is thus
‘The English nation and language have one origin: they go back to the Saxon
and Angle peoples of Germany’.29 Perhaps inevitably, associations like these
evoke comparisons between English and prestigious antique or modern
languages. In his 1605 Remains concerning Britain William Camden thus
notes, ‘I think that our English tongue is, I will not say, as sacred as the
Hebrew or as learned as the Greek; but as Xuent as the Latin, as courteous as
the Spanish, as Court-like as the French, and as amourous as the Italian, as
some Italianated amorous have confessed’.30 More simply and eloquently, in
his Elementarie Mulcaster, having described the English of his day as the best

29 Gill, Logonomia Anglica (1621; rpt. Menston: Scolar Press, 1968), no sig.
30 Camden, Remains Concerning Britain (1605; rpt. Yorkshire: EP Publishing Limited, 1974), 33.
 Say the Right Thing 201

there ever was or would be, waxed rhapsodic on the intrinsic value of his
language and on the intrinsic connections between it and its speakers: ‘I loue
Rome, but London better, I fauor Italie, but England more, I honor the Latin,
but I worship the English.’31
By aligning the character of the English language with the character of
England itself, early grammarians like Camden and Mulcaster fostered a sense
of language that marked change and variation as politically as well as socially
signiWcant. Through a kind of syllogistic logic that recalls the equation of
nationhood with monolingualism, if the achievements of English stood as
testaments to the emerging greatness of the country, then linguistic deviance
could likewise reXect and even contribute to political dissolution. And by
extension, variation already present in the form of regional orthography or
morphology needed to be suppressed in the advance of Standard English. The
written record of this period documents just this homogenization of writing,
as in Early Modern Scots, where documents show the gradual insertion
and dominance of southern English forms for the weak preterite, relative
pronoun, and so forth.32 Stability in the realms of language and social
interaction were therefore connected even before Standard English can be
said to exist, although by the same token this connection is one of the
elements that made Standard English possible. For critics as enthusiastic as
Camden, who praised the achievement of contemporary English, whatever
faults there were in the language were there precisely because the present age
was too given to fashion and novelty. ‘Hitherto will our sparkful youth laugh
at their great-grandfathers English,’ he intones, ‘who had more care to do well
than to speak minion-like, and left more glory to us by their exploiting of
great Acts, than we shall do by our forging of new words and uncouth
phrases.’33 In a similar vein, Verstegan points out that ‘people in former
ages were nothing so curious or delighted with varying their speech, as of
late ages they are grown to bee, but kept their old language as they did their
old fassion of apparel; in both which the world hath of later ages more then in
former bin delighted; and in this age of ours much more than euer’.34 When
Mulcaster grants, therefore, that the English of his day must continue to
change, he phrases his praise of his own English in a way that implicates
the uncertain future of England in the certain variation and decline of its

31 Mulcaster, The First Part of the Elementarie, 254.

32 See Amy J. Devitt, Standardizing Written English: DiVusion in the Case of Scotland 1520–1659
(Cambridge: Cambridge University Press, 1989).
33 Camden, Remains Concerning Britain, 29.
34 Verstegan, A Restitution of Decayed Intelligence, 197.
202 Language Anxiety

language: ‘whatsoeuer shall becom of the English state, the English tung
cannot proue fairer’.
As codiWcation developed, so, too, did this connection between order in
language and order in society. Through the eVorts of writers as diverse
as Thomas Hobbes, John Locke, Edmund Burke, and Dr Johnson, it, too,
might in fact be regarded as having become naturalized—that is, become an
unexamined assumption about language in many Anglophone communi-
ties.35 In a private letter of 1638, even before these other writers had publicly
deWned the issue, Milton observed: ‘For when speech is partly awkward and
pedantic, partly inaccurate and badly pronounced, what does it say but that
the souls of the people are slothful and gaping and already prepared for any
servility? On the other hand, not once have we heard of an empire or state
not Xourishing at least moderately as long as it continued to have pride in
its Language, and to cultivate it.’36 It is just this kind of a presumed and
unquestioned connection between social and linguistic order that underlies
the policy issues I considered in the previous chapter.
In channeling political concerns through language, early modern metalin-
guistic discussion helped fashion change and variation as detrimental, if still
natural, features of human language. Early critics reinforced this connection
when, from a narrower perspective on communication, they described the
purpose and achievement of language to be the transference of pure, unadul-
terated ideas from speaker to listener or from writer to reader. For Carew, ‘the
Wrst and principal point sought in every language is that we may express
the meaning of our minds aptly to each other’.37 Subsequent linguistic
discussion, even when not overtly political, has often reinforced this pre-
Babel notion that human language should serve as the means for relaying
ideas in their totality and that variation and change are therefore counter-
productive processes. In his Essay on Human Understanding, for example,
Locke characterized language as arbitrary and conventional and words as
signs of ideas. Connections between physical words and invisible ideas cannot
be natural, Locke reasoned, or there would be only one language. Because
language proceeds from sense perception, moreover, ‘those of one Country,
by their customs and manner of Life, have found occasion to make several
complex Ideas, and give names to them, which others never collected into

35 For an overview of these connections, see David Simpson, The Politics of American English,
1776–1850 (New York: Oxford University Press, 1986), 32–51.
36 Milton, Complete Prose Works, 8 vols., gen. ed. Don M. Wolfe (New Haven, CT: Yale University
Press, 1953–82), i, 330. The letter, written in Latin and translated in this volume, is to Benedetto
Buonmattei.
37 Carew, ‘The Excellency of the English Tongue’, in Camden, Remains Concerning Britain, 42.
 Say the Right Thing 203

speciWck Ideas. This could not have happened, if these Species were the
steady Workmanship of Nature; and not Collections made and abstracted
by the Mind, in order to naming, and for the convenience of Communica-
tion.’ Very much in line with St Augustine, however, Locke also contends that
arbitrary language can be shaped and managed so as to avoid some kind of
moral relativism and instead recover, so far as is possible, the stability of
God’s truths. Whether through imprecise expression or shifting pronunci-
ation and usage, failures in communication occur when the words speakers
use neither ‘stand . . . for the reality of Things’ nor invoke the intended ‘Ideas in
the minds also of other Men, with whom they communicate’. Indeed, Locke
understands the purpose of any language to be to enable the exchange of
ideas, involving external signs that make one’s invisible ideas known to others:
‘To make Words serviceable to the end of Communication, it is necessary . . .
that they excite, in the Hearer, exactly the same Idea, they stand for in the
Mind of the Speaker. Without this, Men Wll one another’s Heads with noise
and sounds; but convey not thereby their Thoughts, and lay not before
one another their Ideas, which is the end of Discourse and Language.’38
Rather than naturalize variation and change, an argument like this (later
advanced by David Hume and others) points to the importance of maintain-
ing linguistic regularity, for the consequences of imprecision or the misuse of
language are misused and imprecise ideas and hence communication. It is for
reasons like this that Descartes had already provided an argument against
accepting animal communication as language: inasmuch as human language
consists of the exchange of ideas, communication among animals cannot be
linguistic, since, lacking human minds, animals have no ideas to communi-
cate.39 And these same reasons underwrote the period’s notion of universal
grammar, which understood thought as subject to rational principles and
language as the image of thought, so that language perforce had to—or
perhaps should—reXect the laws of reason.40 Even the prevailing alternative
argument about the character of language agreed with philosophers like
Locke on at least this one point. In the eighteenth century, through such
eVorts as Horne Tooke’s Diversions of Purley, this argument took the form of
advocacy for the ‘genius’ of each language, that is for the intrinsic character

38 Locke, An Essay Concerning Human Understanding, ed. Peter H. Nidditch (Oxford: Clarendon
Press, 1975), 432–3, 407, 406, 478.
39 See Roy Harris, The Language Machine (Ithaca, NY: Cornell University Press, 1987), 30. A dated
but still valuable overview of eighteenth-century positions on language can be found in Leonard, The
Doctrine of Correctness in English Usage, 19–44.
40 See further Hans AarsleV, The Study of Language in England, 1780–1860, 2nd edn. (Minneapolis,
MN: University of Minnesota Press, 1983), 13–43.
204 Language Anxiety

of a language’s form and semantics. For the present discussion, the important
point is that any variation from this character would necessarily be, again,
deviation. Locke, Tooke, and even Whorf thus share at least this: the same
essentialist connection between language and culture, as well as the sense that
variation is semantically obfuscatory and socially debilitating.
The ways in which a standard ideology has come to channel social concerns
through change and variation have proliferated in the modern world, aVect-
ing ethnic and social groups as well as individuals and, in turn, national
varieties, sociolects like African American Vernacular English, the argot of
young speakers, and so forth. In fact, the mediation of social concerns
through language, including change and variation, has become so proliWc
that Deborah Cameron has invented the term verbal hygiene to describe ‘the
urge to meddle in matters of language’.41 This meddling can take many forms,
appearing in certain kinds of classroom instruction and linguistic analysis,
as well as in commentary by popular essayists like William SaWre and
John Honey. The focus of discussion may be nominally linguistic—correct
pronunciations, usages, and punctuation—and its implication is always that
perfect communication, as before Babel, is possible and attainable, if only
prescriptive grammatical rules were obeyed. But the discussion more gener-
ally serves as ‘the arena where certain social conXicts Wnd symbolic expres-
sion’. Above all, as with the uproar over Webster’s Third, these conXicts engage
issues like moral relativism, authority, and permissiveness in society. ‘In the
case of language,’ Cameron suggests, ‘it might be argued that investment in
traditional authority manifests not just a general preference for continuity
over change, but also an attachment to values and practices that were im-
pressed on people in the formative stages of their personal linguistic histories.’
In its most direct and brutal form, such investment disallows any speech
judged non-standard, and, more importantly, those who speak it as well. In
most Anglophone areas it may be impossible (or at least diYcult) to refuse to
hire individuals on the basis of their race or sex, but it is certainly legal
in many cases to refuse to do so because of the putatively incomprehensible
variety they speak: ‘linguistic bigotry is among the last publicly expressible
prejudices left to members of the western intelligentsia’. More subtly, in
Cameron’s analysis, in both the United States and the United Kingdom
recent eVorts to revive traditional grammar instruction and aYrm the im-
portance of Standard English have been the means to advance a politically
orthodox agenda:

41 Cameron, Verbal Hygiene (London: Routledge, 1995), vii.

 Say the Right Thing 205

conservatives use ‘grammar’ as the metaphorical correlate for a cluster of related

political and moral terms: order, tradition, authority, hierarchy and rules. In the
ideological world that conservatives inhabit, these terms are not only positive, they
deWne the conditions for any civil society, while their opposites—disorder, change,
fragmentation, anarchy and lawlessness—signify the breakdown of social relations.
A panic about grammar is therefore interpretable as the metaphorical expression of
persistent conservative fears the we are losing the values that underpin civilization
and sliding into chaos.42

The very notion of traditional grammar instruction informing these connec-

tions presents its own diYculties, for the halcyon days it recalls never were.
For one thing, since the modern profession of English instruction came
into existence in the late nineteenth century English teachers have been
describing each generation of student-writers as worse than the previous
one. And for another, even those individuals whose long-ago grammar
instruction may well have been superior to anything available today are also
often remembering and glorifying an era that restricted that kind of education
by sex, class, and race.
From the perspective of formal linguistics, Germanic philology, which
emerged in the nineteenth century as the dominant approach to change and
variation, developed linguistic genealogy with much the same nostalgia and
socializing eVects. As conceptual tools, tree diagrams, showing the progres-
sion of (say) Indo-European to Centum to Germanic, implied the discrete
character of languages and described change as an exceptionless and regular
process. The extension of this analysis into English necessarily promoted
similar notions of linguistic stasis and change. In many ways, indeed, the
history of English as an academic discipline also typically has been the history
of standardization, a history that moves from the dialects of the Old and
Middle English periods, through the codiWcation of the early modern one,
to the emergence of a global modern Standard English in the eighteenth
century, to, at least in some accounts, an implicit culmination in modern
American English.43
Given the historical contexts of standardization and modern linguistics,
this implication of Standard English in what is widely understood to be the
language’s history was perhaps inevitable. From a literal if limited perspective,
it can be said that early critics recognized and deWned—‘invented’ would be

42 Cameron, Verbal Hygiene, 13, 14, 12, 95. On the complaint tradition, also see Bailey, Images of
English, 237–66, and Milroy and Milroy, Authority in Standard Language, 24–46.
43 See, for example, Albert C. Baugh and Thomas Cable, A History of the English Language, 5th edn.
(Upper Saddle River, NJ: Prentice Hall, 2002); and Seth Lerer, Inventing English: A Portable History of
the Language (New York: Columbia University Press, 2007).
206 Language Anxiety

the early modern word—diachronic change and synchronic structure at

the same time. Even as seventeenth-century grammarians like Verstegan or
Cawdrey were codifying contemporary English, that is, bibliophiles like Francis
Junius were inaugurating Anglo-Saxon studies. In this way, the history of
English as a discipline not only emerges from a standard ideology but also,
emphasizing continuity above all, legitimizes the standard variety of English as
the pre-eminent one and non-standard ones as merely deviations. While critics
like H. C. Wyld could thus imagine a history for Standard English and might
even accept the legitimacy of rural dialects, they saw urban varieties as simply
vulgar. ‘It is as though uneducated speakers are not allowed to be involved
in language history,’ comments James Milroy.44 And this use of historical
continuity and stability as qualities that legitimate linguistic forms and varieties
continues today with standard ones but also, in a gesture that again points to
the naturalization of a standard ideology, with non-standard ones.
Creoles oVer a case in point. Seeing them as deWned not historically but
structurally, Salikoko S. Mufwene rejects prevalent conceptions of African
American Vernacular English as a stable variety that developed linearly from
the days of slavery. He argues, indeed, that attempts to trace African American
Vernacular English to a pervasive Plantation Creole and then to something
like the Gullah variety still used on the seacoast islands of South Carolina
reXect the kind of historical imperative and monogenesis that a standard
ideology fosters with regard to variation and change. ‘There is no historical
justiWcation’, he suggests, ‘for assuming that there was ever a time in the
seventeenth or eighteenth century when every African-American spoke the
basilect of a Gullah-like variety. Nor is there any particular justiWcation
for assuming that AAVE must have developed from a Gullah-like variety
and that its speakers must have aimed at speaking like whites.’45 African
American Vernacular English, for Mufwene, should be only an umbrella
term, not the label of a Wxed or even semi-regular variety that the ideology
of standardization encourages critics to Wnd.
In this way, history itself acquires ideological force, providing aYrmation of
a linguistic structure’s naturalness and continuity in the language. The mere
existence and development of particular usages over centuries—including

44 Milroy, ‘The Legitimate Language: Giving A History to English’, in Richard Watts and Peter
Trudgill (eds.), Alternative Histories of English (London: Routledge, 2002), 11. See also Shana Poplack,
Gerard van Herk, and Dawn Harvie, ‘ ‘‘Deformed in the Dialects’’: An Alternative History of Non-
Standard English’, in Watts and Trudgill (eds.), Alternative Histories of English, 87–110; and Milroy, ‘The
Consequences of Standardization in Descriptive Linguistics’, in Tony Bex and Richard J. Watts (eds.),
Standard English: The Widening Debate (London: Routledge, 1999), 16–39.
45 Mufwene, ‘African-American English’, in The Cambridge History of the English Language, vi, John
Algeo (ed.), English in North America (Cambridge: Cambridge University Press, 2001), 321.
 Say the Right Thing 207

Standard English as a variety—become conWrmations of their claim to be

transcendent, essential forms of English. ‘ ‘‘Antiquity’’ as a virtue of the English
language’, notes Richard Bailey, ‘thus emerges as a signiWcant part of the image
of the language. But the fact of antiquity is one thing; its uses are another,
and ‘‘educated’’ people have been expected to employ antiquity to their own
advantage in presenting themselves as ‘‘discriminating’’ users of English.
Highly reWned national and class distinctions have been built around etymo-
logical nuance.’46 In eVect, antiquity thereby works against variation and
change, so that even with over two-thirds of the current global Anglophone
population now living outside of the United Kingdom and the United States,
the antiquity of British English still serves as one of the strongest arguments
for its primacy among the increasingly diverse varieties of Modern English.
It was the putative failure of British English to live up to the responsibility
of antiquity that so delighted my radio interviewer in July of 2000.
With the exception of sociolinguistics, even much modern formal linguistic
theory has furthered the attitudes toward variation and change that are
foundational to the standard ideology and that were developed in eighteenth-
and nineteenth-century metalinguistic discussion. Ferdinand de Saussure’s
landmark Cours de linguistique générale, thus, propelled twentieth-century
criticism in this direction when it emphasized the synchronic state of a
language over its diachronic change. And this is the direction taken by
subsequent structuralist linguists, such as Charles Fries in his 1952 The
Structure of English, and in turn by the transformational approaches that
replaced structuralism. ‘Linguistic theory is concerned primarily with an
ideal speaker-listener’, Noam Chomsky famously declared, ‘in a completely
homogeneous speech-community, who knows its language perfectly and is
unaVected by such grammatically irrelevant conditions as memory limita-
tions, distractions, shifts of attention and interest, and errors (random
or characteristic) in applying his knowledge of the language in actual per-
formance.’47 Less theoretically, it is the same direction taken by Basic English,
Charles K. Ogden’s eVort to reduce and stabilize English as a secondary
language for technical discourse through the reduction of its vocabulary to
just 850 words. Beginning with the suppositions that a global language was a
political necessity (particularly between the two world wars), that an artiWcial
language like Esperanto was not viable, and that of the natural languages
English was the most logical choice, Ogden set about simplifying English,
not producing an interlanguage version. In the words of his colleague and

46 Bailey, Images of English, 274.

47 Chomsky, Aspects of the Theory of Syntax (Cambridge, MA: MIT Press, 1965), 3.
208 Language Anxiety

champion, I. A. Richards, ‘If a language is to be easy to learn we must not

only cut its words down to a minimum and regularize its grammar; we must
also study very carefully the meanings of every one of its words and decide
upon the central, pivotal or key meaning of each one of them. Parallel to the
reduction and ordering of its vocabulary, there must be a reduction and
ordering of the meanings of the words it recommends.’48 Perhaps the
most famous version of this kind of thinking, however, is Wctional as well as
satiric: Newspeak as Oceania’s medium of consistent if rigidly controlled
communication in the novel Nineteen Eighty-four by George Orwell, himself
a sometime supporter of Basic English.
Not only have such theories shaped the status of variation and change
through their overt emphasis on synchronic structure, however. By cultivating
abstract systems like Saussure’s la langue or Chomsky’s ‘competence’, they
have also furthered the notion articulated by both Locke and the Genesis
account of the Tower of Babel that complete communication is the possible
and reachable goal of human speech. In these systems, when speakers fail to
understand one another, they do so not because such failure is an inalienable
part of the human condition but because they do not speak the same langue or
because their linguistic competence has been obscured by the vicissitudes of
daily performance.49 In all these ways, metalinguistic discussions have made a
twofold contribution to the anxiety surrounding change and variation. First,
of course, they have framed language discussions in such a way as to empha-
size a view of language as static. And second, by agreeing on this one point,
even as they disagree on many others, they have rendered it as part of what
might be called a metalinguistic horizon—a set of ideas whose validity is
demonstrated by the fact that they serve as presumptions in discussions of
other issues, even when these discussions are carried out by critics whose
other views diVer as much as do those of Locke, Tooke, Sir William Jones,
Whorf, and Chomsky.50 Like the connection between social and linguistic
order, that between variation and deviation has become naturalized.

Dialects, dictionaries, and pidgins

Metalinguistic discussions that limit or challenge variation have a peculiar,
counter-intuitive characteristic: in order to identify and advocate a standard,
they need also to identify the non-standard forms and varieties that the

48 Richards, Basic English and Its Uses (London: Kegan Paul, 1943), 22.
49 Harris, The Language Machine, 44–5.
50 AarsleV explains that part of what motivated Jones were his objections to the work of Tooke and
others as impressionistic and even whimsical (The Study of Language in England, 127–32).
 Say the Right Thing 209

standard will avoid. While a standard may have one acceptable spelling,
meaning, or pronunciation for a word, there may be many non-standard
ones, and the act of specifying the standard will likely entail specifying, at least
implicitly, all the likely unacceptable alternatives as well. When used as an
adverb, for example, there is the only correct spelling, not their, they’re, or thar,
and the correct pronunciation of this word is [ðEr] or [ðej#r], maybe even
[ðAr], but not [bEr], [ðEn], [ðæt], and so forth. At a theoretical level the
number of possible non-standard variations of any given standard form is
virtually inWnite, although because utterly random variation allows for no
language at all, actual non-standard variation is rather more limited for
spoken and written language. In any case, all such non-standard forms serve
a crucial role in standardization: their existence constitutes the variation from
among which standard forms can be isolated. In a grandly metaphorical way,
they might even be said to represent the competing narrative of variation in
Genesis 10 to that of stasis in Genesis 11. Standards need not simply to
acknowledge non-standard forms, then, but even in a sense to maintain
them. A language that somehow truly lacked variation, with uniform pro-
nunciation and syntax by all users in all circumstances, could not itself be
regarded as a standard, for if there is only one form for any given linguistic
item, there cannot be a selection of that form. By the same token, if there
somehow were only one variety of a given language, already necessarily used
in all domains, there could be no elaboration of that language’s function.
Even codiWcation would seem to be impossible, since all it could do would be
to identify invariable usage: prescriptive and descriptive grammar would
be the same.
In the early modern period, commentary on non-standard varieties in fact
proliferated alongside grammars and dictionaries. One strain of thought at
this time Xirted with contradicting the entire ideology of standardization by
accepting change and variation as inevitable. As the preface to his Logonomia
Anglica, for example, Gill’s description of the history of English and its many
changes (‘quam tandem mutacionem’) serves as the work’s premise, and he
later identiWes six regional varieties of English, oVering the most detailed
comparative account of their structure up to that time. To illustrate the
‘Borealium’ (northern) variety, he points to a distinctive centralized variant
of the diphthong that had developed from Middle English /u/: ‘au for ou, as
gaun, or also geaun, for goun toga’.51 Likewise evidently embracing Genesis 10,
Carew went so far as to regard such regional varieties as one of English’s
strengths: ‘Moreover the copiousness of our Language appeareth in the

51 See Gill, Logonomia Anglica, 16.

210 Language Anxiety

diversity of our Dialects, for we have Court and we have Country English, we
have Northern and Southern, gross and ordinary, which diVer each from
other, not only in the terminations, but also in many words, terms, and
phrases, and express the same thing in diverse sorts, yet all write English
alike.’52 More moderately, Verstegan compares English in particular to the
dispersion of the Germanic languages in general, and notes that change is
entirely natural. In this regard, he cites the diVerences between the speech of
Londoners and that of ‘the countrey people that neuer borrow any words out
of Latin or French’. For Verstegan, it is the relation between plain geographic
distance and variation that most clearly reveals the latter’s naturalness: ‘wee
may note our Cornishmen, who beeing sequestered from the Welshmen,
but by a little arme of the sea do also varie from them in their language,
though not so much as the Britons in France, who are yet more seperated:
and yet was the language of these three originally one, which their speeches
albeit somewhat diVering, do yet suYciently witnes’.53
Verstegan here sounds very much like a nineteenth-century comparativist,
pointing to parallels in other language families and thus to higher-order
linguistic principles as an explanation for English in particular. Yet his largely
descriptive view of English’s social and regional varieties is joined by other far
more judgmental strains of criticism that invested the varieties with social
signiWcance, much as the standardizing tradition did for individual forms.
Over forty years earlier Hart, too, had noted that ‘Tongues haue often
chaunged’. But he went on to suggest that the naturalness of change did not
prevent change from being abused. If ‘the fancies of men’ have produced
linguistic change and variation, they ought also be able to ‘correct the vicious
writing of the speach’.54 For all of his own descriptivism, Gill passes even
stronger judgement, suggesting that use and pronunciation in fact lead to
linguistic corruption and identifying the Western variety of English as so
corrupt as to be scarcely recognizable as English. The ‘Occidentalium’, he
observes, is the ‘greatest barbarity. And in fact if you should hear a farmer in
Somerset, you would easily wonder whether English or some foreign
tongue was being spoken.’55 This view of dialects assumed a kind of formal
status in John Bullokar’s 1616 English Expositor. The Wrst English dictionary
to deWne the word dialect, the Expositor did so in a way that represented
dialects as divergences from some putatively accentless variety—a view that
persists, of course, into the present: ‘Dialect. A diVerence of some words, or

52 Carew, ‘The Excellency of the English Tongue’, 49.

53 Verstegan, Restitution, 195, 198.
54 Hart, An Orthographie, Diir .
55 Gill, Logonomia Anglica, 18.
 Say the Right Thing 211

pronunciation in any language: as in England, the Dialect or manner of

speech in the North, is diVerent from that in the South, and the Western
Dialect diVering from them both . . . So euery countrey hath commonly in
diuers parts thereof some diVerence of language, which is called the Dialect of
that place.’56 But it was Sir Isaac Newton, of all people, who articulated this
judgemental view of regional variation in the broadest critical perspective:
‘The dialects of each language [are] soe divers and arbitrary [that] a generall
language cannot bee so Wtly deduced from them as from ye natures of things
themselves wch is ye same in all nations and by which all language was at
ye Wrst composed.’57 From the primacy of nature descends language, and
from language evolve dialects, which are to that extent farther removed from
and less expressive of nature.
As I noted in Chapter 3, at least a superWcial awareness of England’s dialect
diversity, both social and regional, can be traced to the Middle Ages. In the
thirteenth century, Roger Bacon commented in this way on the general
regional diVerences of English: ‘We see also that within the same language
there are various idioms [diversa idiomata], that is manners and peculiarities
of speech, as there are in English among northerners, southerners, easterners,
and westerners.’58 And a century earlier, in his Gesta PontiWcum, William of
Malmesbury not only acknowledged regional variation but also complained
that some varieties were not mutually intelligible with others and hinted at
the emergence of sociolects as well: ‘Indeed, the entire language of the
Northumbrians, especially in York, grates so stridently that none of us
southerners is able to understand it. This situation came about because the
north is in the proximity of the barbarians, with the result that it was distant
from the former English kings and the current Norman ones, who are known
to be situated more in the south than the north.’59 While Malmesbury ties
speech patterns to social position, and oVers more than a hint of regional
pride and even disdain, he does not explicitly situate regional varieties within
an interpretive hierarchy that connects social features like rank, occupation,
and sex to linguistic features like vocabulary and pronunciation. Both Bacon
and Malmesbury, indeed, oVer remarks more suggestive of idle curiosity
than evocative of the systematic identiWcation of discrete varieties whose

56 Bullokar, An English Expositor: Teaching the Interpretation of the Hardest Words Used in Our
Language (London: John Legatt, 1616), s. v. Blank identiWes this as the earliest such deWnition in‘The
Babel of Renaissance English’, 214.
57 Quoted in Edward G. Gray, New World Babel: Languages and Nations in Early America (Prince-
ton, NJ: Princeton University Press, 1999), 67.
58 Bacon, Opera Quædam Hactenus Inedita, ed. J. S. Brewer (London: Longman, 1859), 467.
59 William of Malmesbury, De Gestis PontiWcum Anglorum, ed. N. E. S. A. Hamilton, Rolls Series 52
(London: Longman, 1870), 209.
212 Language Anxiety

social meanings creative writers could exploit in the way I earlier described
as narratives of language change. And this remains the case through the
Middle Ages, with Chaucer, John Trevisa, Caxton, and others all testifying
to an awareness of regional variation but not to a modern sociolinguistic
stratiWcation of dialects. Within late-medieval England’s linguistic repertoire,
indeed, the pre-eminent distinction remained not that among varieties of
English but between English and non-English languages.60
What changed in the early modern period, as part of the linguistic attitudes
that a standard ideology fosters, was the emergence of a metalinguistic
discussion to frame the status of regional variation and change. When meas-
ured against a standard, such variation became—had to become—inaccur-
acies and errors that need to be corrected. Beyond this, non-standard
varieties’ essential roles in the enforcement of a standard increased, since
the errors they demonstrated acquired social as well as linguistic signiWcance.
Put another way, when vested with social implications, variation and change
became more potent, enforcing social and linguistic norms in mutually
supportive ways. And this connection between linguistic and social structure
has become so strong in the history of English since the early modern period
as to constitute, I would argue, a kind of symbiosis. Without the embedding
of social and linguistic practices in one another—speciWcally, the channeling
of social issues through variation and change—Standard English would itself
not have been possible.
While Carew may have thought that ‘all write English alike’, despite the
regional variations in their language, Gill describes a linguistically and socially
variegated speech community. But the linguistic variation he sees in this
community is itself socially charged. He sees the speech of the cultured class
as inherently stable, and variation and change as natural only among what he
describes as ‘rustics’: ‘And what I say about these dialects, I would wish you to
understand relates only to rustics; for among gentler temperaments, there is
everywhere only one speech, pronunciation, and meaning—more cultivated
by upbringing.’61 In his 1589 The Arte of English Poesie, George Puttenham
imbeds variation even more deeply in the geographic and social landscape.
For Puttenham, in a passage I cited in full in Chapter 2, the varieties spoken in
the country and by cliques of academics or lower-rank, uneducated individ-
uals are linguistically inferior—odd, misshapen, and false—compared to that
spoken by the cultured and educated. Anyone looking for the best language,
according to Puttenham, should not look in ‘marches and frontiers’ or even

60 See Machan, English in the Middle Ages (Oxford: Oxford University Press, 2003), 71–110.
61 Gill, Logonomia Anglica, 19.
 Say the Right Thing 213

universities or ‘in any vplandish village or corner of a Realme’, but rather

should follow ‘generally the better brought vp sort . . . men ciuill and gra-
ciously behauoured and bred’.62 The role of education in Puttenham’s analysis
points to a signiWcant development in the use of linguistic performance to
channel other kinds of anxiety. By his analysis, not simply did variation
and change become identiWed with non-standard varieties but the speakers
of those varieties became fated, at least for the foreseeable future, to perpetu-
ate them. This is so because Standard English, spoken or written, was an
art acquired at home and cultivated in schools and universities, already
the special provenance of an intellectual, economic, and political elite.
Not to attend a university or the right school is not to acquire the right
variety of English and therefore not to belong with England’s foremost
sociolinguistic group.
Connections between non-standard varieties, social status, and variation
and change developed at both the level of the varieties themselves and at that
of the details of their individual grammars. Invested with social signiWcance,
speciWc linguistic forms like titles of address or the second person personal
pronoun (whether thou or ye) served as tokens of social display that identiWed
speakers’ varieties and their concomitant social standing. Master, as a preface
to a surname, thus marked a gentleman, just as surely as thou projected
disdain onto one’s interlocutor—a point wittily made in Shakespeare’s
Twelfth Night, when Sir Toby Belch, in his eVorts to arouse Sir Andrew
Aguecheek to compose a letter of challenge to what he imagines to be a
rival suitor, observes: ‘Taunt him the license of ink. If thou thou’st him
some thrice, it shall not be amiss.’63
More generally, Puttenham’s comments on class and culture portend what
would harden into a discourse for talking about dialects, as in Thomas
Sheridan’s 1762 Course of Lectures on Elocution, where he identiWes two general
varieties in London: Cockney, which is current ‘in the city’, and the ‘polite
pronunciation’, which is found ‘at the court-end’. Of these two, the court
dialect is far and away the most fashionable, and it can be learned only by
speakers who have already been admitted to its speech community and whose
speech, in circular fashion, conWrms their social standing: ‘by conversing with
people in polite life, it is a sort of proof that a person has kept good company’.
‘All other dialects’, says Sheridan, ‘are sure marks, either of a provincial,
rustic, pedantic, or mechanic education; and therefore have some degree of

62 Puttenham, The Arte of English Poesie (1589; rpt. Menston: Scolar Press, 1968), 120.
63 Twelfth Night, 3.2.39–40. See Joseph M. Williams, ‘ ‘‘O! When Degree is Shak’d’’: Sixteenth-
Century Anticipations of Some Modern Attitudes toward Usage’, in T. W. Machan and Charles T. Scott
(eds.), English in Its Social Contexts (New York: Oxford University Press, 1992), 69–101.
214 Language Anxiety

disgrace annexed to them.’64 As disinterested as Old English and even most

Middle English writers may have been in associating social standing and
achievement with linguistic forms and varieties, this association between
politeness, good company, and stable, standard language—and, conversely,
between provincialism, variation, and non-standard language—had in fact
become commonplace in Sheridan’s day. In 1789, for example, Philip Withers
matter-of-factly observed in a way that could well have inspired Shaw’s
Henry Higgins not only that language remained the best opportunity for
the upper class to demonstrate its rank but also that the language of the
lower classes was not even worthy of attention:
It is no Part of my Plan to notice all the Phrases current in Covent Garden and the
Purlieus of St. Giles. They are below rational criticism. My Animadversions will extend
to such Phrases only as People in decent Life inadvertently adopt.
In former Ages, a Gentleman was easily distinguished from the Multitude by his
DRESS. In the present Period, all external Evidence of Rank among Men is destroyed.
Every outward Distinction is also lost in the female World . . . It is utterly impossible,
on a Wrst View, to determine the Rank of this Boarding-School Miss. But the Moment
the unhappy Girl attempts to speak, her Origin is disclosed, and her Finery and
aVected Airs excite sentiments of Pity and Contempt. Hence the Importance of early
attention to Purity and Politeness of Expression: it is the only external Distinction
which remains between a Gentleman and a Valet; a Lady and a Mantua-maker.65

Bad language has provided one particularly eVective way by which vari-
ation might further these discriminations between gentlemen and their valets,
both literal and metaphorical. It’s reasonable to suppose that as long as
humans could speak, they have sworn and blasphemed, and the Ten Com-
mandments in fact expressly forbid taking God’s name in vain. In the twelfth
century Henry I attempted to control this type of variation by imposing a
system of Wnes for swearing in royal residences that was proportionately
linked to social standing—to whether the speaker was a lord, squire, yeoman
or page.66 But it is again the early modern period, especially the seventeenth
and eighteenth centuries, that oVered the Wrst consistent distinctions between
good language and various kinds of bad language, including swearing and
blasphemy, and that systematically tied those distinctions to social standing.
Just as an increased attention to the use of titles of address helped language to
stabilize shifting social ranks in early modern England, so the identiWcation
and proscription of improper language allowed merchants, yeomen, and the

64 Sheridan, A Course of Lectures on Elocution (1762; Menston: Scolar Press, 1968), 30.
65 Withers, Aristarchus, or the Principles of Composition, 2nd edn. (London: J. Moore, 1789), 160–1.
66 Ashley Montagu, The Anatomy of Swearing, 2nd edn. (London: Macmillan, 1973), 108.
 Say the Right Thing 215

like to distinguish themselves from a lower rank of rougher speaking, less

educated but socially aspirant individuals. If in some situations swearing can
be a form of dominance, that is, in others so, too, can its avoidance be.
What began as a sixteenth-century Puritan eVort to censor bad language
assumed increasing urgency and still greater moral overtones in the decades
to come. Following the Restoration, thus, there emerged several societies
devoted to cultivating religion and manners, and within this context language
variation in the form of improper language served as a symptom of larger
social ills. For this reason, groups like the Society for the Reformation of
Manners and the Society for the Promotion of Christian Knowledge took
action focused on the elimination of cursing, often through manuals of good
behavior. One of these, Richard Allestree’s 1731 The Whole Duty of Man,
devoted twelve pages to swearing as both a religious and social transgression:
a sin against God but also a mark of unreliability and dishonesty.67 While the
SPCK pursued its ends through charity schools using books like Allestree’s,
the SRM attempted to legislate against bad language in a formal way, and both
groups generally channeled several social issues through the variation from
good speech that bad language emblemized. Bad language, thus, became
regarded as a characteristic of the lower ranks, even if, as with phonological
features like ‘dropped hs’, it may be in reality no more common among them
than among the educated middle and upper ranks. And what makes bad
language bad are likewise the larger issues it channels: it points to social
unrest, it produces social unrest, and it constitutes social unrest. To the extent
that at least early metalinguisitic discussion stressed the importance for
women in particular to use good language, this form of linguistic variation
acquired gendered qualities as well, leading, ironically, to the sociolinguistic
disparagement of feminized speech, whether by women or men. In other
words, to vary from a putative norm and speak as a woman, even if this
means avoiding a marked discursive habit like swearing, has often been the
justiWcation for social marginalization.68
While swearing and bad language may have become less provocative now
than in the eighteenth century, they can still serve as vehicles for issues of class
and social propriety. In the 1960s and 1970s, Mary Whitehouse, founder of the
National Viewers’ and Listeners’ Association, was a particularly insistent
British voice on the need to curtail all forms of bad language in books and

67 Tony McEnery, Swearing in English: Bad Language, Purity and Power from 1586 to the Present
(London: Routledge, 2006), 94–7.
68 For a historical survey of this issue, see Bailey, Images of English, 246–66. For broader consider-
ation of the sociolinguistic marginalization of women’s speech, see Deborah Cameron, Feminism and
Linguistic Theory, 2nd edn. (New York: St Martin’s Press, 1992).
216 Language Anxiety

on television and radio. Warning often of obscenity, blasphemy, and drugs,

and taking various kinds of legal action against particular works and produc-
tions, Whitehouse had larger objectives consistent with this history of man-
aging social issues through language: ‘ ‘‘Let’s all be kids together’’, cry the
adults as if only the fantasies of child’s play help them to come to terms with
what they have made of the world. And in so doing, they deny the young
the basic security without which they cannot grow to a mature and respon-
sible exercise of freedom.’69 Like the critics of Webster’s Third, writing at
nearly the same time, Whitehouse was most troubled about features of
language that were not linguistic at all.
In this context, I want to return to dictionaries and their own roles in
specifying Standard English and fostering variation and change as conduits of
extra-linguistic issues.70 The earliest dictionaries in England are medieval
wordlists like the Medulla Grammatice or the Promptorium Parvulorum,
which tend to list Latin and English in one-word equivalents and which
circulated in relatively few copies. While such works would seem to have little
extra-linguistic impact, the social implications of dictionaries and their iden-
tiWcations of language, change, and variation do Wgure already in the early
modern period. As I noted in Chapter 3, the earliest dictionaries to focus on
slang, cant, and other linguistic variation did so in order to enlighten good
citizens of the coded language that thieves and vagrants used to plot against
them. Writing in 1610 about how one Cock Lorrell, in 1501, had organized a
group of vagrants, Samuel Rid observes, ‘Wrst of all they thinke it Wt to deuise a
certaine kinde of language, to the end their cousenings, knaueries and villanies
might not be so easily perceiued and knowne, in places where they come’.71
This same association between dictionaries’ potential to stabilize language and
society as one maneuver in general is foundational in the development of
English lexicography following Cawdrey’s 1604 A Table Alphabeticall. As an
aspect of codiWcation, indeed, English dictionaries emerged against a back-
ground of religious and national self-deWnition, cultural aggrandizement, and
global expansion. If France, Spain, and Holland oVered competition for
colonies and foreign markets, Latin and Greek (again) represented languages
with expansive and nuanced lexicons, traditions of grammatical exegesis, and
unimpeachable records of rhetorical achievement. In their constancy, regular-
ity, and imperviousness to change, they were what speakers of every early
modern vernacular hoped for in their own language.
69 Quoted in McEnery, Swearing in English, 125.
70 In fact, the earliest citation in the OED to ‘Standard English’ in its current sense is from the
dictionary’s own original 1859 proposal for publication (s. v., standard, n. (a.), B.I.3.e).
71 Quoted in Blank, ‘The Babel of Renaissance English’, 226. See further Julie Coleman, A History of
Cant and Slang Dictionaries, i, 1567–1784 (Oxford: Oxford University Press, 2004).
 Say the Right Thing 217

For English dictionaries, the challenge was to negotiate a compromise

between the perceived need to supplement the native vocabulary with im-
portations of Greek and Latin terms and thereby enhance the reputation as
well as expressibility of English vis-à-vis the classical languages and other
vernaculars; and a reluctance to transform the language into something so
monstrous that it could no longer stand as a symbol of English achievement.
In the same way that Webster would later frame discussion of English in
America, change might have been accepted as necessary, but only up to some
negotiable point, at which it would be replaced by stability. On one hand,
then, Thomas Blount’s Glossographia of 1656 accepted and recorded words
like advigilate (‘to watch diligently’) and adoxy (‘ignominy’). On the other,
Edward Phillips’s 1658 The New World of English Words dispensed such
inkhorn terms, including cinerulent (‘full of dust or ashes’) and cacologie
(‘evil communication’), in a list of barbarous words and those ‘illegally
compounded’ or derived from Latin and Greek.72 At either extreme, early
dictionaries in general volatilely reXected social change (whether for good
or ill) in the words they recorded or excluded, and they thereby functioned
as something of far greater consequence than did medieval wordlists.
Faced with the kinds of theoretical and practical diYculties occasioned by
language change and variation, lexicographers have perhaps not surprisingly
been more Xexible and less authoritarian than many of their critics. What
were certain and Wxed for Defoe, Swift, Follett, or Macdonald—the character
of English lexicon, morphology, and registers—were provisional and evolving
for Gove and predecessors like James Murray and Dr Johnson. The latter
began his lexicographical work exhibiting conWdence and certitude in his
linguistic judgements. His 1747 Plan of an English Dictionary meticulously
examines the many practical diYculties involving borrowed words, pronun-
ciation, orthography, etymology, inXections, and so forth. Summing up his
objectives, Johnson announces to the Earl of ChesterWeld, ‘This, my Lord, is
my idea of an English dictionary; a dictionary by which the pronunciation
of our language may be Wxed, and its attainment facilitated; by which its
purity may be preserved, its use ascertained, and its duration lengthened’.73
Elsewhere, in place of ‘purity’, Johnson and others focus in particular on
‘propriety’ as a distinctively eighteenth-century way once more to meld
together grammatical correctness and social stability.74 And in this regard,
it’s signiWcant that while Dante’s deWnition of grammar emphasized primarily
a stabilizing quality, English deWnitions recalled Quintilian in also emphasiz-
ing the notion that grammar was the art of speaking and writing correctly.
72 Starnes and Noyes, The English Dictionary from Cawdrey to Johnson, 1604–1755, 41–2, 55–6.
73 Johnson, The Plan of a Dictionary of the English Language (London: J. and P. Knapton, 1747), 32.
74 See Lerer, Inventing English, 179–80.
218 Language Anxiety

By the time he completed his 1755 A Dictionary of the English Language,

however, Johnson had discovered that his earlier views about linguistic
constancy and purity were theoretically and practically unsustainable. He
still held strongly enough to a prescriptive view of grammar to describe
shabby, for instance, as ‘A word that has crept into conversation and
low writing; but ought not to be admitted into the language’.75 But in the
preface Johnson described language change as being as natural as, and there-
fore no more preventable than, diurnal change in general. As all humans age,
Johnson notes, so, too, do all languages continue to supplement, diminish,
and change their lexicons:
When we see men grow old and die at a certain time one after another, from century
to century, we laugh at the elixir that promises to prolong life to a thousand years; and
with equal justice may the lexicographer be derided, who being able to produce no
example of a nation that has preserved their words and phrases from mutability, shall
imagine that his dictionary can embalm his language, and secure it from corruption
and decay, that it is in his power to change sublunary nature, or clear the world at once
from folly, vanity, and aVectation.76

The comparison here to folly and vanity is striking, suggesting as it does that
language change may well be sometimes regrettable, sometimes detrimental,
and sometimes unnecessary. But it remains inevitable. And this inevitability
leaves lexicographers struggling to decide not only what words to include but
how to deWne their core senses, from which regional or other non-standard
usages might be seen to deviate.
Given the social issues implicated in the deWnition of standard and non-
standard language, a sociolinguistic gesture as powerful as this has predictably
produced diVerences of opinion. To many of Gove’s critics, for example,
despite their belief that Webster’s Third had shirked its historical responsibil-
ities, that dictionary’s presentation of the oldest deWnition as an entry’s initial
sense ran contrary to the best interests of users who wanted to know the most
widespread current deWnition. For Johnson, writing against a backdrop of
universal languages and the continuing inXuence of the Tower of Babel, this
same conXict took shape as one between modern usage and etymological
meaning, and Johnson’s sentiments, unlike many of his contemporaries,
rested with the former.77 In this regard at least, Gove thus oddly appears the
conservative and Johnson the modernist.

75 Johnson, Dictionary of the English Language (1755; rpt. New York: AMS Press, 1967), s. v.
76 Johnson, Dictionary of the English Language, i, Ciir .
77 See Allen Reddick, The Making of Johnson’s Dictionary, 1746–1773 (Cambridge: Cambridge
University Press, 1990), 45–51.
 Say the Right Thing 219

In the Oxford English Dictionary, Murray confronted this same contradic-

tion between the desire to delineate and deWne the vocabulary of English and
the recognition that vocabulary and usage are fundamentally open. Following
Johnson’s lead, his preface also draws a comparison between lexicography and
observations of the natural world, in both of which distinctions among
individuals (objects, creatures, or words) become complicated by the changes
of time and the variation of species. A naturalist may identify certain proto-
typical objects or individuals that unmistakably form a core classiWcation and
use them to draw distinct species boundaries, even though the boundaries
themselves must remain permeable to synchronic and diachronic variants.
‘So the English Vocabulary’, the preface continues,
contains a nucleus or central mass of many thousand words whose ‘Anglicity’ is
unquestioned; some of them only literary, some of them only colloquial, the great
majority at once literary and colloquial—they are the Common Words of the language.
But they are linked on every side with other words which are less and less entitled
to this appellation, and which pertain ever more and more distinctly to the domain of
local dialect, of the slang and cant of ‘sets’ and classes, of the peculiar technicalities
of trades and processes, of the scientiWc terminology common to all civilized nations,
and of the actual languages of other lands and peoples. And there is absolutely
no deWning line in any direction: the circle of the English language has a well-deWned
centre but no discernible circumference.78

Murray spoke like a scientist and philologist (maybe even, with his echo of
Alain de Lille’s deWnition of God, as a theologian)—someone well versed in
languages and linguistic theory. But if the OED, like many dictionaries,
aspired to pure objectivity in its descriptions of words and their histories,
culturally determined issues of propriety and prestige inevitably shaped its
deWnitions and the limits of which linguistic changes (in vocabulary or word
meaning) were acceptable as English. As a Victorian project, in fact, the OED
excluded common vulgarity and labeled certain words and usages as archaic,
colloquial, and dialectal.79 The project was the 1857 brainchild of Richard
Trench, the archbishop of Dublin, who conceived the dictionary to sustain
ethics and faith and not merely lexicography. And it was a project that in
documenting the history of English constituted a national achievement,
perhaps mirroring the 1856 opening of the National Portrait Gallery and the
1882 publication of the Dictionary of National Biography. All of these contem-
poraneous projects, according to John Willinsky, were understood to give
‘proof of an advanced and advancing civilization’.80

78 Murray et al., A New English Dictionary on Historical Principles, 10 vols. (Oxford: Clarendon
Press, 1888–1928), i, xvii.
79 See further Mugglestone, Lost for Words, 143–78.
80 Willinsky, Empire of Words, 24.
220 Language Anxiety

The metalinguistic discussions of variation and change that I have been

tracing here have several hallmarks. First, developing as they do from the
standardization of English and the ideology of a standard, they involve very
much a set of historical circumstances. And this means that as powerful and
naturalized as the notion of variation as an aberration or deviation may be
today, this notion has speciWc origins and does not reXect a transcendent
or inevitable quality of language. Second, these origins establish Standard
English and its putative absence of variation and change as, in eVect, the birth
language of a ruling elite who had the most to gain from this connection, who
alone (at least early on) had the means to acquire and use one particular
variety of the language, and who socially as well as linguistically had the most
to gain by preventing change in general. And it should be added that while
Standard English may have helped to sustain this elite’s privileges, it did not
create or deWne them; wealth, education, and social inXuence did that. Third,
established in this way standard language and the variable non-standards
it required served as general sociolinguistic principles and precedents that
might be activated in response to any number of speech communities: they
were templates that could be Wlled with various social meanings rather than
simply isolated, unique, and Wxed constructs. If linguistic variation could be
used to demarcate the educated from the uneducated, for instance, or the
provincial from the sophisticated, so, too, might it discriminate in sex, sexual
orientation, ethnicity, national identity, character, personal outlook, and even
patriotism. Variation and change could funnel, in fact, any social issue. And
fourth, in its attention to grammar as well as the linguistic repertoire, the
focus of this metalinguistic commentary has characteristically been arbitrary
and opportunistic. Verbal inXections could have been invested as easily and
logically with social meaning as personal pronouns, while the English of Kent
or Wessex might have as easily been stigmatized as that of Somerset or, more
frequently, northern England. And from a diVerent perspective, the emer-
gence of a London variety as the basis of Standard English depended on the
proximity of printing presses and social and economic power rather than on
any intrinsic virtues of that variety. When cultural and linguistic traditions
reinforce one another, as in the stigmatization of northern British English or
African American Vernacular English, they may mystify this opportunism,
but they cannot validate it.
What all this means is that by the time dialectology began to develop as
a subdiscipline of linguistics in the nineteenth century, metalinguistic
commentary had already fashioned variation and change as degenerative
processes and the regional and social varieties they produce as conWrma-
tions thereof. All such varieties, further, could carry implications for class,
 Say the Right Thing 221

education, social prestige, and so forth. Given the ideology of a standard, and
the imagined constancy and consistency of Standard English, dialects could
appear, variously, as atavistic evocations of a simpler, antique historical
moment, or as corruptions born of ignorance and dissipation. Foreshadowed
in the confusion produced by the use of ‘egges’ for ‘eiren’ by Caxton’s
merchant, both views share the sense that dialects are aberrations that, in
the presence of Standard English, should be or already are disappearing.
In his monumental 1905 English Dialect Grammar, even Joseph Wright
invoked the expectations of standard ideology in both his notion of the ‘purity’
of dialects and in his sense of their evanescence in the evolution of a superior,
stable variety. ‘There can be no doubt that pure dialect speech is rapidly
disappearing even in country districts’, Wright observes, ‘owing to the spread
of education, and to modern facilities for intercommunication. The writing of
this grammar was begun none too soon, for had it been delayed for another
twenty years I believe it would by then be quite impossible to get together
suYcient pure dialect material to enable any one to give even a mere outline of
the phonology of our dialects as they existed at the close of the nineteenth
century.’ In many cases, for Wright, inculcation of the standard through
education is evidently succeeding—or at least believed to be succeeding—in
erasing variation. Put another way, the disappearance of pure dialects proves
the success of English culture, just as does Wright’s contention that in lower-
class, uneducated, urban London ‘the dialects are hopelessly mixed and now
practically worthless for philological purposes’.81 As Jonathan Marshall has
pointed out, the presumption underlying such foundational dialectology—a
presumption that embodied the expectations of a standard and that Mufwene
identiWes in analyses of African American Vernacular English—was that even
regional dialects were (or should be) inherently stable: ‘The methods used were
not designed to deal with the fact that the same speaker may use a very wide
variety of diVerent pronunciations, and explanations for the variation were not
normally to be found. Traditional dialectology focused on regions having
‘‘place’’, ‘‘diVerence’’, and ‘‘distinctiveness’’ as their most prominent features
to be analyzed.’82 What varieties of English could not do within the context of a
standard ideology was function, simply, as testaments to natural processes of
change and variation.
81 Wright, The English Dialect Grammar (Oxford: Oxford University Press, 1905), vii, viii. For an
account of the development of English dialectology in the nineteenth century, see Ossi Ihalainen, ‘The
Dialects of England Since 1776’, in The Cambridge History of the English Language, v, Robert BurchWeld
(ed.), English in Britain and Overseas: Origins and Development (Cambridge: Cambridge University
Press, 1994), 197–274.
82 Marshall, Language Change and Sociolinguistics: Rethinking Social Networks (New York: Palgrave
Macmillan, 2004), 2.
222 Language Anxiety

From this perspective, the historical reception of global varieties—whether

colonial, commonwealth, or acquired as a second language—shows signi-
Wcant parallels with that of indigneous British varieties. In each case, it has
been a dynamic of opposition—standard opposed to non-standard, or British
(and sometimes American) English opposed to transplanted varieties—that
has enabled conceptualization of all the varieties. It’s worth remembering,
indeed, that while 1776 might be advanced as the originary moment of
American English, this same moment, in its creation of the Wrst national
variety outside the United Kingdom, also did much to foster a sense of British
English as well, for British English can exist as a conceptual category only
when it co-exists with other varieties from among which it can be selected.83
And likewise in each case this linguistic dynamic has been invested with social
meanings of the kind witnessed in metalinguistic discussions of British
dialects. For speakers of these transplanted varieties, the sociolinguistic sep-
aration could be turned to purposes of positive self-identiWcation, as was the
case in the early American republic. A 1774 letter to the Royal American
Magazine thus opined, ‘The English Language has been greatly improved in
Britain within a century, but its highest perfection, with every other branch of
human knowledge, is perhaps reserved for this land of light and freedom’.84
And since the early modern period the advancement of Scots as a distinct
language and not a variety of English has been a prominent feature in
discussions of language and nationhood in Scotland.85
But often the variation and change that have produced national varieties
outside of the United Kingdom have been regarded with the same doubt and
derision that have been directed at non-standard varieties. British (and
American) global expansion has in fact exacerbated the complexities of
variation and change in two signiWcant ways in light of the ideology of a
standard. First, in the very centuries when contact between English and other
languages increased in number, kind, and linguistic complexity (i.e., with
non-Indo-European languages), thereby producing probably more varieties
than ever before, at the same moment traditions of codiWcation worked the
hardest to stabilize language and reduce and stigmatize variation. And second,
even as the spread of English has conWrmed the spread of Anglo-American
culture in Africa, the antipodes, Latin America, and Asia, the inevitable

83 John Algeo, ‘What is a Briticism?’, in A. N. Doane, Joan Hall, and Richard Ringler (eds.), Old
English and New: Essays in Language and Linguistics in Honor of Frederic G. Cassidy (New York:
Garland, 1992), 287–304.
84 Quoted in Bailey, Images of English, 103.
85 For a good survey of the historical issues, see J. Derrick McClure, ‘Scotland’, in BurchWeld (ed.),
English in Britain and Overseas, 23–93.
 Say the Right Thing 223

alterations of English in post-colonial situations have furthered anxiety about

the state of the language and the signiWcance of change and variation. To put
the matter more pointedly: sociolinguistic impulses to stabilize language in
fact have worked in a diametric contradiction with economic impulses to
expand markets, which necessarily also created more contact situations and
more second-language learners. This contradiction animates the poet Robert
Bridges’s summation of the motives of the Society for Pure English, founded
in Oxford in 1913. Without irony or any recognition of inconsistency, he
notes that two considerations called the society into existence: the ‘English
language is spreading all over the world’, and
It would seem that no other language can ever have had its central force so dissi-
pated—and even this does not exhaust the description of our speciWc peril, because
there is furthermore this most obnoxious condition, namely, that wherever our
countrymen are settled abroad there are alongside them communities of other-
speaking races, who, maintaining among themselves their native speech, learn
yet enough of ours to mutilate it, and establishing among themselves all kinds
of blundering corruptions, through habitual intercourse infect therewith the neigh-
bouring English.86

British disparagement of American English provided much of the counter-

point that drove Webster’s arguments for a uniquely American language, but a
strain of skepticism about the legitimacy of English in the United States has
remained to the present. As recently as 1979, a peer in the House of Lords
exclaimed, ‘If there is a more hideous language on the face of the earth than
the American form of English, I should like to know what it is!’87 Similar
skepticism over transplanted varieties of English has also Wgured in the larger
metalinguistic discourses for talking about all such varieties, including those of
Jamaica, Canada, Australia, South Africa, and New Zealand. As in the United
states, in all these cases it was the divergence of the transplanted varieties that
raised concerns. In 1857, for instance, the Revd A. Constable Geikie, addressing
the Canadian Institute and providing the Wrst recorded instance of the phrase
‘Canadian English’, described this variety as ‘a corrupt dialect growing up
amongst our population, and gradually Wnding access to our periodical litera-
ture, until it threatens to produce a language as unlike our noble mother tongue

86 Bridges, ‘The Society’s Work’ (Society for Pure English, tract XX, 1925), in W. F. Bolton and
David Crystal (eds.), The English Language, 2 vols., ii, Essays by Linguists and Men of Letters, 1858–1964
(Cambridge: Cambridge University Press, 1969), 87–8.
87 Quoted in Richard W. Bailey, ‘American English Abroad’, in Algeo (ed.), English in North
America, 495.
224 Language Anxiety

as the negro patua or the Chinese pidgeon English’.88 An 1886 editorial in a

Hawaiian paper saw the divergences of a developing indigenous variety—
including the use of a pidgin—from United States English as a development
requiring classroom attention: ‘It is a sober duty for every instructor of Hawaiian
youth to check the use of pigeon-English. Very much can be done by watchful-
ness in this particular. And as we look forward into the years, and think of the
possibilities, there is every incentive to make teachers chary in their use of
doubtful English, and alert to correct the language of playground and street.’89
And in 1910, E. W. Andrews had characterized New Zealand English as arising
from simple failure to duplicate correct British English: ‘There is not enough
diVerence between the environments of the Englishman and the New Zealander
to produce the existing diVerence in pronunciation. It should evidently be the
teacher’s aim to stay the process, and if possible restore to the New Zealand
speech the culture it has unfortunately lost. We must, therefore[,] examine the
faults one by one, and enumerate the deWnite sounds of English that the colonial
ear has failed to catch and reproduce.’90
India provides still another paradoxical illustration of how the divergence
of a transplanted variety from particularly Standard English has marked out
that variety and its speakers for calumny and derision. While Macaulay’s 1835
minute encouraged the spread of English among indigenous people of India,
for example, the inevitably distinctive form English in India took from natural
processes of borrowing, nativization, and the like served as the object of
ridicule. In his 1891 Baboo English as ’tis Writ, Arnold Wright described native
Indian papers as ‘badly printed, badly written’, and saw their English, over a
century after its introduction, as not a variety but a corruption born of
ignorance and insensitivity to nuance. These papers, he continues, ‘are for
the most part edited by aspiring native students, whose imperfect knowledge
of English leads them to perpetuate most ridiculous blunders. The injudicious
use of metaphors and idioms is perhaps the greatest stumbling block of
the native writers. He has learned a number of expressions by rote, and
is not content unless he is always pressing them into his writings whether
the occasion warrants or not.’91 In eVect, the English of such indigenous

88 Quoted in J. K. Chambers, ‘ ‘‘Lawless and Vulgar Innovations’’: Victorian Views of Canadian

English’, in Sandra Clarke (ed.), Focus on Canada (Amsterdam: John Benjamins, 1993), 6. See further,
Richard W. Bailey, ‘The English Language in Canada’, in Bailey and Manfred Görlach (eds.), English as
a World Language (Ann Arbor, MI: University of Michigan Press, 1982), 134–76.
89 Quoted in Albert J. Schütz, The Voices of Eden: A History of Hawaiian Language Studies
(Honolulu, HI: University of Hawaii Press, 1994), 301.
90 Quoted in Gordon et al., New Zealand English, 6.
91 Wright, Baboo English as ’tis Writ: Being Curiosities of Indian Journalism (London: T. Fisher
Unwin, 1891), 17.
 Say the Right Thing 225

populations serves as both conWrmation and justiWcation of Anglo-American

political and economic inXuence: its existence conWrms the existence of
the latter, even as its quality also provides justiWcation for the latter’s intro-
duction. Given the inXuence of the standard ideology in metalinguistic
thinking, speakers of non-standard or national varieties can become compli-
cit in their own subordination, furthering negative stereotypes of their own
speech. This continues to be the case for some American speech communities
and was so, at least until relatively recently, in Australia and other post-
colonial countries.92
The interlanguages that have resulted from English’s contact with indigen-
ous languages through global expansion should also be considered in this
light. Born of immediate social necessity, pidgins and creoles have terriWc
social utility, testifying for the vitality of languages and speech communities
in contact with one another. And so it comes as no surprise that on occasion
the speech communities that use them can rival those using natural languages
in duration and number of speakers. Mobilian Jargon, composed from
English and several indigenous languages, persisted across the southern
United States from the early eighteenth century until the middle of the
twentieth, for instance, and in Papua New Guinea Tok Pisin has become a
national language, used alongside English in parliamentary debates, educa-
tion, and various domestic settings. In the Philippines, Taglish is the most
widely used variety of English.
What may be surprising is that for all this, Anglophones and metalinguistic
theory have historically responded to these interlanguages and the variation
and change they embody with the same double bind that I have described for
transplanted varieties of English. As products of trade, missionary contact, and
colonial expansion in general, pidgins and creoles are inevitable byproducts
of history and might even be seen as aYrmations of Anglophones’ success. At
the same time, they have been regarded as monstrous, the vulgar speech of
vulgar people. Early North American colonists, thus, routinely commented on
the stupidity of the American Indians and the pidgins used to communicate
with them. This inability to empathize linguistically with other speech
communities is at times striking, as in the Revd John Megapolensis’s attempts,
in the 1640s, to learn Mohawk by means of the structural categories of Latin.
Whether his informants used Mohawk or a pidgin to provide equivalents for

92 On Australia see John Gunn, ‘Social Contexts in the History of Australian English’, in Machan
and Scott (eds.), English in Its Social Contexts, 204–29. On the processes of linguistic subordination
more generally, see Rosina Lippi-Green, English with an Accent: Language, Ideology, and Discrimination
in the United States (London: Routledge, 1997). For additional examples of British disparagement of
transplanted English, see Bailey, Images of English, 130–3.
226 Language Anxiety

English vocabulary is unknown, for Megapolensis could understand the

speech only through its failure to adhere to recognizable (i.e., Latinate)
linguistic structure: ‘one tells me the word in the inWnitive mood, another in
the indicative; one in the Wrst, another in the second person; one in the present,
another in the preterit’.93 Megapolensis’s attitude is mostly of disbelief, but
Anglophones in Hawaii took a more socially aggressive approach in the
1920s, by which time Hawaiian had all but disappeared, replaced by English,
Hawaiian Pidgin English, and Hawaiian Creole English. The speech groups of
the languages diVered along class and ethnic lines, with the latter two being
used nearly exclusively by indigenous people and immigrants. As a conWrma-
tion of this distinction, and as a mechanism to maintain it, a two-tier educa-
tional system developed that used proWciency in English as a placement
criterion and that continued into the 1960s.94
For its part, metalinguistic discussion has often sustained such approaches
to interlanguages. Finding no place for them in its stemmatic classiWcation of
language families, and no parallel for them in the linguistic habits and
developing nationalism of Europe, nineteenth-century philologists used
terms like ‘baby talk’ and ‘mongrel dialect’ to refer to pidgins and creoles as
badly learned versions of presumptively stable European languages. Perhaps
necessarily, such linguistic and cultural judgements were mutually reinfor-
cing. In 1911, noting the omission of copulas in pidgins and creoles, William
Churchill generally observed, ‘The savage of our study, like many another
primitive thinker, has no conception of being in the absolute; his speech has
no true verb ‘‘to be’’ ’. And a year later, Beatrice Grimshaw commented in this
way on the indigenous variety of English of Rossel Island, southeast of Papua
New Guinea: ‘To be addressed in reasonably good English of the ‘‘pidgin’’
variety, by hideous savages who made murder a profession, and had never
come into actual contact with civilisation, is an experience perplexing enough
to make the observer wonder if he is awake.’95 To a more recent critic
pidgin English, generically conceived, remains a ‘means of speaking which is
not quite English, and yet is nearer to English than to anything else’.96 For a
time in the middle of the twentieth century, pidgin dialectalism even appeared
as a learning disability alongside reading deWciencies and cleft palates. For its

93 Quoted in Axtell, ‘Babel of Tongues: Communicating with the Indians in Eastern North
America’, in Edward G. Gray and Norman Fiering (eds.), The Language Encounter in the Americas,
1492–1800: A Collection of Essays (New York: Berghahn Books, 2000), 36.
94 Viv Edwards, Multilingualism in the English-Speaking World: Pedigree of Nations (Oxford: Black-
well, 2004), 128–9.
95 Quoted in Peter Mühlhäusler, Pidgin & Creole Linguistics (Oxford: Blackwell, 1986), 26, 27.
96 Charlton Laird, The Miracle of Language (New York: Fawcett, 1953), 238–9.
 Say the Right Thing 227

part, the very term ‘creole’ compels a distinct classiWcation of language that,
structurally, may be no diVerent from a language like English, which has
continuously borrowed forms and words throughout its history.97
While linguistics today generally takes a more sympathetic and expansive
approach to interlanguages, vestiges remain of the anxiety associated with
the change and variation they demonstrate. Such vestiges appear in argu-
ments that the grammar of pidgins and the language of children reproduce
the syntactic simplicity of a proto-language used prior to the cultural and
linguistic expansions of humans somewhere around 100,000 bc.98 They are
even more apparent in the suspicion surrounding whether to label, say, Kriol,
spoken in Belize, as an independent language or a variety of English. ‘Sim-
plicity is confused with inadequacy,’ suggests Loreto Todd, ‘and creole
speakers are often judged to be speaking bad English rather than eYcient
creole.’99
The most general diYculty with the anxieties generated, one way or
another, by the creation of new varieties and interlanguages is that the
categories that give rise to them, like the categories of change and variation,
can themselves be impressionistic value judgements. Interlanguages, new
varieties, debased varieties, and varieties that aren’t really varieties but rather
collections of forms that reXect inadequate acquisition easily blend into one
another, and this is so because any such distinctions rest as much (or more)
on interpretive frameworks as on any structural phenomena. While nine-
teenth-century critics saw transplanted varieties of English as at best imper-
fect and at worst barbarous, more recent critics have exercised greater subtlety
by working to deWne individual varieties’ grammars, histories, and relation
to their speakers’ cultures. Of the South Asian English once ridiculed by
Wright, for example, Braj Kachru observes that the variety can ‘be character-
ized both in terms of its linguistic characteristics and in terms of its contextual
and pragmatic functions’.100
Susan Butler provides one framework for categorical distinctions among
new varieties and interlanguages by specifying Wve criteria that deWne a
distinctive variety: a pronunciation standard, communal words and phrases
to describe speciWc circumstances where the language is used, a self-conscious

97 On the structural similarities between natural languages and creoles, as conventionally deWned,
see Sarah Grey Thomason and Terrence Kaufman, Language Contact, Creolization, and Genetic
Linguistics (Berkeley, CA: University of California Press, 1988), and Mufwene, The Ecology of Language
Evolution (Cambridge: Cambridge Univeristy Press, 2001).
98 Derek Bickerton, Language and Human Behavior (Seattle, WA: University of Washington
Press, 1995).
99 Todd, Modern Englishes: Pidgins and Creoles (Oxford: Blackwell, 1984), 248.
100 Kachru, Asian Englishes: Beyond the Canon (Hong Kong: Hong Kong University Press, 2005), 43.
228 Language Anxiety

awareness of the variety’s history, a literature, and reference works.101 By these

criteria, Philippine English, African American Vernacular English, South
Asian English, Chicano English, and so forth do indeed qualify as distinctive
varieties of English. And in this vein it’s worth noting that one of the most
prominent features of recent debates over the validity of world Englishes
has been the emergence of national dictionaries, which provide the codiWca-
tion that is necessary for any standard. A Jamaican English dictionary, thus,
appeared in 1967, an Australian English dictionary in 1988, a South African
one in 1996, and a New Zealand one in 1997.102
Randolph Quirk has provided a wholly diVerent framework for interpret-
ing the variation that informs the English used in many such communities.
Challenging the judgement of what he regards as imperfectly learned English
to be a variety tantamount to long-recognized regional and social varieties
like Canadian English or United States English, he suggests that to use
variation in this way, to legitimate social groups, is to practice what he archly
labels ‘liberation linguistics’.103 Focusing on the social consequences of vari-
ous linguistic forms and what he sees as linguists’ ethical obligations, Quirk
elsewhere suggests:
we need to ask ourselves who beneWts if we encourage the institutionalizing as norms
of certain types of language activity that could alternatively be seen as levels of
achievement. It may temporarily comfort an individual to be told that his English is
a communicatively adequate basilect . . . it may comfort a Ministry of Education in
requiring less funding than a more ambitious language-teaching programme; it may,
above all, seem comfortingly democratic. But will it serve the individual’s own needs

101 Butler, ‘Corpus of English in Southeast Asia: Implications for a Regional Dictionary’, in M. L. S.
Bautista (ed.), English Is an Asian Language: The Philippine Context: Proceedings of the Conference Held
in Manila on August 2–3, 1996 (Sydney: Macquarie Library, 1997), 103–24.
102 F. G. Cassidy and R. B. Le Page (eds.), Dictionary of Jamaican English (Cambridge: Cambridge
University Press, 1967); W. S. Ramson (ed.), The Australian National Dictionary: A Dictionary of
Australianisms on Historical Principles (Melbourne: Oxford University Press, 1988); Penny Silva
(ed.), A Dictionary of South African English on Historical Principles (Oxford: Oxford University
Press, 1996); H. W. Orsman (ed.), The Dictionary of New Zealand English: A Dictionary of New
Zealandisms on Historical Principles (Auckland: Oxford University Press, 1997). Representative studies
of newly recognized varieties are: Carmen Fought, Chicano English in Context (New York: Palgrave,
2003); Lisa Kim (ed.), Singapore English: A Grammatical Description (Amsterdam: John Benjamins,
2004); Braj B. Kachru, ‘English as an Asian Language’, in Bautista (ed.), English Is an Asian Language,
1–23; Andrew B. Gonzalez, ‘When Does an Error Become a Distinctive Feature of Philippine English?’,
in R. B. Noss (ed.), Varieties of English in Southeast Asia (Singapore: Singapore University Press, 1983),
150–72; and Bolton, Chinese Englishes.
103 Quirk, ‘Language Varieties and Standard Language’, English Today, 21 (1990), 3–10. Quirk’s
remark might be juxtaposed with David Crystal’s lament: ‘There has been little perception of the
need for a ‘‘green linguistics’’ ’ (Language Death [Cambridge: Cambridge University Press, 2000], 32).
A useful overview of the continuing argument about the status of putatively new varieties of English
can be found in Bolton, Chinese Englishes, 1–49.
 Say the Right Thing 229

when he or she looks for a better job? Will it help the Ministers of Education when
they look for a bigger supply of professionals? Will it serve democracy’s goal of the
individual’s mobility within a coherent free society?104

As Quirk implies, it’s certainly the case that recognition of indigenous var-
ieties often has a political edge, as in the argument that Black South African
English should become a kind of national language for the unity it would
provide its speakers.105 It’s also of course the case, as my earlier discussion
demonstrated, that Standard English has its own political edge.
The thing about competing frameworks for approaching transplanted
varieties and interlanguages is that their diVerences are unresolvable. Philip-
pine English qualiWes as a variety of English only if one accepts Butler’s
interpretive categories, but the fact that it Wts her categories cannot validate
the conclusion that any collection of forms that does so qualiWes as a variety,
unless one has already accepted her overall model. Similarly, the history of a
variety can be determinative only for someone to whom that category is
already signiWcant, and it isn’t necessarily so to Quirk. Indeed, Quirk’s
analysis, rooted in an ethical framework rather than a structural or social
one, would reject the concept of Philippine English, and however consistent
this rejection is with his framework, it likewise cannot prove, any more
than Butler’s acceptance of Philippine English can prove, the transcendent
accuracy of his interpretation.
The ethical orientation that motivates Quirk is itself fraught. For him, this
orientation assigns linguists a moral responsibility to resist movements like
‘liberation linguistics’ when their ultimate impact will be the further social
marginalization of particular speakers: ‘And if the native varieties go their
own way to a greater degree than seems at present likely, linguists may be
called on to suggest some interference with nature as radical as crop biologists
or livestock breeders have long taken for granted.’106 But if in Quirk’s view
linguists’ ethical obligation is to intervene in and manage the development of
natural languages, Susan Dicker reXects the views of a good many other
linguists when she adopts this same moral posture to argue that linguists
need to labor to undermine any social attitudes or institutions that might
sustain the very sociolinguistic positions that Quirk takes as givens. In
response to the assertion that particular varieties or discourse practices can
limit social and economic opportunities in the United States, she advocates

104 Quirk, ‘Language Variety: Nature and Art’, in Noss (ed.), Varieties of English in Southeast Asia, 13.
105 See Peter Titlestad, ‘English, the Constitution and South Africa’s Language Future’, in Vivian de
Klerk (ed.), Focus on South Africa (Amsterdam: John Benjamins, 1996), 163–73.
106 Quirk, ‘Language Variety: Nature and Art’, 15.
230 Language Anxiety

neither more education in Standard English nor even broader sociolinguistic

tolerance but rather an aggressive subversion of the cultural values and
institutions that Standard United States English supports, with linguists,
educators, and government oYcials actively promoting nationwide bilingual-
ism. Rather than manage natural languages, linguists for Dicker must
manage nature itself, enacting what Quirk might well have called ‘guerrilla
linguistics’.107
What Quirk and Dicker share—and share with Butler as well—is an anxiety
and even alarm associated with the processes of language variation and
change. For Quirk, it is tolerance of linguistic change that causes concern,
while for Dicker and Butler change and variation serve as vehicles for socially
disruptive transformation. All three critics seek to shape the meaning of
change, and in so doing all three elide the very distinction between variation
and its social signiWcance that their analyses foreground so clearly.
I assemble these competing views not in order to suggest that the classiWca-
tion of a set of forms as corruptions, a social or regional variety, or an
interlanguage is impossible to make or of no consequence. Quite the contrary.
All speakers, and not just linguists, make such judgements all the time, and
terriWc consequences sometimes follow from them. In the United States
perhaps one of the most prominent recent examples of this very point is the
controversy surrounding a 1996 resolution by the Oakland, California, School
Board. Confronting persistently low standardized test scores and graduation
rates among primary and secondary students—who were and are predomin-
antly low-income African Americans—and presuming to follow in the spirit
of a 1979 Supreme Court decision that had rendered it incumbent on school
districts to take into account diVerences between Standard United States
English and African American Vernacular English as a home language,
the Oakland School Board declared the latter (which it labeled Ebonics) a
‘genetically distinct’ language. The implications and objectives of this declar-
ation were never entirely clear, but what is clear is that within a matter of
days it became national news—the subject of critical commentary, vitriolic
editorials, academic debate, and racist jokes.
And what drove the response was not the fact of a collection of linguistic
forms and practices that the Board called Ebonics—everyone seemed to know
what was being talked about—but rather the challenge of categorizing
just what this fact was and is. Was Ebonics a language? Was it a sociolect?
Was it, to use Quirk’s terms, the institutionalization of an inadequate level of
achievement? Was the declaration itself a serious linguistic judgement, a

107 Dicker, Languages in America: A Pluralist View (Philadelphia, PA: Multilingual Matters, 1996).
 Say the Right Thing 231

largely empathetic response to a social situation, or a cynical attempt to

manipulate public sentiment and attract federal funds? And above all, what
would follow—pedagogically, linguistically, and socially—from the various
answers to these questions? On the surface, this response, conducted mostly
by individuals who lived far from Oakland and had no children attending
school in the district, argued over a categorical judgement about what forms
and varieties could be included in the history of English. But at its heart,
reception of the Oakland decision, like responses to all of the transplanted
varieties I have just considered, argued about issues of race, class, education,
citizenship, taxation, and social and individual responsibility. Like questions
over the status of Scots or Kriol—whether varieties of English or languages
unto themselves—an example such as this points to the way large social issues
of self-identiWcation can be triggered by variation and change and played out
through them. And this is why the classiWcation of linguistic varieties is
important.
The Oakland decision on Ebonics, later reissued in a slightly modiWed
version, led to months of political commentary and caustic remarks in the
popular as well as academic presses, the impact of which, a decade later, has
been minimal on Oakland students and racial attitudes. Neither, I suspect, has
the controversy had much impact on popular conceptions of change and
variation. The very lack of a clear resolution suggests that in Oakland, as in
other contexts, a great deal in fact depends on the interpretation speakers
assign to change and variation in their language. It also suggests that unam-
biguous, non-controversial assignments are not easily made. At an abstract
level, it may well be possible to draw neat distinctions between varieties,
interlanguages, corruptions, and the social issues they mediate, but in the
messy reality of language use, where most speakers live, this is not the case.108

Beyond language
Metalinguistic discussion of language has been nearly as changeable as English
itself. And it has changed in ways that not only reXect social context but also
conWgure language as supportive of that context. What St Augustine saw as
a Xawed but crucial means of communication, its vocabulary crafted by
ancients so as to reveal the immanent, transcendent truths of the universe,
Dante saw as an art, something these same ancients devised in order to
transcend the immutable truth of change and preserve communication. To

108 For a measured assessment of the decision and its aftermath, see Theresa Perry and Lisa Delpit
(eds.), The Real Ebonics Debate: Power, Language, and the Education of African-American Children
(Boston, MA: Beacon Press, 1998).
232 Language Anxiety

the early modern period in England in particular, grammar was the measure
of English’s inadequacies. Without a formal grammar—and, ultimately, a
standard variety—English could not achieve stability and thereby attain
the communicative achievements of Greek and Latin. Without these, neither
could it conWrm its (or England’s) pre-eminence in the world. This same
grammar, when invested in the seventeenth and eighteenth centuries with
social meanings about class and status, has come to serve as a measure of
the very social categories it also helped maintain in the cultivation of written
and spoken standards.
Securely on this chronological side of this semantic development, George
Sampson, in his widely read 1921 book English for the English, could see English
and the study of its language and literature as moral forces that would give life,
rather than merely a living, to working-class children and would preserve
humanity and culture in the aftermath of nineteenth-century industrializa-
tion. Elementary education, he felt, was the most important level of education,
and the cultivation of correct grammar and the study of English literature were
its most important features: ‘But, as we have said, English is really not a subject
at all. It is a condition of existence rather than a subject of instruction. It is an
inescapable circumstance of life, and concerns every English-speaking person
from the cradle to the grave. The lesson in English is not merely one occasion
for the inculcation of knowledge; it is part of the child’s initiation into the
life of man.’109 This same period witnessed the foundation of the Society of
Pure English, championed by Bridges and dedicated to preserving grammat-
ical distinctions and protecting English from changes initiated by speakers
both at home and abroad. By the 1975 Bullock report on the teaching of
English in the United Kingdom, metalinguistic emphasis had shifted from
the redemptive power of English to the importance of classroom methods and
the encouragement of pedagogical and social acceptance of non-standard
varieties. A decade later the Swann Report acknowledged the importance of
ethnicity and the value of maintaining non-English languages at home, but
opposed bilingual education on the grounds that instruction in languages
other than English was not the business of schools. And this testament to
the importance of preserving a particular variety of English appeared even
as other metalinguistic discussion has championed the diversity of the lan-
guage in the ways I have described in Chapter 4.110

109 Sampson, English for the English: A Chapter on National Education (1921; rpt. with introduction
by Deng Thompson, Cambridge: Cambridge University Press, 1970), 44.
110 On recent educational policy in the United Kingdom, see Linda Thompson, Michael Fleming,
and Michael Bryam, ‘Languages and Language Policy in Britain’, in Michael Herriman and Barbara
Burnaby (eds.), Language Policies in English-Dominant Countries: Six Case Studies (Clevedon: Multi-
lingual Matters, 1996), 99–121. On the contrary embrace of variation, see David Crystal, The Stories of
English (Woodstock, NY: Overlook Press, 2004).
 Say the Right Thing 233

One constant in these discussions has been the channeling of non-linguistic

issues through language, whether in St Augustine’s approach to language as
evidence of divine design or through contemporary pedagogical discussions
of the role of Standard English in furthering or delaying social integration and
advancement. Variation and change in particular have consistently Wgured in
this regard. And they have done so by being both evidence for non-linguistic
social arguments and the means to conduct the arguments themselves. If
linguistic variability motivated the record that dictionaries like Webster’s
Third provide, serving in eVect as their driving force, it has also provided
the counterpoints that make a variety like Standard English possible—the
competing narrative, to reprise Hayden White’s terminology. Because of this,
the metalinguistic framing of variation and change can be pre-determinative
of sociolinguistic experience. I have already suggested how ‘creole’ as a label
draws a structural distinction among what are often accidents of history.
Within the ideology of standard language, similarly, certain linguistic
data—forms, pronunciations, words, and usages—become non-standard be-
fore they can be mere variation, and they accordingly will carry with them
whatever social meanings have already been invested in them. ‘Linguistic
stereotypes’, Bailey remarks, ‘eventually emerge with opprobius social evalu-
ations attached to them; once a given feature is associated with that stereo-
type, it joins the value system already established for it . . . the image, in short,
anticipates the evidence.’111 And so the evidence of sociolinguistic experi-
ence—the vulgarity of non-standard forms, varieties, and speakers—almost
inevitably conWrms the interpretations that speakers inherit.
In this chapter I have argued for ways in which metalinguistic commentary,
like literary narratives and public policy, can be determinative in speakers’
attitudes towards change and variation. The picture that emerges here is far
from clear and oVers no easy conclusions. As I noted in the opening chapter,
I certainly do not claim that language change is the only phenomenon
to mediate extra-linguistic anxiety or that mediating such anxieties is the
only thing that language does. In fact, the metalinguistic discussions I have
traced contain a good many contradictions. Standard language, as I’ve just
observed, requires non-standard language to be deWned and maintained as a
distinct variety. British English requires American, Australian, and South
African English to preserve its own identity. And colonialism and inter-
national activity produced transplanted varieties of English that have been
at once conWrmation of Anglo-American global achievement and examples of
linguistic decline.

111 Bailey, Images of English, 134.

234 Language Anxiety

Metalinguistic commentary on change and variation has produced its own

contradictions, perhaps the most prominent of which has been the interpreta-
tion of print and standard language as both socially liberating and conWning.
Indeed, already in the eighteenth century Sheridan (the father of the play-
wright) envisioned print as a means for eliminating the linguistic and social
stratiWcation that Standard English fosters. By advancing literacy through a
standard language, the press, for Sheridan, would be the means towards
an egalitarian society.112 Volumes like the 1880 The Letter H Past, Present,
and Future oVered a similar promise, that by eliminating stigmatized, variant
pronunciations, speakers might exceed restrictions of class, employment, and
education.113 What a book like The Letter H also does, of course, is reinforce
the social stigmatization of ‘dropped h’ not only among those speakers whose
varieties don’t drop it but also among those whose varieties do—and who
thereby become complicit in the subordination of their own speech. And what
it further does is hold out what I take to be a false promise: that by changing
language alone, individuals can also change their social circumstances.
Democracy’s goal, as Quirk implies, may in fact be ‘the individual’s mobility
within a coherent free society’, but whether most adult individuals, with the
aid of The Letter H or Jones’s Pronouncing Dictionary, truly could or can
change their accents is doubtful. And still more doubtful is whether such
change, if it were possible, could also eVect a change in an individual’s social
status and opportunity. For every Margaret Thatcher who is able to reinvent
herself through dialect transformation, the examples of Dickens’s Wction and
fabricated Maori English suggest that there are many more individuals whose
social standing will precede conceptions of whatever actually comes out of
their mouths.
Emerging foremost from these contradictions is the fact that, whatever the
nuances of any particular metalinguistic criticism, it tends to present vari-
ation and change as volatile issues whose most important implications are
extra-linguistic. For Cameron, perhaps the crux of the verbal hygiene issue is a
conXict over the possibility—or impossibility—of achieving perfect commu-
nication through value-free language and thereby preserving a stable repre-
sentation of the world. ‘The common language needed for public purposes’,
she observes, ‘is portrayed above all as a neutral and universal language,
one that is available to all parties equally and does not predetermine the
outcome of their discussions.’114 When speakers manipulate traditional,

112 Sheridan, A Course of Lectures on Elocution, 256–62.

113 Alfred Leach, The Letter H Past, Present, and Future: A Treatise, with Rules for the Silent H Based
on Modern Usage, and Notes on Wh (London: GriYth and Farran, 1880).
114 Cameron, Verbal Hygiene, 120.
 Say the Right Thing 235

standard language by avoiding the generic masculine pronoun, they certainly

expose the partiality of their own language, but they also reveal the partiality
of the putatively neutral standard. In these circumstances, speakers must
make choices over which partiality they prefer, and such choices have poten-
tially explosive powers of social reform and reaction.
The volatility of precisely this need to make a choice appears in more
general metalinguistic debates over the meaning of variation and change.
To critics like Tony Crowley and Robert Phillipson, linguistic variability has
transformative power. It must be maintained, because if it is erased, through
the spread of a standard or global English, then linguistic regularity becomes a
means of repression and social domination.115 John Honey shares Crowley’s
and Phillipson’s sense of language’s political consequences, though he sees this
same variability as change that must be avoided, for it brings with it
the repression of those denied access to Standard English. If Crowley and
Phillipson champion non-standards as forms of self-expression and social
validation, Honey judges that failing to instruct speakers in Standard English
renders them, eVectively, as wrong speakers subject to the attendant social
consequences: ‘causing children to learn standard English is an act of em-
powerment which will give them access to a whole world of knowledge and
to an assurance of greater authority in their dealings with the world outside
their own homes, in a way which is genuinely liberating’.116 It is in this sense
that Quirk has challenged the practice of what he sees as a kind of well-
meaning but misguided ‘liberation linguistics’. But there are two crucial
points for me in all these views, whatever their individual diVerences: Wrst,
variation and change serve as Xashpoints for extra-linguistic political con-
cerns; and second, they do so in ways that the ideology of a standard renders
natural and inherent in language—maybe even invisible—rather than con-
structed and situational.
Dr Johnson once observed, ‘I am not answerable for all the words in my
Dictionary’.117 The furor over Webster’s Third and the tensions fostered by
metalinguistic discussions of English in general suggest something else.
Not only have lexicographers like Johnson (and Gove) been answerable for
their words, but their words have been answerable for cultural achievements
far beyond language alone. While Dr Johnson’s disavowal may be character-
istically arch, it also points to a signiWcant contradiction. It is true enough

115 Crowley, The Politics of Discourse: The Standard Language Question in British Cultural Debates
(London: Macmillan, 1989), and Robert Phillipson, Linguistic Imperialism (Oxford: Oxford University
Press, 1992).
116 Honey, Language is Power, 42.
117 Quoted in Reddick, The Making of Johnson’s Dictionary, 10–11.
236 Language Anxiety

that individuals are generally not the creators of the words and grammar they
use or even the conceptions of language that frame them. We learn language
through observation and use, and much of what we say and how we speak
therefore goes unexamined. At the same time, there would seem to be limits
to the responsibilities of linguistic systems in comparison to those of their
speakers. If we suppose that Webster’s Third somehow displays linguistic
permissiveness—which is a supposition that depends on the prior suppos-
ition that such permissiveness exists—there remains another signiWcant men-
tal leap from this permissiveness to any consequences it might have for
personal virtue and child-rearing. In precisely what way, one might ask,
does the refusal to designate certain words as slang produce a permissive
society? Given the fact that Anglophones typically consult English dictionaries
for only meaning and orthography and not for moral or social purposes,118
what evidence is there that such consultation will likewise inXuence personal
much less cultural attitudes? And if the OED was meant as ‘proof of an
advanced and advancing civilization’, is there reason to believe that the
dictionary itself contributed to an advance of that civilization’s inhabitants?
Any logic that regards answers to these questions as a self-evident ‘plenty’ is
merely associational, I think. It is an assertion by Wat of cause-and-eVect
connections. Even for education and usage, areas most directly aVected by
what’s in a dictionary, connections between the putative permissiveness of
Webster’s Third (or any other quality of any other dictionary) and extra-
linguistic social activity are only and fundamentally intuitive. Or, perhaps,
counter-intuitive, since for every pedant like Follett or Macdonald there must
be thousands more Anglophones who merely speak, irrespective of what
words a dictionary includes or how it labels and deWnes them. Most people,
I suspect, rarely consult dictionaries at all.
Speakers have a good many choices in the language they use, including
whether or not they wish to leave it unexamined. The ideology of a standard is
itself the product of choice in historical circumstances and not an intrinsic
quality of natural language, and the success of its contributions to the English
language and Anglophone culture similarly reXects historically conditioned
and therefore not inevitable choices. In other words, the eVects of standard
language depend on the ways in which speakers use the variety, for in itself a
standard is neither inherently liberating nor inherently conWning. Variation
and change, however, are inherent in natural language, and they necessarily
do occur in any speech community. By extension, the metalinguistic sense we
make of them likewise reXects choices about society, its values, and the role of

118 Béjoint, Tradition and Innovation in Modern English Dictionaries, 140–68.

 Say the Right Thing 237

language in it. And whether we follow Honey and decide standard language is
liberating or follow Crowley and decide it is repressive, we will invest variation
and change with the sociolinguistic signiWcance that supports our decisions.
But such decisions remain choices for which we as individual speakers are
answerable. It is to this issue that I devote the concluding chapter.
 6

Fixing English

When English is the deWnition

One of the courses I regularly teach is the History of the English Language.
This is a curricular staple in many universities that oVer advanced degrees in
English language and literature, and for its historical breadth and the range of
language-related issues that it raises, it’s also one of my favorites. Inevitably,
history of the language courses are structured chronologically, beginning
with the Indo-European language and people of around 5,000 bc, moving
through the Primitive Germanic grouping of about 2,000 bc to the Western
Germanic grouping of about 500 bc, and proceeding from there to the
beginning of Old English.
When exactly this beginning occurred presents one of the Wrst challenges in
the course, perhaps because by this historical point an actual written record
(as opposed to hypothetical reconstructions) begins to appear, requiring
greater care with empirical argument, and because the history also becomes
sociolinguistically close enough to have consequences for modern speakers
and how they view themselves, their language, and their world. Does English
begin the moment speakers of varieties of West Germanic set foot in what
would become Great Britain—is its origin in fourth-, Wfth-, and sixth-century
social and political action? Is its origin purely linguistic—does it begin later,
in perhaps the eighth century, when these transplanted British varieties of
West Germanic might well have become mutually unintelligible with the
Old Frisian and Old Saxon languages that remained on the Continent and
that otherwise are the Germanic varieties structurally most similar to Old
English? Does Old English begin in the seventh century, when the written
record begins, in which case its origins are material? Is the origin still later,
in the late ninth century, when an Anglo-Saxon nation-state begins to
emerge from the political and cultural achievements of Alfred the Great?
Or, in view of the fact that a Channel crossing alone would likely have little
impact on linguistic structure, does English really begin on the Continent?
 Fixing English 239

And if so, at what point—when the West Germanic variety diverged suY-
ciently from the Germanic variety, when that variety diverged suYciently
from a larger grouping of generally Western Indo-European languages desig-
nated the Centum grouping, or when that grouping diverged suYciently from
Indo-European?
Textbooks for a course like this—and many are available—typically mimic
this chronological progression, oVering an introductory chapter or two on
continental matters and then proceeding to English and its chronological
development.1 Such books perhaps inevitably present a teleological narrative
line, for they cannot be histories if they do not identify beginnings, develop-
ments, and, perhaps most importantly, conclusions, and to do this means to
identify what linguistic and social facts qualify for inclusion within the history
and what merit exclusion. Relations between facts and their representation
have loomed large in historical and literary discussions of the past quarter
century, but as long ago as 1946 R. G. Collingwood deWned the issues with
a clarity that I think transcends anything written since. ‘History is not a
spectacle,’ said Collingwood simply. ‘The events of history do not ‘‘pass
in review’’ before the historian. They have Wnished happening before he
begins thinking about them. He has to re-create them inside his own mind,
re-enacting for himself so much of the experience of the men who took part in
them as he wishes to understand.’2 In its selection of events and larger
deWnition of its own concerns, historical writing is thus (as Hayden White
later noted) an act of imagination and judgement—one that preWgures what’s
important and what’s not—and as such it must conceptualize and validate,
in addition to the past, some version of the present towards which the history
inexorably advances.
What all this means for the history of English and its textbooks is that the
decision about where to begin Old English is not peculiar or unique but
emblematic of every decision about structure, variation, and change that a
linguistic historian must make. In each case, in the identiWcation of what
constitutes the language, what constitutes its variants, and what constitutes
plain error, English is not so much a simple event that passes in review before
the historian but a preWguration of the Weld, its relevance to the present, and
the relevance of both to the future. As in David Lightfoot’s inventive simile,
‘a language, like Odysseus, turns out to be a mythical, imaginary creature.
It may be a convenient and useful Wction, like Odysseus and the setting of the

1 One exception is Barbara Strang’s A History of English (London: Methuen, 1970), which begins
with the present and regresses in time.
2 Collingwood, The Idea of History (Oxford: Clarendon Press, 1946), 97.
240 Language Anxiety

sun, but in reality it is a derivative, the aggregate output of some set of

grammars. We shall see that language is not a coherent, deWnable entity.’3
While the history of English necessarily rests on empirical data, that history’s
claim that the language is deWned by the acceptance of certain forms, vari-
ation, and change and the disavowal of others is only apparently or partially
empirical. The history of English, I would suggest, is less a topic than an
argument—or a value judgement, as I called it in Chapter 2—and part of
what dictionaries do is provide conWrmation for this argument. As an argu-
ment, moreover, this history is ultimately circular. It presumes some trans-
historical identity for English, whatever alterations the language might
experience, as a means for distinguishing English forms from non-English
ones; and it then uses these same distinctions to deWne the language. Rather
than a topic rooted in the empirical data of linguistic structure, indeed, the
history of English is an argument serving shifting social expectations.
What all this means for linguistic change and variation is that they play
crucial roles in the deWnition of English. As I also argued in Chapter 2, for the
most part change and variation are regular and constant, if not always entirely
rational—they proceed systematically in patterns that are sometimes predict-
able and often explicable retrospectively. Variation is so much a part of
language, in fact, as almost to challenge the claim for any stable category
like ‘English’, replacing it, à la Lightfoot, with an amorphous, continually
shifting and reconWguring structure. Or, as in Roger Lass’s memorable phrase,
a ‘language is a population of variants moving through time, and subject to
selection’.4 It is for this very reason that emergent linguistics has increasingly
come to see speakers’ articulation of variation as the quality that produces
and constitutes natural language: to vary is to speak. And this means that
alongside the axiom of linguistic relativism—the notion that because all
languages communicate what their speakers need them to communicate
there can be no better or worse languages—stands another axiom that’s
been suggested by others and that I have teased out throughout this book:
change and variation are structurally neutral. They simply and continuously
happen. What I’ve also teased out, however, is the notion that if change occurs
neutrally within often predictable patterns, its direction and, more import-
antly, its meanings are shaped by extra-linguistic factors. Here I oVer a
classroom metaphor.

3 Lightfoot, The Development of Language: Acquisition, Change, and Evolution (Malden, MA:
Blackwell, 1999), 77.
4 Lass, Historical Linguistics and Language Change (Cambridge: Cambridge University Press,
1997), 377.
 Fixing English 241

At any one synchronic moment, there is an aggregate of grammatical and

lexical tokens uttered by those speakers who have some claim to be speaking
English, whether that claim is historical (e.g., they live in primarily Anglo-
phone countries like Canada or New Zealand), socioeconomic (e.g., they
live in the United Kingdom or the United States), pedagogical (e.g., they are
students of English as a second language), or even pragmatic (e.g., they
are speaking to someone who also has a claim to be speaking English). And
for many of these tokens there are alternatives—variants—whether among
the speakers of one community or between those speakers and another
community’s. If one were to imagine a chalkboard or dry-erase board at the
front of the ubiquitous history of the language class, one could represent all of
the linguistically meaningful utterances produced by speakers with some
claim to be speaking English as dots on that board. English would then be
the subset of dots that are judged a part of and that constitute the language.
By extension, diachronic representation—for which a much larger board
would be necessary!—would begin with dots for all the utterances of anyone
anywhere at any time who had some claim to be speaking English and would
proceed to identiWcation of that subset of utterances transhistorically accept-
able as English. Given the ubiquity of change and variation, with possibly
hundreds of millions of dots representing hundreds of millions of utterances
on the board I’ve imagined, abstractions like the neogrammarians’ claim
that change admits no exceptions, or Saussure’s langue, or Chomsky’s ideal
speaker-listener play vital roles in our understanding of language, its vari-
ation, and its change. Without them, we could not proceed, for we cannot
examine every individual utterance by every individual English speaker of
the past 1,500 years.
To return to Collingwood, however, diachronic and synchronic subsets of
the utterances that qualify as English do not pass in review before the
grammarian or historian who draws them. They are not simple facts but
reXections, and justiWcations, of some pre-Wguring of the Weld of synchronic
and historical English linguistics that is itself, necessarily, based on some
heuristic frame. Someone on some principle, that is, decides which dots to
include and which to exclude, and in many ways decisions about such matters
were much easier to make in the past than they are now. When Murray began
editing the OED in 1879, cultural context allowed him to know what was and
wasn’t English, and even he, in the preface passage I quoted in Chapter 5,
allowed for some imprecision. Since then, with increased recognition of
global English dialects, the language’s identity has lost much more precision,
leaving lexicographers with far less certitude than Murray enjoyed. And when
the cost of a dictionary’s production is taken into account—as any modern
242 Language Anxiety

publishing house must do—yet another principle factors into lexicographical

selection. Such a nexus of linguistic, social, and economic principles has
indeed shaped the various corrections and revisions of the OED since its
initial completion in 1928.5 There’s nothing insidious about any of this—it’s
how human cognition makes sense of the world, by identifying and categor-
izing experience according to some principle. And to do this—to decide
whether varieties and grammatical forms go away, and if they do, where
they go—necessarily draws on perceptual as well as empirical data. Put very
simply, an individual’s sense of English rests as much on lived experience as
on linguistic structure.
At the same time, it is precisely this process that makes the history of English
an argument. To ask who makes this argument, or how it is made, is also to ask
perhaps the largest question I have implicitly asked throughout this book:
What is English? And this is not an idle question, since any claim to speak
English carries with it signiWcant cultural, socioeconomic, and political reper-
cussions. Any answer to this question requires some sort of structural baseline:
a grammar of the phonology, morphology, syntax, and lexis against which all
possible English utterances can be evaluated. Do they conform to this baseline,
do they vary from it but in ways that still qualify them as English, or is their
variance so distinctive that they can no longer be considered part of the
language? And these kinds of comparisons aVect assessment not only of
individual grammatical features, such as whether speciWc pronunciations,
words, or inXectional morphology qualify as English, but also of the concen-
tration of such features in what may or may not be considered a variety. Like
grammatical features, regional dialects, sociolects, pidgins, and creoles all
derive their identities through their distinctions from some putatively base
form of English. Without this kind of baseline, indeed, no meaningful gener-
alizations of any sort could be drawn about English: we would simply be
unable to talk about synchronic variation and diachronic change unless
we had some ideal against which to measure the variation and change.
But as conceptually necessary as such a baseline might be, it can also
occasion, in two distinct ways, anxiety about language, its variation, and its
change. First, whatever heuristic is used to identify the baseline, whatever
frame historians and grammarian use to deWne the linguistic present in their
own minds or re-enact linguistic history (to invoke Collingwood one Wnal
time), the end result necessarily advances a form of English used only by some
speakers. Concomitantly, it advances the social and economic opportunities

5 See Charlotte Brewer, Treasure-House of the Language: The Living OED (New Haven, CT: Yale
University Press, 2007).
 Fixing English 243

of those speakers over those of speakers who use another variety. Given
my argument that no linguistic phenomena have meaning in and of them-
selves, it is always speakers who invest signiWcance in them, deciding how to
classify them, which are important, and what they express. And the truth of
the matter is that in assigning such meaning, some speakers, to paraphrase
George Orwell’s Animal Farm, are more equal than others. If by birth I should
happen to speak a variety that’s particularly close to this baseline—and I do—
I will have less occasion to worry over or attend to the grammatical forms I use
than would someone whose variety qualiWes as more peripheral English or
even a creole. Change and variation might occasion anxiety for both of us but
for diVerent reasons: for me, over the investment I might have in maintaining
the sociolinguistic distinction and priority of my own variety, which would in
fact leave me in the more inXuential position to decide whether the other
speaker’s variety is English, a creole, or an undeveloped interlanguage; and
for the other speaker, over the sociolinguistic diVerences implicated in the
variation between our speech patterns. General issues like these undermine
the speciWc assumption that somewhere there is a stable, ‘real’ English pro-
viding stable comparative detail for assessing other varieties. Where and how
would the identity of this ‘real’ English be drawn today? And for that matter,
since English has changed so much, where and how might it be drawn
historically? Is ‘real’ English what was used prior to the loss of grammatical
gender and the dual pronoun? Or after the Great Vowel Shift but before
the inXux of Latinate words? Or after the conclusion of the Second World
War but before the appearance of transplanted varieties in Asia and the
introduction of Spanish lexical items in the United States?
The previous chapter argued that as it has been written since the early
modern period, the history of English as a conceptual category has been
implicated in the development of standard English, and vice versa. Not only
did the cultivation of both disciplines occur simultaneously in the sixteenth,
seventeenth, and eighteenth centuries, but both also identiWed the same
baseline in a learned variety of English concentrated in southeast England.
It was this variety that formed the basis of standard spoken and written
English, just as it was the development of Standard English that has served
as a kind of master narrative in histories of English. English colonizing and
merchandising eVorts provided two bulwarks for this narrative: they
depended heavily on literacy and Standard English in educational, mission-
ary, and governmental activities, and they largely originated in the southeast
of England, so that when English spread around the globe, a generally south-
east variety became a literal model in the utterances of inXuential personages
in the Raj, early America, and other colonial communities. To an extent,
244 Language Anxiety

indeed, all non-British varieties of English are ultimately dialectal develop-

ments of a southeast variety, since even when transplanted British dialects
co-existed (as in the United States) this one typically proved the most
inXuential.6 If American English eventually became an alternate baseline, it
did so only after early British speakers repeatedly faulted it for failing to meet
expressed standards and only as a baseline whose claim to authenticity rested
on an entirely diVerent claim: not historical, literary, or grammatical but
socioeconomic and political. In this way, of course, the anxiety occasioned
by divergence from some imagined baseline of English is neither countered
nor neutralized; it is instead accepted as the necessary byproduct of institu-
tionalized variation, which now exists between two competing standards
as well among the varieties measured against those standards.
This institutionalized anxiety has appeared again more broadly in the
proliferation of English and its speakers over the past few centuries, which
replays the nervous tension expressed in eighteenth-century British responses
to the development of United States and New Zealand English. Whether
through calls to contain the diversiWcation of English or through arguments
to acknowledge world Englishes, variation and change from a standard,
putatively Wxed form of the language serve as Xashpoints for linguistic and
political discussion. Histories of English have themselves contributed to the
anxiety I describe. Underwritten by a standard ideology, they still often
describe a process of decreasing synchronic variation since the Middle Ages,
leading to the early modern triumph of English, the emergence of standard
language, and culmination in some modern form like United States English.
And this historical spectacle is imagined despite a backdrop of a conserva-
tively estimated one billion contemporary speakers of English as a Wrst or
second language, in a plethora of indigenous, post-colonial, and what Braj
Kachru calls expanding-circle domains. All this means that contrary to the
narrative teleology of histories of the language and the structural inadequacies
that comparison with a standard can imply, contemporary English embodies
structured synchronic variation (i.e., sociolects and regional dialects) that
must well exceed that of any point in the language’s history. English, I would
say, is more alive than it has ever been.
The other way by which a comparative grammatical baseline can cultivate
anxiety is the way this book has examined in detail: by the channeling of
educational, social, religious, and governmental issues through variation and

6 See Roger Lass, ‘Where Do Extraterrestrial Englishes Come from? Dialect, Input, and RecodiWca-
tion in Transported Englishes’, in Sylvia Adamson et al. (eds.), Papers from the 5th International
Conference in English Historical Linguistics (Amsterdam: John Benjamins, 1990), 245–80.
 Fixing English 245

change. Alfred the Great’s discussion of the decay of learning after ninth-
century Viking raids, the late-medieval Lollard controversy, the cultivation
and codiWcation of standard written English, the suppression of indigenous
languages in New Zealand and Hawaii, attempts to deWne a nation through
English, arguments over whether the varieties spoken in post-colonial regions
qualify as national varieties—all these sociolinguistic episodes displace anx-
iety over social concerns to linguistic change and variation. They channel
concerns over economics, race, ethnicity, sex, and class into language issues
that can then be managed through both corpus planning of English usage
and status planning of the relative roles of varieties of English as well as of
the role of English in relation to contact languages around the globe. Within
a heuristic emblemized by the biblical account of Babel and articulated
in narratives from the earliest days of English, Anglophone policies and
institutions have designated the linguistically natural (change and variation)
as unnatural and debilitating, and have therefore invested in it terriWc social
consequence. Whether from the reactionary perspective of St Augustine, John
Hart, or the United States English Only movement, or from the skeptical
and conspiratorial one of those who practice what Quirk labels liberation
linguistics, change and variation serve, weirdly, as at once euphemisms and
Xashpoints. They stand in for social issues that are truly volatile and thereby
neutralize them; but in so doing they themselves also occasion anxiety about
the issues they represent. Blaming language and managing it through educa-
tional and governmental policy thus become almost instinctual responses to
any social, racial, or economic conXict. ‘The American people believe English
should be the oYcial language of the government,’ former United States
House Speaker Newt Gingrich accordingly once claimed, conXating language,
class, and race. ‘We should replace bilingual education with immersion
in English so people learn the common language of the country and they
learn the language of prosperity, not the language of living in a ghetto.’7
Given the disjunction between language variation and the meanings
that accrue to it, these meanings and the social concerns that exercise linguis-
tic liberationists as well as Gingrich can be eVectively addressed only by
separating them from these linguistic associations. In the Wrst instance,
what provides language with social meanings—and what allows it to serve
euphemistically for a variety of non-linguistic concerns—is the individual
and collective decision to identify some variation as natural and some as

7 ‘It’s English or Tongues of ‘‘Ghetto’’ Life, Gingrich Says’, Milwaukee Journal-Sentinel, 1 April
2007, 19A.
246 Language Anxiety

erroneous. In eVect, it is a decision to construct the history of English through

a non-linguistic argument by allowing the language’s putative identity
to underwrite political and social agenda. And so the way to change the
situation, I would maintain, itself begins with a decision: a decision not to
allow language to stand in for non-linguistic concerns but directly to address
these concerns, whether as individual citizens or as members of political
entities like cities, states, and countries.
To argue this much, and to act upon this argument, is not to embrace lax
standards in education, social chaos, or the gradual, inexorable decline of
both the art of using English well and the structural integrity of the language.
For Deborah Cameron, in fact, ‘The way to intervene in public debates like
the one about English grammar is not to deny the importance of standards
and values but to focus critically on the particular standards and values being
invoked and to propose alternatives’.8 However much synchronic variation
English today displays, moreover, there remains some kind of shared cultural
space embedded in and realized through English, and a proWtable line of
inquiry would examine this shared space and extrapolate from it both to
revised views of grammar and to distinctions between language and social
meaning.9 To argue as I have, then, is to accept responsibility for what we as
Anglophones do, and until now a signiWcant part of our actions has been to
respond to certain diYcult non-linguistic issues as if they were linguistic and
in the process avoid dealing with their full implications. The early modern
codiWcation of English shows that social stability cannot be maintained by
means of descriptive grammar; Wction’s use of non-standard varieties shows
that racism and sexism cannot be transformed through rhetoric; the linguistic
histories of New Zealand and the United States show that ethnic divisiveness
cannot be eradicated, nor social equality achieved, through the elimination
of indigenous and immigrant languages; and the persistence of a kind
of transmuted ethnicity in former Celtic-speaking areas shows that global
homogenization of culture through international business does not erase
cultural identity, even as desultory attempts to revive Celtic languages
suggest that the objectives of such business cannot be thwarted or advanced
merely by the identiWcation and protection of endangered languages. When
change and variation do not constitute the problem, they also cannot oVer
the solution.

8 Cameron, Verbal Hygiene (London: Routledge, 1995), 115.

9 See, for example, Anna Wierzbicka, English: Meaning and Culture (Oxford: Oxford University
Press, 2006).
 Fixing English 247

The cost and value of variation

Without doubt, as natural and inevitable as variation and change may be, they
impose a signiWcant cost on speakers, their communities, and even, in a
metaphorical sense, languages themselves. One way to measure this cost
would be by imagining the kinds of sociolinguistic beneWts that would be
lost if standard language, in the way it counteracts variation and change,
ceased to exist. In the absence of standard language, for example, print would
also be diYcult, for though in England it dates to the pre-standard Wfteenth
century, its emergence as the deWning medium for written language—and all
the social and institutional developments that emergence entailed—correlates
with Standard English. Developing together as they did, print and standard
language have provided a way to achieve a linguistic condition in which, even
if speakers can still misunderstand one another, written language does not in
fact change or at least changes relatively slowly. In turn, print and Standard
English have enabled the growth of education, the proliferation of literacy,
and the expansion of Anglophone politics and markets. They also make
it possible for geographically and chronologically distant speakers to com-
municate with one another—for South Africans to write to Canadians, and
for modern Australians to read what Jane Austen wrote. Print and standard
language may get us as close as we can be to the pre-Babel condition of being
‘of one language’.
A further measure of the cost of variation is that even with print and
Standard English, confusion still arises, for Anglophones can use the same
written and spoken language with such diversity as to thwart communication.
Trivial communicative disconnects can occur whenever speakers of diVerent
sociolects or regional varieties encounter one another when shopping or,
particularly, traveling. While these disconnects may be the source of literary
achievement or innocent amusement—local pronunciation of place names
seems a continual source of humor to all Anglophones—they can also
produce frustration, confusion, and anger. And this can be the case both for
an isolated traveler unable to understand directions or, like Caxton’s mer-
chant, to order a meal in a restaurant, as well as for social groups and nations
attempting to achieve workable, legal relations with one another. The 1840
Treaty of Waitangi in New Zealand created the social confusion it did partly
because of a disparity between the English and Maori versions, just as
Macaulay’s 1835 minute used the variation between the English of British
colonials and that of the indigenous people of India to codify social and
class distinctions, and just as Webster saw American dialect variation as the
248 Language Anxiety

means to and conWrmation of the claim of the United States to a cultural identity
distinct from that of England. At issue here, too, is identity and its investment in
the forms and varieties of language we use. To the extent that language does serve
as a symbol of group, ethnic, or national identity, its variation and change
necessarily can work against this identity by foregrounding and enabling the
emergence of various subgroups within a language community. If diYcult
issues like these arise against a backdrop of Standard English, which to some
extent limits linguistic diversiWcation, one could reasonably expect that without
a standard such diYculties would increase.
On a more literal level, the cost of variation and change can be seen in their
impact on social services, voting, education, and so forth. Funded by taxes,
such programs respond directly to the variation present in Anglophone
populations by providing, or withholding, money tied speciWcally to an
individual’s command of English. For native speakers, the primary investment
is a Wnancial one in programs designed to bring a non-standard speaker or one
who lacks command of standard written English to a given level of compe-
tency. However this level is identiWed, a proportionate relationship necessarily
exists among a speaker’s current competency, the level to which the speaker
will be brought, and the Wnancial commitment by a school or public oYce.
Very simply, it costs money to erase social and regional variation, and the
farther a speaker is from the standard, the more it will cost to erase that
distance—which might not in fact be erasable. For non-native speakers,
the question of investment may also be Wnancial, involving the cost of ESL
programs in relation to speakers’ disposable incomes. But non-native speakers’
investment can be more pointedly emotional and personal, turning on
issues like speakers’ initial command of English, the structural similarities
between their native language and English, the favorability (or otherwise) of
speakers’ sociolinguistic context to the acquisition of English, and the extent to
which their identity is interconnected with their native language.
If community Wnancial decisions about sociolinguistic issues like these
represent a choice about the relative use of tax dollars—what amount of
money should go to national defense and what to immigrant education?—
they also represent a choice about how a city or country wants to represent
itself. Does it see itself as hostile, inviting, or indiVerent to language variation
among its own Anglophones or between them and the non-native speakers in
their midst? And if it sees itself as inviting, as many cities and countries in fact
do, does it see the most sincere, beneWcial kind of invitation to non-native
speakers to be a funded entitlement program or the mere assertion that
they have the opportunity, like countless immigrants before them, to learn
English, assimilate, and advance in society? How does a city or country see the
 Fixing English 249

relationship between what beneWts the native Anglophones and what beneWts
the non-native immigrants? While it’s easy to demonize answers to questions
like these—and particularly those who give the answers—the questions
themselves resist easy solutions, and therein lies part of their cost. In the
summer of 2006 the mayor of the small town of Arcadia, Wisconsin (pop.
2,402), proposed making English the town’s oYcial language, requiring the
presence of an American Xag beside that of any other country, and mandating
that federal oYcials be notiWed about any undocumented immigrant workers.
Responding to a large inXux of Spanish-speaking laborers from Mexico
into the local dairy industry, he asserted, ‘I do want to close the doors to
those that would come there and drain our resources by living and working
here illegally, while beneWtting from our tax-supported services’.10 If the
mayor’s solution seems excessive and even caustic—and it ultimately was
rejected by the local city council—the Wnancial and social costs and problems
that he identiWed in connection with language variation are very real.
Education is the venue where this cost has been perhaps the most prom-
inent. Part of this prominence emerges from the plain emotive fact that
children provide the most visible face of education. Part emerges from the
role classrooms play both as forums for non-native speakers’ introduction to
English and Anglophone communities and in this way as emblems of how a
society sees itself or what it sees itself becoming. And part emerges from the
breadth of those aVected by educational programs for native and non-native
speakers. Inasmuch as public education is a government-funded and regu-
lated enterprise, that is, it aVects every member of a community, even those
who are not themselves students in schools or the parents or guardians of
students. Every member of a community, consequently, is likely to have an
opinion on education.
For all these reasons, the costs of language change and variation in educa-
tion are wide as well as high; as steep as the monetary value of bilingual,
transitional, and even monolingual programs may be, the price of these
programs in a community’s view of itself and its future is higher still. And
the more non-native speakers move into Anglophone regions or seek to
acquire English as a second language, the greater the cost of variation and
change. In the United States, an inXux of non-Anglophone workers from Asia
and Latin America has occurred over the past few decades. Not primarily
targeting large cities, as nineteenth-century European emigrants tended to do,
many of these speakers have settled in rural areas in particular—like Arcadia,
Wisconsin. Bringing their children with them, they have also dramatically

10 Quoted in Brian Voerding, ‘Aracadia Mayor: No Más’, La Crosse Tribune, 19 August 2006, A2.
250 Language Anxiety

aVected the direction and programs of many school districts. Since the federal
Bilingual Education Act lapsed in 2002, local districts have borne responsibility
not simply for educating these non-native speakers but for bringing them to
federally mandated performance levels, and the cost of this education—
in several senses—can be staggering. Between 1993 and 2002, the number
of limited-English students in several states tripled; in North Carolina it
quintupled. And if it can be Wnancially onerous simply to accommodate
such increases, it’s also onerous to Wnd qualiWed teachers, since comparatively
few of those in teacher training specialize in English as a second language.
Consequently, if school districts are to attract credentialed applicants, they
need to supplement salaries with various kinds of bonuses.11 And all this
means that tax payers in the district, monolingual as well as multilingual,
have a vested and vocal interest in how the school district responds to variation
and change.
This leaves yet one more cost of variation and change, and that has been
the level of anxiety it has produced, both among those who see these
linguistic phenomena as problems and would return to a pre-Babel era and
among those who would celebrate linguistic variation as a mark of human
diversity. Dialogue, discussion, and dissent may be inevitable means towards
progress and conXict resolution, but on this topic at least they have
more typically produced discord than resolved it. For critics like Honey,
Macdonald, and Quirk linguistic variation, while unavoidable, only serves
to advance social disruption and inequity, while for critics like Phillipson and
Crowley it reXects and furthers the identity of the individual groups who all
speak what is only nominally the same language and who have the right to
speak languages other than English. What both critical outlooks share is the
conviction that in their response to variation, Anglophones respond also
to each other’s dignity, which they promote by either suppressing or fostering
linguistic change. But despite this common goal, the dialogue between
what Cameron calls verbal hygienists and what Quirk calls linguistic liber-
ationists is unresolvable and Wnally, I believe, unconstructive. Indeed, it has
become similar to discussions between political adversaries that begin with a
question like, ‘Do you want world peace?’ And just as few politicians would
answer ‘no’, so would few Anglophones, at least publicly, assert that they
wanted actively to suppress the human rights and dignity of other speakers.
Both sides of the linguistic debate thus claim a kind of moral high ground
and thereby allow little possibility for a compromise meeting somewhere in

11 See Yilu Zhao, ‘Wave of Pupils Lacking English Strains Schools’, New York Times, 5 August
2002, A1, 11.
 Fixing English 251

the middle, at least so long as discussion foregrounds language as the problem

and the solution in all manner of social issues. It is in this sense, then,
that another cost of variation and change for Anglophones has been the
increase in anxiety about them that extended discussion and consideration
have produced. In doing little to ameliorate the channeling of social issues
through linguistic ones, this discussion has in fact done much both to
naturalize the connection and to amplify the concerns associated with it.
All that being said, linguistic variation and change also have clear value,
and I do not here advance merely an intellectual interest in linguistic diversity
or refer simply to some association between it and social diversity. Rather,
as the inevitable byproducts of language contact and social achievement,
variation and change testify to the vitality of an individual language and of
the collectivity of languages in general. While linguistic change often may
proceed mechanistically, with found facts like push chains accounting for a
sequence of phonological alterations, social phenomena just as often prove
to be the instigators of this change. When language changes in these instances,
it does so because society has advanced in some way, and examples include
everything from technological progress (which, as in the nineteenth century,
might supplement lexicon), to cultural revolution (which, as in the early
modern period, might lead to syntactic change in response to translation
needs), to social movement (which, as at the time of the Great Vowel
Shift, might utilize language as a means of group distinction).
Beyond serving as responses to expanding sociolinguistic experience, vari-
ation and change also constitute foundational aspects of English grammar
and usage. At a synchronic level even more fundamental than borrowed
lexicon or syntactic innovation, they enable features like strong verbs, variable
morphological rules, and vowel alteration as a plural marker. They also
produce variable stress patterns for related words (photograph as opposed to
photographer and photographic); for the same word functioning as two diVer-
ent parts of speech (present the noun as opposed to present the verb); and for a
word fulWlling two diVerent discursive functions (pronunciation of a direct
object when it’s topicalized to appear as the Wrst word in a sentence, as
opposed to when it’s not). This last issue points to the role variation more
generally plays in communicative competence. If grammatically competent
speakers are those who are able to produce well-formed sentences with a
language’s syntax, morphology, lexicon, and so forth, communicatively com-
petent ones are those who know how to adapt this grammatical competence
to the speciWc frames of particular domains. They know how to vary speech
depending on their interlocutors, topic, and locale; they know that one uses
diVerent vocabulary with children and adults, that one doesn’t bare one’s
252 Language Anxiety

soul to a stranger at a bus stop or tell raucous jokes at a funeral, and that
accommodation to another’s sociolect or regional dialect can be a means to
show solidarity. All this is to say that such knowledge is knowledge about
variable phenomena, and a speaker without it—a pre-Babel speaker who had
no sense of the need to alter speech depending on circumstances—would
qualify as what has memorably been labeled a ‘faulty interactant’.12 Whatever
social meanings accrue to variation and change and however detrimental
these meanings might be to certain speakers, the intrinsic value of variation
to expression and speakers is thus no less essential.
This connection between variation and communicative competence deep-
ens through consideration of the Xip-side of my earlier comment about how
linguistic variation can work against a large community identity. As a form of
social display, language accomplishes more than the transference of ideas
that Locke described so well. It foments revolution, expresses passion, and
releases tension. And also, as I have noted before, like clothing, hairstyles, and
musical taste, grammar and discursive habits can be ways of announcing—
and interpreting—social identity. All these things can constitute individuals’
self-representation and provide those around them with an impression of
their age, background, education, ethnicity, and social outlook. In the ter-
minology of M. A. K. Halliday, which I noted in the opening chapter,
language functions as a social semiotic by encoding social distinctions in
age, class, education, and so forth.13 At the broadest level, going back to the
ancient Greeks and their suspicion of foreigners, language has been a crucial
means for establishing ethnic and national identity, and in its own history,
English has indeed served and continues to serve distinct roles in the deWni-
tion of group culture, whether the group is one deWned by age-graded
features, by lexicon that emerges from ethnic or non-native traditions, or
by a conviction that language stands as a distinctive characteristic of national
identity. Adolescents and speakers of national as well as non-standard var-
ieties use language so eVectively in representing themselves as members of
a cohort that, Stephen May suggests, if language is not the pre-eminent
determinant of a cultural group that Herder and other German Romanticists
believed it to be, it is still a signiWcant factor: ‘language may not be intrinsic-
ally valuable in itself—it is not primordial—but it does have strong and felt
associations with ethnic and national identity’.14

12 See R. A. Hudson, Sociolinguistics, 2nd edn. (Cambridge: Cambridge University Press, 1996),
112–16.
13 Halliday, Language as Social Semiotic: The Social Interpretation of Language and Meaning
(London: Edward Arnold, 1978).
14 May, Language and Minority Rights: Ethnicity, Nationalism and the Politics of Language
(New York: Longman, 2001), 129.
 Fixing English 253

By extension, both historically and today, as the geographical, ethnic, and

social circumstances of Anglophones have changed, so, too, has English. The
earliest Germanic groups, perhaps even before they left the Continent,
encountered Roman paved roads for the Wrst time and borrowed what
became street for them. As Anglo-Norman barons occupied the Welsh
marches later in the Middle Ages, they borrowed the Welsh word crag to
describe the rocky outcrops they encountered. Later still, when English
explorers returned from overseas, they brought with them not only goods
from around the world but the words to describe them—like ketchup (Chi-
nese), bwana (Swahili), and chintz (Hindi). More complexly, change and
variation in the history of English testify for the development of sociolects
derived from contact with non-native speakers or through native speakers’
exploitation of what Halliday labels ‘the sociolinguistic play potential of one’s
own variety of the language’.15 Speakers of what may well be a non-standard
variety, that is, foster its covert prestige and their own sociolinguistic identity
by developing coded linguistic activity in syntax, lexicon, and discursive
practices; in maintaining the integrity of their own variety, such speakers
not only help to preserve their own social identity but also establish barriers
of intelligibility between themselves and other, more powerful groups. And
in this way, varieties as disparate as African American Vernacular English,
Geordie, and Singapore English oVer self-preserving group identiWcation in
much the same way as would have the divergence of the Ingvaeonic languages
(Old Saxon, Old Frisian, and Old English) from Primitive Germanic or
Webster’s declared independence of United States from British English. The
relatively recent presence of non-standard and even non-native speakers
in publicly prominent positions like that of network news broadcaster
points to this same value in variation as a reXection of changing group
identities.16 On a narrower but I tend to think no less signiWcant scale,
multilingual department store Santa Clauses, able to greet children as easily
with ‘Maligayang Pasko’ and ‘Feliz Navidad’ as with ‘Merry Christmas’,
likewise emblemize the Xexibility of language in deWning group identity.17
Variation and change thus prove as consequential to representation as they
do to grammatical structure and discourse conventions.
The advantages of variation can be pursued further, into a more
abstractly linguistic realm. It’s true enough that a concomitant of the twen-
tieth and twenty-Wrst century’s global economy—and particularly of the role

15 Halliday, Language as Social Semiotic, 160.

16 Mireya Navarro, ‘Breaking the Sound Barrier’, New York Times, 23 July 2006, 9, 1 and 10.
17 Patricia Leigh Brown, ‘Santa Claus Packs a Bag of Languages at a California Mall’, New York
Times, 21 December 2003, A5.
254 Language Anxiety

of Anglophone countries in this economy—has been the curtailment of

some languages and the contraction and even death of others. But as Muf-
wene has noted, another concomitant, typically overlooked in discussions
that focus on language loss, has been something akin to the process of
‘speciation in population genetics’.18 From a grammatical point of view,
change and variation point to the vitality of a language, since it is only living
languages that can change, but also constitute the medium by which any
language—to push Mufwene’s biological metaphor—produces new varieties
and languages. And in the area of speciation, English has of late been a
particularly fertile language.
It remains unclear at least to me how statistically signiWcant generalizations
can be made about the degree of current global linguistic diversity in com-
parison to that in any pre-literate, statistically unanalyzable moment.19 None-
theless, I will grant that the approximately 6,000 natural languages used today
may well represent a relative decrease from the number of languages spoken in
some pre-modern epoch. At the same time, however, English itself—in its
regional dialects, sociolects, and the various interlanguages (creoles and
pidgins) to which it has contributed—embodies more structural and socio-
linguistic diversity than it has at any moment in its recorded history. I wonder,
then, if the situation of the modern industrialized world is not simply whether
it does or does not display less linguistic diversity than did the pre-Columban
world but whether its diversity skews diVerently. Today, that is, variation
may predominate as a reXection and measure of what we now consider
intra-linguistic variation in languages used by large groups of speakers (Span-
ish and Chinese, perhaps, alongside English) rather than of the divergence
among structurally distant and distinct languages each of which has com-
paratively few speakers. And I would advance this line of reasoning still
further. Placing the common claim that variation among languages has
declined precipitously over time within a sociolinguistic context that none-
theless guarantees change, I wonder whether the implicit identiWcation
of linguistic variation with national languages isn’t yet another naturalized

18 Salikoko S. Mufwene, ‘Language Endangerment: What Have Pride and Prestige Got to Do with
It?’, in Brian D. Joseph et al. (eds.), When Languages Collide: Perspectives on Language ConXict,
Language Competition, and Language Coexistence (Columbus, OH: Ohio State University Press,
2003), 330.
19 One study claims, for example, not only that language loss has accelerated in the modern period
but that the perhaps 6,000 to 6,500 languages spoken today represent a signiWcant drop from the
15,000 languages that the study somehow estimates were spoken between 100,000 and 10,000 bc. See
Anna Ash, Jessie Little Doe Fermino, and Ken Hale, ‘Diversity in Local Language Maintenance and
Restoration: A Reason for Optimism’, in Leanne Hinton and Ken Hale (eds.), The Green Book of
Language Revitalization in Practice (San Diego, CA: Academic Press, 2001), 19–35.
 Fixing English 255

legacy of Romantic thought. I wonder whether linguistic variation isn’t

really a kind of zero-sum game. I wonder whether variation might have
remained constant between the Indo-European migrations and the present,
though it is now more intra- than inter-linguistic, and only time can tell
whether the pendulum of variation might some day swing back.
The title of this chapter turns on a pun, embracing Wx both in the sense
‘repair’ and in the sense ‘stabilize’. It’s a pun, or at least a double meaning,
present in metalinguistic discussions of English already in the early modern
debates that helped fashion a grammar of Standard English. In his 1712 Pro-
posal for Correcting, Improving, and Ascertaining the English Tongue, Jonathan
Swift advocates that ‘some Method should be thought on for ascertaining and
Wxing our Language for ever’, a notion reprised forty years later by David
Hume, who speaks of ‘Eminent and reWned geniuses . . . [who] . . . Wx the
tongue by their writings’.20 It is Dr Johnson, however, who perhaps predict-
ably makes the best use of this double meaning. In the preface to his
Dictionary, he notes, ‘Those who have been persuaded to think well of my
design, require that it should Wx our language, and put a stop to those
alterations which time and chance have hitherto been suVered to make in
it without opposition’.21 These two senses are often in play, not only in
metalinguistic discussions of English but also in the dynamics between
language and culture. Much of what I have described in this book, indeed,
could well be summarized by the seductive notion that if the language can
be stabilized, the culture will be repaired, and vice versa.
But English isn’t broken. It varies and changes because, despite the power
of the myth of Babel, that’s what natural languages do. While the written
channel might well be regarded as relatively stable right now, the spoken
channels—the many global sociolects and regional dialects—will never be.
And whatever other connections English may have to the social institutions
and actions of those who speak it, its own variability points to one thing
above all: that linguistically, these social groups are vital and healthy.

Roses and any other names

While I was writing the Wrst draft of this book, a controversy of the sort that
could arise only in a university occurred in my home institution. It was the
kind of controversy that I could imagine Wguring in one of David Lodge’s
academic novels. It involved the nickname for the university’s athletics teams.

20 OED, s. v. Wx, v.
21 Johnson, Dictionary of the English Language (1755; rpt. New York: AMS Press, 1967), i, Ciir .
256 Language Anxiety

In the early years of the university, several diVerent names had been used,
but beginning in the early 1950s the teams were known as the ‘Warriors’
and the image of the teams—on banners, uniforms, shirts, and so forth—was
that of a non-descript American Indian. The selection of Warriors was not
random, since the university was named for an Indian missionary, founded
by an order of Catholic priests with a long history of working on missions
and reservations, and located in an area that was once entirely inhabited by
various Algonquian nations. Already in the 1960s some students, faculty, and
local citizens began to express concern that the Warrior name and attendant
imagery showed disrespect to Native Americans and ought therefore to be
dropped. In the early 1970s the university did eliminate ‘Willie Wampum’, a
student who dressed as a caricatured Native American and appeared at
athletic events, especially basketball games, to cheer on the team and inspire
the crowd’s support. But the name Warriors remained into the 1990s, when,
following a vote by students and faculty and against the wishes of a great
many alumni, it was changed to Golden Eagles.
Largely because of disgruntled alumni, the nickname controversy never
entirely died, and a decade later it again reached a moment of crisis when
the university’s board of trustees agreed to review the situation and make a
deWnitive recommendation on whether the name should remain Golden
Eagles, revert to Warriors, or become something entirely new. Leading up
to the board’s deliberations and vote, various university and civic groups
expressed various opinions, which generally fell at one of two extremes: that
Warriors was and always would be oVensive to Native Americans, both
as people and in the public conception of them, and that Warriors not only
had a historical link to the university but need not be oVensive, as witnessed
by the many other universities that have retained the nickname, sometimes
(but not always) tacitly morphing the warrior from a Native American to
a medieval knight. For their part, Native Americans took positions on
both ends of this spectrum. Campus rallies were held, interviews given to
the local media, and opinion pieces published in magazines and newspapers.
One of these, endorsed by several members of the faculty, appeared in the
student-run campus newspaper.
The letter argued that faculty ‘have a responsibility to comment on the
issues of language and representation involved in the current debate’ and,
more generally, that words ‘accrue meaning—both literal and connotative—
only through historical usage and in real human communities’. Asserting that in
the United States warriors ‘connotes Wrst and formatively American Indians’, the
letter maintains that whether ‘this linkage ascribes to American Indians
traits that seem negative or positive, it denies them the dignity and right
 Fixing English 257

of self-representation. In the case of sports, it appropriates and misuses

American Indian experience.’ For all these reasons, the letter concluded,
‘Whatever the intentions of the Board of Trustees in considering reversing
this name change, there can be no doubt that the wider community [of the
university, city, and country] would perceive the return of the Warriors as
a return to an era of racial insensitivity’. What direct impact this letter had on
the board’s decision, I don’t know; it was only one of several expressions of
this same basic sentiment during the months that the controversy dragged
on. Eventually, the board did explicitly reject Warriors and, after brieXy toying
with the idea of yet another new name, agreed to retain Golden Eagles.
Though asked to sign this letter, I did not do so—not because I liked one
name over the other in the abstract, nor because I felt it was appropriate to
insult or limit the dignity of Native Americans or any other group. Partly
I didn’t sign because I didn’t think it much mattered what the university called
its teams; I thought that Warriors was a Wne nickname for an athletic team,
that it was odd to change it to Golden Eagles, that Golden Eagles was a Wne
nickname, and that amidst current curricular and institutional crises in
higher education it was still odder to be debating the issue at all, much less
contemplating a return to Warriors. Mostly I didn’t sign, however, because
I saw the argument as fundamentally Xawed, as the sort of argument
that anxiously muddies language change and social meanings in ways I have
examined throughout this book.
Without in any way questioning the motives or integrity of the letter’s
signers, I am struck by the way it both faults speakers for the language they use
and denies them responsibility for creating that language and, hence, the
opportunity to change their language or, more importantly, their thinking
and action. Responsibility rests instead with the language, in which meanings
may accrue ‘through historical usage and in real human communities’
but which at some point—now, evidently—become impervious to these
same usages and communities. While King Alfred, the Lollards, and Webster
had all maintained that language should change only up to a point that
they had determined and then become Wxed there, this argument takes
the mirror tactic that asserts change cannot take place after a point that the
writers had determined. Warriors is judged necessarily to connote Native
Americans, who, in turn, are judged necessarily delimited and demeaned
by this same connotation. And this necessity is predicated on a pattern of
usage throughout the history of the university and country. To use warriors
again, then, is necessarily to invoke earlier usage patterns and therefore
declare that the present university embodies the racial insensitivity, if not
racism, present in these patterns.
258 Language Anxiety

To my mind, then and now, there are serious practical and theoretical
problems with this line of reasoning that once more demonstrate how extra-
linguistic social anxieties can be channeled through language change and
variation. Derived from Old French werreieor, English warrior, in the sense
‘One whose occupation is warfare; a Wghting man, whether soldier, sailor, or
(latterly) airman; in eulogistic sense, a valiant or an experienced man of war’,
occurs already in Robert of Gloucester’s late thirteenth-century Metrical
Chronicle, where the connotation is obviously of an armored soldier: ‘KniZtes
and oþer worrerours, þat to þis londe wende’. This same sense predominates
in the fourteenth, Wfteenth, sixteenth, and seventeenth centuries—alongside,
I should note, an odd and expansive set of Wgurative usages involving women
and animals—and the Wrst recorded connotation of what we would today call
indigenes does not occur until 1788: ‘Many of their warriors, or distinguished
men, we observed to be painted in stripes, across the breast and back.’
Coming from John White’s Journal of a Voyage to New South Wales, however,
this passage refers not to Native Americans but to Aborigines. The Wrst OED
example with warrior as a reference to American Indians is from James
Fenimore Cooper’s 1826 Last of the Mohicans: ‘A swarthy band of the native
chiefs . . . with the warriors of their several tribes’.22
The practical point here is not that warriors was never used to connote
Native Americans until 1826; surely, there must have been signiWcantly earlier
(if unattested) usages in this sense. The practical point, as the faculty letter in
fact notes, is rather that words do indeed have histories. But what the letter
does not note as well is that these histories reXect the usage decisions of
human speakers and not some kind of atavistic spirit—something akin
to Milton’s etymological fallacy—within the language itself. Warrior did not
and does not always ‘Wrst and formatively’ connote American Indians, even
in the United States. A borrowing from French, it reXects the large-scale late-
medieval English importation of French courtly practices (like chivalry) and
the vocabulary associated with them; to be like the French would have
required (or at least beneWted from) words for speciWcally French institutions
and practices, words such as harness, majesty, and manor, all of which are
borrowed from French and Wrst recorded in English at about the same time
as warrior is.
Application of the word to indigenous peoples whom Anglophones
encountered as they spread around the globe equally reXects decisions of
individual and collective speakers. Given the European habit of framing
perception and reception of New World phenomena within the context

22 OED, s. v., warrior.

 Fixing English 259

of historical experience, particularly that of the Antique as described

and depicted in books, it was inevitable that Anglophones would think,
probably early on, to designate as warriors the ‘Wghting men’ they encountered
in North America and elsewhere around the globe.23 This was the label that
would have best categorized and explained such people. Additionally, in view
of the word’s history, such extrapolated designations would seem at least
initially to have been not merely descriptive but complimentary (or ‘eulogis-
tic’ as the OED labels it), building on a lexical tradition that began with the
appropriation of what was originally a term for a landed, noble Wghting man.
Two decades prior to Cooper’s The Last of the Mohicans, for instance, Words-
worth wrote ‘Character of the Happy Warrior’, whose eponymous hero is
not a mounted knight but someone who is diligent, moral, and compassion-
ate, and who,
. . . while the mortal mist is gathering, draws
His breath in conWdence of Heaven’s applause:
This is the happy Warrior; this is He
That every Man in arms should wish to be.24

And two centuries prior to Wordsworth, in translating Giovanni Biondi’s

History of the Civil Wars of England, the Earl of Monmouth refers unambigu-
ously to a seventeenth-century champion, again clearly not a mounted
knight, who ‘was rightly ranked in the number of the chiefe warriers of that
age’. Within this lexical context alone, but also within the rhetorical context of
representing indigenes as worthy opponents, there is by extension no denial of
indigenous dignity in either White’s or Cooper’s use of warrior as cited above.
And generalization from ‘soldiers’ to all Native Americans is a likewise logical
choice and parallels the semantic extension of a word like viking, which in
popular and academic usage has had its reference broaden from what might
best be called a sea-going pirate of particularly the ninth and tenth centuries
to all of the men, women, and children who lived in medieval Scandinavia.
Positive connotations for a warrior as a description of a Native American
‘Wghting man’ continue throughout the nineteenth century. A New York Times
headline of 23 April 1875, at the height of the so-called Plains Uprising, notes
the ‘Surrender of Hostile Comanche Chiefs and Thirty-Six Warriors’, and a
headline like this can be newsworthy only if the surrender itself is—which

23 On the Antique as a frame for early modern perceptions of the New World, see Anthony Grafton,
with April Shelford and Nancy Siraisi, New Worlds, Ancient Texts: The Power of Tradition and the Shock
of Discovery (Cambridge, MA: Harvard University Press, 1992).
24 Wordsworth, The Complete Poetical Works, ed. A. J. George (Cambridge, MA: Houghton MiZin,
1932), 341.
260 Language Anxiety

would require that the ‘Warriors’ be formidable foes and not individuals
lacking dignity. The article itself in fact goes on to distinguish the ‘thirty-six
braves’ from ‘140 women and children’, thereby drawing a clear distinction
between indigenous ‘Wghting men’ and indigenes in general.25 And even on
15 July 1899, by which point any Native American aggressive resistance to
Anglo expansion was long a thing of the past, the Times could still represent
Native American Wghting men as redoubtable with the headline ‘Pursuit of
Sioux Warriors’ and the subheadline ‘Three Parties of Indian Police are
After Swift Bear and His Band—An Uprising Feared’.26 The Times may well
have exaggerated or even demonized the Xeeing Sioux by describing them as
warriors, of course, but such rhetoric would prove the point that in reference
to Native Americans the word still had, at that time, neutral and even positive
connotations in the way that it validates and perhaps gloriWes pursuit by the
United States army. Without the dignity of being ‘Wghting men’, Swift Bear
and his band could ill justify fears of an uprising. Just over a decade later, a
positive, metaphoric use of warrior is traditionally ascribed to Glenn ‘Pop’
Warner in the pre-game pep talk that he gave to the football team of the
Carlisle Indian School prior to its 1912 game against, of all places, the United
States Military Academy at West Point, whose backWeld featured, of all people,
Dwight D. Eisenhower: ‘These men playing against you today are soldiers.
These are the Long Knives. You are Indians. Tonight we will know whether
or not you are warriors.’27
Any caricaturing or denial of dignity in the connotations of warrior
must have been one semantic step farther on. The experiences of Native
Americans who fought in the First World War may have been a factor, since
stereotypes of Indians as instinctive, natural Wghters could have joined with

25 New York Times, 23 April 1875, 1.

26 New York Times, 15 July 1899, 1.
27 Quoted in Alexander M. Weyand, The Saga of American Football (New York: Macmillan, 1955),
101. Although the New York Times report the day after the game does not list Weyand as having
participated, he was the captain of the 1915 Army team and so it’s reasonable to suppose he knew
people who did. He claims to have gotten the quotation from Bill Stern, a popular sports broadcaster
in the 1930s, 1940s, and 1950s. Pop Warner barely mentions the incident in his autobiography, but two
Carlisle players (Gus Welch and Pete Calac) did attribute these sentiments to Warner in his pep talk,
though they made no attempt to quote him verbatim. Evidently, Warner, who was a strong advocate of
Carlisle and who was very loyal to his players, in general liked to use the fact that they were Indians
playing white men as a motivational tool. See Robert M. Wheeler, Jim Thorpe: World’s Greatest Athlete
(Norman, OK: University of Oklahoma Press, 1975), 128; Bill Crawford, All American: The Rise and Fall
of Jim Thorpe (Hoboken, NJ: John Wiley and Sons, 2005), 188; Lars Anderson, Carlisle vs. Army: Jim
Thorpe, Dwight Eisenhower, Pop Warner, and the Forgotten Story of Football’s Greatest Battle (New York:
Random House, 2007), 278; Sally Jenkins, The Real All Americans: The Team that Changed a Game, a
People, a Nation (New York: Doubleday, 2007), 7. Though Weyand’s book is Jenkins’s source, she oVers
a slightly diVerent version of the quotation.
 Fixing English 261

racial stereotypes to undermine the very success implicit in representations

of Native American soldiers; their success, that is, conWrmed that they
were ‘blood-thirsty warriors’.28 But warrior clearly retained positive, self-
representational qualities on the two-year anniversary of Armistice Day,
when, at a memorial at the Tomb of the Unknown Soldier in Washington,
D C, the Crow chief Plenty Coups concluded his speech by saying, ‘I hope that
the Great Spirit will grant that these noble warriors have not given up their
lives in vain and that there will be peace to all men hereafter’.29 And even today
the word warrior can preserve not only its medieval associations with knight-
hood, as at those American universities that use the name and a mounted
knight as a mascot, but also its positive connotations for at least some Native
Americans, as among the Portland, Oregon, group ‘Warriors for Christ’.
The goal of this group is ‘to encourage and strengthen spiritual growth
among Native Americans through evangelism and discipleship training, and
to strengthen and plant Native churches on Indian reservations and urban
areas throughout North America’, and its web page emblazons its name over a
picture of buValo, a plains teepee, and a bonneted young man.30 From a
global perspective, the word likewise can retain laudatory, self-representa-
tional connotations for other indigenous peoples, as in the 1994 novel Once
Were Warriors, which holds out the word and its associations as symbols
of the status lost to Maori culture in the brutal and economically debilitated
urban housing projects of the New Zealand present.31
These practical problems in evaluating how meaning accrues to a word
like warrior point to a larger theoretical problem. And that is what I’ve come
to think of, in a metaphor that’s perhaps grotesque but nonetheless apropos,
as having your Sapir-Whorf hypothesis but eating it, too. The positions
expressed in the faculty letter share many educational and governmental
sentiments that I have traced throughout this book by attributing to language
and language change an overarching and preemptive power to shape reality—
to deWne experience in such a way, through grammar and discursive patterns,
as (in this instance) to deny individuals ‘dignity and [the] right of self-
representation’. At the same time, like these other sentiments the letter’s
arguments are predicated on the notion that some individuals, at least, can
transcend the determinative power of language, historically by creating and
then limiting a word’s connotations once and for all and currently by being
able to evade the linguistic determination of experience in order to evaluate
28 Thomas A. Britten, American Indians in World War I: At Home and at War (Albuquerque, NM:
University of New Mexico Press, 1997), 99–115.
29 Quoted in Britten, American Indians in World War I, 160.
30 [Link]
31 Alan DuV, Once Were Warriors (Honolulu, HI: University of Hawaii Press, 1994).
262 Language Anxiety

language itself and the process by which this same creation took place. By
this line of reasoning, indeed, a word like warrior behaves in semantically
peculiar, even unique ways. For unexplained reasons, the implicit claim seems
to be that warrior may have been borrowed into English without Anglophones
appropriating and misusing the experience of French chevaliers, and it
may have had its meaning extended from ‘mounted, armed Wghting man’
simply to ‘Wghting man’ or even ‘virtuous man’ without similarly aVecting the
experience of early modern soldiers or morally upright individuals. But once
applied to indigenous Wghting men in the United States, the denotation
and connotation of the word became permanently negative. It’s as if the
‘real human communities’ that give the word its meaning exclude all those
that existed before and after some historical moment at which the meaning
of warrior is presumed to have become forever Wxed.
Language thereby becomes both the manufacturing force that the strictest
versions of Sapir-Whorf describe—the force that fashions reality and that
renders individuals who speak diVerent languages as the inhabitants
of eVectively diVerent worlds—but also the literal subject of certain speakers,
particularly malfeasant ones. In this way, language change once more chan-
nels social anxiety, for it is by linguistic change that the word warrior allegedly
came Wrst and foremost to mean ‘Native American’ in a way that denied
indigenes ‘the dignity and right of self-representation’, and it is through
resistance to linguistic change, the letter asserts, that the university could
resist ‘a return to an era of racial insensitivity’. Change and variation also
become mystiWed yet again, since while speakers have the ability to invent
words and change at least some of their meanings across time, after some
point, for unexplained reasons, they lose this ability: they cannot now,
evidently, recuperate warrior. And the ascription of all such developments
to some vague historical agency like ‘social power’ as reXected in dominant
institutions not only further mystiWes change but also exonerates speakers
from the responsibility for the language they use and their decisions to change
it or leave it as is. It is just this kind of adaptation of the Sapir-Whorf
hypothesis that I have opposed in this book, arguing instead that it is speakers
who invest linguistic forms with meanings, which may in fact then shape their
reality but for which they remain responsible. And I have argued that just as
all interlocutors in a conversation are accountable for the direction and sense
of their conversation, all Anglophones are accountable for English, for as
James Milroy has observed, ‘it is speakers, and not languages, that innovate’.32

32 Milroy, Linguistic Variation and Change: On the Historical Sociolinguistics of English (Oxford:
Blackwell, 1992), 169.
 Fixing English 263

By extension when we respond to speakers through language alone, we

depersonalize and obfuscate this accountability.
It may well be that there are some words that are so semantically contamin-
ated that they cannot be rehabilitated in any way. It’s diYcult to imagine
an athletic team called the nazis or the niggers, for example, because it’s diYcult
to imagine that these words could ever become separated from their horriWc—
and not just demeaning—connotations. But even as I say this, I want to add
that when it comes to language, human will can be a powerful mechanism
for shaping pragmatics. Both vandal and cossack, for instance, would seem to
have irrevocably negative connotations of brutality and destruction, and yet
each is now used for United States university athletic teams. And these appear
alongside various kinds of devils, demons, buccaneers, vikings, and, of course,
warriors. There is even an institution that proudly and without any sense of
irony or ethnic stereotyping designates its teams the ‘Fighting Irish’.
From its historical associations with slavery and racism, nigger would seem
particularly unlikely ever to be semantically ameliorated: it has become
perhaps the most notorious word in English, so notorious, in fact, that its
utterance in certain circumstances has been likened to a violent speech act
akin to yelling ‘Wre’ in a crowded theater.33 With the widespread use of the
form [nig@] in rap music, however, the word has lost these inXammatory
associations in some domains, in the same way that gay and perhaps queer
have themselves become ameliorated. In all cases, by casually using the words
and not directing them as insults or simply avoiding them, as had been
customary, speakers demystify them and denude them of at least some of
their power to shock and oVend. In some usages nigger has in fact come to
serve almost as an endearment, particularly among young African Americans.
But the limits of this usage—and the responsibility of speakers for fashioning
not only what they say but what they do—appeared in an assault case in
New York City in 2006. The fact that one young man beat another was not
in dispute; what was in dispute was whether the attack qualiWed as a hate
crime based on race, for which the penalty would be greater. And this
qualiWcation turned on the signiWcance of the defendant’s use of the ‘n-
word’, as prosecutors and defense attorneys regularly referred to it throughout
the trial, avoiding uttering the word nigger themselves. The defendant, having
grown up in a racially mixed neighborhood, maintained that he used the
word as part of his ordinary discourse and that it consequently did not have
its overwhelmingly negative historical connotations. The prosecutors argued

33 Randall Kennedy, Nigger: The Strange Career of a Troublesome Word (New York: Pantheon, 2002).
264 Language Anxiety

that it did, and in the end were able to secure conviction of Wrst- and second-
degree robbery as a hate crime.34
As the faculty letter’s reference to historical usage indicates, the history
of words matters: it oVers a window on communities that used language and
on how it and their world have shaped one another as well as the present.
But it should be the whole history that matters, not just an expedient
selection, and words that have the historical, visceral impact of nazi and
nigger are few and far between in English. Even granting the historical
associations of warrior with American Indians, one would be hard-pressed
to locate usages and a social consensus that warrior was one of these contam-
inated words. Everything I have traced—from Robert of Gloucester to Pop
Warner to modern New Zealand—suggests otherwise. And to my mind,
further, to try to increase the corpus of damaged, unusable words only further
serves to mystify language change, divert responsibility for usage from
speakers, and (thereby) avoid the real social issues associated with the
words. Prevention of a return to warrior oVers a temporary satisfaction, but
in doing so it also suppresses consideration of the very issues that seem to
make the word into a problem. It is worth remembering, indeed, that words
acquire negative and positive connotations in precisely the same way:
speakers’ usage. Just as speech communities have fashioned any word’s
history and negative connotations, so might they rehabilitate it. Just as they
borrowed a word like warrior and changed its meaning in various ways,
so might they continue to change it, its connotations, and, most importantly,
the social issues it raises.
Words obviously do matter, and changing language can sometimes
change the world in which we live. Words can aVect how we see this world
and can be at least a partial means for altering that world. Such was the case in
the 1960s when, in the Wrst wave of feminism, many speakers came to resist
strongly the casual use of girls for women who worked in oYces—a usage that
required the untenable presumption that labeling these women as children or
adolescents did not adversely aVect others’ perception of them. Forty years
later, at least in the United States, this usage is almost literally unheard of and
could well be, in any case, grounds for a sexual harassment law suit. And at
the very least, this change in usage reXects some positive change in the status
of female oYce-workers, though avoidance of girl has scarcely produced a
society free of sexism.

34 Corey Kilgannon, ‘Tolerance for a Racial Slur Is a Test for Potential Jurors’, New York Times, 18
May 2006, C17; and Kilgannon, ‘Attacker Guilty of Hate Crimes in Howard Beach’, New York Times, 10
June 2006, B1.
 Fixing English 265

But it is my argument that throughout the history of English, and perhaps

particularly in the wake of another of Orwell’s works, ‘Politics and the English
Language’,35 change and variation—or their absence in an imagined pre-Babel
world—have served far too easily as channels for non-linguistic issues and
have thereby distorted discussion of vital social concerns. They have been
used to imagine a world of perfect, unvarying communication, which in turn
has been used to formulate all manner of institutional activity and to produce
all manner of social anxiety. If the study of natural languages shows that such
a world is impossible, the Tower of Babel—the myth that seems to explain
so much—shows that even if perfect communication were possible, it would
also be lethal, both oVering an aVront to God and leading to the scattering
of humanity. As Orwell said so well, language can serve to mask otherwise
unacceptable political and social activity, and it is easier to change human
language than human actions and beliefs. In that sense, as with girl, manipu-
lation of language can have positive beneWts beside the negative ones Orwell
describes—it can at least encourage social change by causing speakers to think
about what they say.36
Ultimately, however, it is social activity, not the language, that is both
the problem and its solution. Language is not responsible for sexism, racism,
or various other kinds of social injustice. It may be a tool to such injustice, but
it is a tool used by speakers, who have the power not merely to invent words
and use them with negative connotations but also to reject these connota-
tions. More importantly, speakers have the power to elect not to blame
language change and variation for the social issues they encounter in the
world—not to transfer non-linguistic anxieties onto linguistic topics—but
directly to address these issues. Perhaps above all, my discussion of warrior
illustrates just this point: that diVerent speakers can invest diVerent meanings
into linguistic activity and that language does not determine social reality (as
I understand the faculty letter to presume) but rather exists in a kind of
symbiotic relationship with it. When we blame language and use it as a
euphemism for something extra-linguistic, the something else always seems
to Wnd a way to persist and take on new forms. ‘The euphemism treadmill’, as
Stephen Pinker calls it, ‘shows that concepts, not words, are in charge: give

35 Orwell’s essay, which has been widely circulated, particularly in rhetoric handbooks, Wrst
appeared in Horizon in April of 1946.
36 Cf. Ann Curzan’s comment: ‘Whether or not changing language eventually changes attitudes
remains an open question; clearly, however, the simple fact that language reform requires speakers to
think about a linguistic construction and its possible social implications—be they sexist, racist, or
otherwise discriminatory—brings a level of awareness of these issues to a speech community that
might not otherwise be achieved’ (Gender Shifts in the History of English [Cambridge: Cambridge
University Press, 2003], 180).
266 Language Anxiety

a concept a new name, and the name becomes colored by the concept; the
concept does not become freshened by the name’. Having noted the replace-
ment in common discourse of Negro by black and then by African American,
he continues: ‘We will know we have achieved equality and mutual respect
when names for minorities stay put.’37
To assign blame to language, and to minister to it alone, provides short-term
solutions to long-term social problems. It also creates still more problems,
since to do so allows speakers to avoid the cause of a problem and examine
only its symptom. The narrative, political, religious, educational, and linguis-
tic issues I have examined here suggest just how debilitating such avoidance
can be. Totalitarianism can exist with or without the word nazi, racism
with or without nigger or warrior, and sexism with or without girl. Simply to
change language, or to worry over what language change reveals about a
changing society, allows speakers to evade responsibility for what they say.
Of far greater consequence, it also allows them to deny responsibility for
what they do.

37 Pinker, ‘The Game of the Name’, New York Times, 5 April 1994, A21. A version of the issue Pinker
describes aVects even the naming of varieties of English predominantly used by minorities. What
school teachers a half century ago might have called simply ‘dialect’ became, in the 1970s, Black English
Vernacular, then African American English, then Ebonics, and then African American Language; and
since the name for this variety has been so unstable and even contentious, there has never been at any
one time consensus over just what label to use. Transformation of the name into an abbreviation—so
African American Language becomes AAL, in opposition to the Language of Wider Communication,
or LWC—reiWes the name and further distances it not only from the variety to which it refers but also
from whatever social issues are associated with that variety. See the usage in, for example, Geneva
Smitherman, Word from the Mother: Language and African Americans (New York: Routledge, 2006).
 Works Cited

AarsleV, Hans, The Study of Language in England, 1780–1860, 2nd edn. (Minneapolis,
MN: University of Minnesota Press, 1983).
Achebe, Chinua, Anthills of the Savannah (New York: Anchor, 1987).
Adams, J. N., Bilingualism and the Latin Language (New York: Cambridge University
Press, 2003).
Aitchison, Jean, Language Change: Progress or Decay?, 3rd edn. (Cambridge:
Cambridge University Press, 2001).
Algeo, John, ‘What is a Briticism?’, in A. N. Doane, Joan Hall, and Richard Ringler
(eds.), Old English and New: Studies in Language and Linguistics in Honor of Frederic
G. Cassidy (New York: Garland, 1992), 287–304.
—— (ed.), The Cambridge History of the English Language, vi, English in North
America (Cambridge: Cambridge University Press, 2001).
—— ‘External History’, in Algeo (ed.), English in North America, 1–58.
Allen, Chadwick, Blood Narrative: Indigenous Identity in American Indian and Maori
Literary and Activist Texts (Durham, NC: Duke University Press, 2002).
Alvarez, Lizette, ‘It’s the Talk of Nueva York: The Hybrid Called Spanglish’, New York
Times, 25 March 1997, A1, 14.
Anderson, Benedict, Imagined Communities: ReXections on the Origin and Spread
of Nationalism, rev. edn. (London: Verso, 1991).
Anderson, Lars, Carlisle vs. Army: Jim Thorpe, Dwight Eisenhower, Pop Warner,
and the Forgotten Story of Football’s Greatest Battle (New York: Random
House, 2007).
Arnold, Matthew, On the Study of Celtic Literature and On Translating Homer
(New York: Macmillan Company, 1904).
Ash, Anna, Jessie Little Doe Fermino, and Ken Hale, ‘Diversity in Local Language
Maintenance and Restoration: A Reason for Optimism’, in Hinton and Hale (eds.),
The Green Book of Language Revitalization in Practice, 19–35.
Aston, Margaret, ‘Wyclif and the Vernacular’, in Anne Hudson and Michael Wilks
(eds.), From Ockham to Wyclif, Studies in Church History, Subsidia 5 (Oxford:
Blackwell, 1987), 281–330.
Axtell, James, ‘Babel of Tongues: Communicating with the Indians in Eastern North
America’, in Gray and Fiering (eds.), The Language Encounter in the Americas,
1492–1800, 15–60.
Bacon, Roger, Opera Quædam Hactenus Inedita, ed. J. S. Brewer (London: Longman,
1859).
Bailey, Charles-James N., Essays on Time-Based Linguistic Analysis (Oxford: Clarendon
Press, 1996).
268 Language Anxiety

Bailey, Richard W., ‘The English Language in Canada’, in Bailey and Manfred Görlach
(eds.), English as a World Language (Ann Arbor, MI: University of Michigan
Press, 1982), 134–76.
—— ‘American English Abroad’, in Algeo (ed.), English in North America, 456–96.
—— Images of English: A Cultural History of the Language (Ann Arbor, MI: University
of Michigan Press, 1991).
Baron, Dennis, The English-Only Question: An OYcial Language for Americans?
(New Haven, CT: Yale University Press, 1990).
Bauer, Laurie, ‘English in New Zealand’, in BurchWeld (ed.), English in Britain and
Overseas: Origins and Development, 382–429.
Baugh, Albert C. and Thomas Cable, A History of the English Language, 5th edn.
(Upper Saddle River, NJ: Prentice Hall, 2002).
Bautista, M. L. S. (ed.), English Is an Asian Language: The Philippine Context:
Proceedings of the Conference Held in Manila on August 2–3, 1996 (Sydney:
Macquarie Library, 1997).
Beadle, Richard, ‘The Virtuoso’s Troilus’, in Ruth Morse and Barry Windeatt (eds.),
Chaucer Traditions: Studies in Honour of Derek Brewer (Cambridge: Cambridge
University Press, 1990), 213–33.
Beal, Joan C., English Pronunciation in the Eighteenth Century: Thomas Spence’s ‘Grand
Repository of the English Language’ (Oxford: Clarendon Press, 1999).
Bede, The Venerable, Expositio Actuum Apostolorum et Retractatio, ed. M. W. Laistner
(Cambridge, MA: Medieval Academy of America, 1939).
—— A History of the English Church and People, trans. Leo Sherley-Price (Harmonds-
worth: Penguin, 1955).
Béjoint, Henri, Tradition and Innovation in Modern English Dictionaries (Oxford:
Clarendon Press, 1994).
Ben-Rafael, Eliezer, Language, Identity, and Social Division: The Case of Israel (Oxford:
Clarendon Press, 1994).
Benton, Richard A., ‘Language Policy in New Zealand: DeWning the IneVable’,
in Herriman and Burnaby (eds.), Language Policies in English-Dominant Countries,
62–98.
Bernstein, Richard, ‘A Snappy Slogan? In German? Don’t Smile. Try English’, New York
Times, 21 December 2004, A4.
Berthelette, Thomas (ed.), Jo. Gower de Confessione Amantis (London: Berthelette,
1532).
Bex, Tony and Richard J. Watts (eds.), Standard English: The Widening Debate
(London: Routledge, 1999).
Y Bibl Cyssegr-lan: Sef yr Hen Destament a’r Newydd (London: Robert Barker, 1630).
Bickerton, Derek, Language and Human Behavior (Seattle, WA: University of
Washington Press, 1995).
Blake, N. F., Non-Standard Language in English Literature (London: Andre
Deutsch, 1981).
 Works Cited 269

—— (ed.), The Cambridge History of the English Language, ii, 1066–1476 (Cambridge:
Cambridge University Press, 1992).
Blank, Paula, ‘The Babel of Renaissance English’, in Lynda Mugglestone (ed.),
The Oxford History of English (Oxford: Oxford University Press, 2006), 212–39.
BloomWeld, Leonard, Language (New York: Holt, 1933).
Bodine, Ann, ‘Androcentrism in Prescriptive Grammar: Singular ‘‘They’’, Sex-indeWnite
‘‘He’’, and ‘‘He or She’’ ’, in Deborah Cameron (ed.), The Feminist Critique of
Language: A Reader (London: Routledge, 1990), 166–86.
Bolton, Kingsley, Chinese Englishes: A Sociolinguistic History (Cambridge: Cambridge
University Press, 2003).
Borst, Arno, Der Turmbau von Babel: Geschichte der Meinungen über Ursprung
und Vielfalt der Sprachen und Völker, 4 vols. in 6 (Stuttgart: Anton Hiersemann,
1957–63).
Boulton, Marjorie, Zamenhof: Creator of Esperanto (London: Routledge, 1960).
Bradley, S. A. J., Anglo-Saxon Poetry (London: Dent, 1982).
Brewer, Charlotte, Treasure-House of the Language: The Living OED (New Haven,
CT: Yale University Press, 2007).
Bridges, Robert, ‘The Society’s Work’ (Society for Pure English, tract XX, 1925),
excerpted in W. F. Bolton and David Crystal (eds.), The English Language, 2 vols.,
ii, Essays by Linguists and Men of Letters, 1858–1964 (Cambridge: Cambridge Uni-
versity Press, 1969), 86–99.
Brinton, Laurel and Margery Fee, ‘Canadian English’, in Algeo (ed.), English in North
America, 422–40.
Britten, Thomas A., American Indians in World War I: At Home and at War
(Albuquerque, NM: University of New Mexico Press, 1997).
Brooke, James, ‘For Mongolians, E is for English, F is for Future’, New York Times,
15 February 2005, A1, 9.
Brown, Patricia Leigh, ‘Santa Claus Packs a Bag of Languages at a California Mall’,
New York Times, 21 December 2003, A5.
Bryson, Bill, The Mother Tongue: English & How It Got That Way (New York:
W. Morrow, 1990).
Bucken-Knapp, Gregg, Elites, Language, and the Politics of Identity: The Norwegian
Case in Comparative Perspective (Albany, NY: State University of New York Press,
2003).
Bullokar, William, The Amendment of Orthographie for English Speech (1580;
rpt. Amsterdam: Theatrum Orbis Terrarum, 1968).
—— An English Expositor: Teaching the Interpretation of the Hardest Words Used in
Our Language (London: John Legatt, 1616).
BurchWeld, Robert (ed.), The Cambridge History of the English Language, v, English in
Britain and Overseas: Origins and Development (Cambridge: Cambridge University
Press, 1994).
—— ‘Introduction’, in BurchWeld (ed.), English in Britain and Overseas: Origins and
Development, 1–19.
270 Language Anxiety

Burgess, Anthony, A Clockwork Orange (London: Penguin, 1972).

Burke, Peter, Languages and Communities in Early Modern Europe (Cambridge:
Cambridge University Press, 2004).
Burling, Robbins, The Talking Ape: How Language Evolved (Oxford: Oxford University
Press, 2005).
Burridge, Kate, Blooming English: Observations on the Roots, Cultivation and Hybrids
of the English Language (Cambridge: Cambridge University Press, 2004).
Butler, Susan, ‘Corpus of English in Southeast Asia: Implications for a Regional
Dictionary’, in Bautista (ed.), English Is an Asian Language, 103–24.
Bybee, Joan L., Phonology and Language Use (Cambridge: Cambridge University Press,
2001).
—— Frequency of Use and the Organization of Language (Oxford: Oxford University
Press, 2007).
—— and Paul Hopper, ‘Introduction’, in Bybee and Hopper (eds.), Frequency and
the Emergence of Linguistic Structure, 1–24.
—— and Paul Hopper (eds.), Frequency and the Emergence of Linguistic Structure
(Amsterdam: John Benjamins, 2001).
Cable, Thomas, ‘The Rise of Written Standard English’, in Aldo Scaglione (ed.),
The Emergence of National Languages (Ravenna: Longo Editore, 1984), 75–94.
Calvet, Louis-Jean, Language Wars and Linguistic Politics, trans. Michel Petheram
(Oxford: Oxford University Press, 1998).
Camden, William, Remains Concerning Britain (1606; rpt. Yorkshire: EP Publishing
Limited, 1974).
Cameron, Deborah, Feminism and Linguistic Theory, 2nd edn. (New York: St Martin’s
Press, 1992).
—— Verbal Hygiene (London: Routledge, 1995).
Campbell, Alistair, Old English Grammar (Oxford: Clarendon Press, 1959).
Carew, Richard, ‘The Excellency of the English Tongue’, in Camden, Remains
concerning Britain, 42–51.
Cassidy, Frederic G. (chief editor), Dictionary of American Regional English, 4 vols.
(Cambridge, MA: Belknap Press of Harvard University Press, 1985–).
—— ‘Geographical Variation of English in the United States’, in Richard Bailey
and Manfred Görlach (eds.), English as a World Language (Ann Arbor, MI:
University of Michigan Press, 1982), 177–209.
—— and R. B. Le Page (eds.), Dictionary of Jamaican English (Cambridge: Cambridge
University Press, 1967).
Cawley, A. C. (ed.), The WakeWeld Pageants in the Towneley Cycle (Manchester:
Manchester University Press, 1958).
Caxton, William, The Prologues and Epilogues of William Caxton, ed. W. J. B. Crotch,
EETS os 176 (1928; rpt. New York: Burt Franklin, 1971).
Chambers, J. K., ‘ ‘‘Lawless and Vulgar Innovations’’: Victorian Views of Canadian
English’, in Sandra Clarke (ed.), Focus on Canada (Amsterdam: John Benjamins,
1993), 1–26.
 Works Cited 271

—— Sociolinguistic Theory: Linguistic Variation and Its Social SigniWcance (Oxford:

Blackwell, 1995).
Chaucer, GeoVrey, The Riverside Chaucer, ed. Larry D. Benson, 3rd edn. (Boston,
MA: Houghton MiZin, 1987).
Ch’ien, Evelyn Nien-Ming, Weird English (Cambridge, MA: Harvard University Press,
2004).
Chomsky, Noam, Aspects of the Theory of Syntax (Cambridge, MA: MIT Press, 1965).
Clanchy, M. T., From Memory to Written Record: England 1066–1307, 2nd edn. (Oxford:
Blackwell, 1993).
Clyne, Michael, Dynamics of Language Contact: English and Immigrant Languages
(Cambridge: Cambridge University Press, 2003).
Cobarrubias, Juan, ‘Ethical Issues in Status Planning’, in Cobarrubias and Joshua
A. Fishman (eds.), Progress in Language Planning: International Perspectives (Berlin:
Mouton, 1983), 41–85.
Coleman, Julie, A History of Cant and Slang Dictionaries, i, 1567–1784 (Oxford: Oxford
University Press, 2004).
Collingwood, R. G., The Idea of History (Oxford: Clarendon Press, 1946).
Connor, George Alan et al., Esperanto: The World Interlanguage (London: A. S. Barnes,
1948).
Cooper, Robert L. (ed.), Language Spread: Studies in DiVusion and Social Change
(Bloomington, IN: Indiana University Press, 1982).
—— Language Planning and Social Change (Cambridge: Cambridge University Press,
1989).
Crawford, Bill, All American: The Rise and Fall of Jim Thorpe (Hoboken, NJ: John
Wiley and Sons, 2005).
Crawford, James (ed.), Language Loyalties: A Source Book on the OYcial English
Controversy (Chicago, IL: University of Chicago Press, 1992).
—— Bilingual Education: History, Politics, Theory, and Practice, 4th edn. (Los Angeles,
CA: Bilingual Educational Services, 1999).
—— At War with Diversity: US Language Policy in an Age of Anxiety (Clevedon:
Multilingual Matters, 2000).
Crowley, Robert (ed.), Piers Plowman (London: Crowley, 1550).
Crowley, Tony, The Politics of Discourse: The Standard Language Question in British
Cultural Debates (London: Macmillan, 1989).
—— ‘Curiouser and Curiouser: Falling Standards in the Standard English Debate’,
in Bex and Watts (eds.), Standard English: The Widening Debate, 271–82.
Crystal, David, The Cambridge Encyclopedia of the English Language (Cambridge:
Cambridge University Press, 1995).
—— Language Death (Cambridge: Cambridge University Press, 2000).
—— Language and the Internet (Cambridge: Cambridge University Press, 2001).
—— English as a Global Language, 2nd edn. (Cambridge: Cambridge University
Press, 2003).
—— The Stories of English (Woodstock, NY: Overlook Press, 2004).
272 Language Anxiety

Crystal, David, The Fight for English: How Language Pundits Ate, Shot, and Left
(Oxford: Oxford University Press, 2006).
Cursor Mundi, ed. Richard Morris, EETS os 57, 59, 62, 66, 68, 99, 101 (London:
Trübner, 1874–93).
Curzan, Ann, Gender Shifts in the History of English (Cambridge: Cambridge Univer-
sity Press, 2003).
Dalby, Andrew, Language in Danger: The Loss of Linguistic Diversity and the Threat
to Our Future (New York: Columbia University Press, 2003).
Daley, Suzanne, ‘In Europe Some Fear National Languages are Endangered’, New York
Times, 16 April 2001, A1, 10.
Daniel, Samuel, Musophilus: Containing a General Defense of Learning, ed. Raymond
Himelick (West Lafayette, IN: Purdue University Studies, 1965).
Dante, De Vulgari Eloquentia, in Alex Preminger et al. (ed. and trans.), Classical and
Medieval Literary Criticism: Translations and Interpretations (New York: Frederick
Ungar, 1974), 412–46.
—— The Paradiso, trans. John Ciardi (New York: Mentor, 1970).
Defoe, Daniel, Essays upon Several Projects (London: Thomas Ballard, 1702).
DeJong, David H., Promises of the Past: A History of Indian Education in the United
States (Golden, CO: American Press, 1993).
De Klerk, Vivan (ed.), Focus on South Africa (Amsterdam: John Benjamins, 1996).
Del Valle, Sandra, Language Rights and the Law in the United States: Finding Our Voices
(Clevedon: Multilingual Matters, 2003).
De Tocqueville, Alexis, Democracy in America, ed. Phillips Bradley, 2 vols. (New York:
Vintage, 1945).
Deutscher, Guy, The Unfolding of Language: An Evolutionary Tour of Mankind’s
Greatest Invention (New York: Henry Holt, 2005).
Devitt, Amy J., Standardizing Written English: DiVusion in the Case of Scotland
1520–1659 (Cambridge: Cambridge University Press, 1989).
Dickens, Charles, Great Expectations, ed. Angus Calder (London: Penguin, 1965).
—— Nicholas Nickleby, ed. Mark Ford (London: Penguin, 1999).
Dicker, Susan J., Languages in America: A Pluralist View (Philadelphia, PA: Multilingual
Matters, 1996).
Dixon, R. M. W., The Rise and Fall of Languages (Cambridge: Cambridge University
Press, 1997).
Dobson, E. J., English Pronunciation 1500–1700, 2nd edn., 2 vols (Oxford: Clarendon
Press, 1968).
Dryden, John, The Works of John Dryden, ed. Edward Niles Hooker and H. T.
Swedenberg, Jr., 7 vols. (Berkeley, CA: University of California Press, 1956–).
DuV, Alan, Once Were Warriors (Honolulu, HI: University of Hawaii Press, 1994).
Eakin, Emily, ‘Going at the Changes in, Ya Know, English’, New York Times, 15
November 2003, A15, 17.
Echevarrı́a, Roberto González, ‘Is ‘‘Spanglish’’ a Language?’, New York Times, 28 March
1997, A29.
 Works Cited 273

Eco, Umberto, The Search for the Perfect Language (Oxford: Blackwell, 1995).
Edwards, Viv, Multilingualism in the English-Speaking World: Pedigree of Nations
(Malden, MA: Blackwell, 2004).
Elton, G. R., The Tudor Constitutions: Documents and Commentary, 2nd edn.
(Cambridge: Cambridge University Press, 1982).
Elyot, Sir Thomas, The Boke Named the Gouernour (1531; rpt. Menston: Scolar Press,
1970).
‘English-Only Bill Vetoed’, New York Times, 18 April 2005, A15.
Fairclough, Norman, Language and Power (London: Longman, 1989).
Fisher, John H., ‘British and American, Continuity and Divergence’, in Algeo (ed.),
English in North America, 59–85.
—— The Emergence of Standard English (Lexington, KY: University of Kentucky
Press, 1996).
—— , Malcolm Richardson, and Jane L. Fisher (eds.), An Anthology of Chancery
English (Knoxville, TN: University of Tennessee Press, 1984).
Fishman, J. A. (ed.), Advances in Language Planning (The Hague: Mouton, 1974).
Forster, Peter G., The Esperanto Movement (The Hague: Mouton, 1982).
Fought, Carmen, Chicano English in Context (New York: Palgrave, 2003).
Fyler, John, Language and the Declining world in Chaucer, Dante, and Jean de Meun
(Cambridge: Cambridge University Press, 2007).
Gazophylacium Anglicanum (1689; rpt. Menston: Scolar Press, 1969).
Gellrich, Jesse M., The Idea of the Book in the Middle Ages: Language Theory, Myth-
ology, and Fiction (Ithaca, NY: Cornell University Press, 1985).
Gerald of Wales, The Jewel of the Church, ed. and trans. John J. Hagen (Leiden:
E. J. Brill, 1979).
Ghosh, Kantik, ‘Manuscripts of Nicholas Love’s The Mirror of the Blessed Life of Jesus
Christ and WycliYte Notions of ‘‘Authority’’ ’, in Felicity Riddy (ed.), Prestige,
Authority, and Power in Late Medieval Manuscripts and Texts (Woodbridge:
D. S. Brewer, 2000), 17–34.
Gibson, Campbell and Kay Jung, ‘Historical Census Statistics on Population Totals by
Race, 1790 to 1990, and by Hispanic Origin, 1970 to 1990, for the United States,
Regions, Divisions, and States’, Population Division, US Census Bureau, September
2002, Working Paper Series No. 56.
Gill, Alexander, Logonomia Anglica, 2nd edn. (1621; Menston: Scolar Press, 1968).
Goddard, Ives, ‘The Use of Pidgins and Jargons on the East Coast of North America’,
in Gray and Fiering (eds.), The Language Encounter in the Americas, 1492–1800,
61–78.
Gonzalez, Andrew B., ‘When Does an Error Become a Distinctive Feature of
Philippine English?’, in Noss (ed.), Varieties of English in Southeast Asia, 150–72.
González, Roseann Dueñas and Ildikó Melis (eds.), Language Ideologies: Critical
Perspectives on the OYcial English Movement, i, Education and the Social Implica-
tions of OYcial Language (Urbana, IL: NCTE, 2000).
274 Language Anxiety

Gordon, Elizabeth et al., New Zealand English: Its Origins and Evolution (Cambridge:
Cambridge University Press, 2004).
Görlach, Manfred, Aspects of the History of English (Heidelberg: Carl Winter, 1999).
—— Explorations in English Historical Linguistics (Heidelberg: Carl Winter, 2002).
Gough, Daniel, ‘Black English in South Africa’, in de Klerk (ed.), Focus on South Africa,
53–77.
Gower, John, The English Works of John Gower, ed. G. C. Macaulay, EETS es 81
(London: Oxford University Press, 1900).
Grace, Patricia, Baby No-Eyes (Honolulu, HI: University of Hawaii Press, 1998).
Graddol, David, The Future of English?: A Guide to Forecasting the Popularity of the
English Language in the 21st Century (London: British Council, 1997).
Grafton, Anthony, with April Shelford and Nancy Siraisi, New Worlds, Ancient Texts:
The Power of Tradition and the Shock of Discovery (Cambridge, MA: Harvard
University Press, 1992).
Gray, Edward G., New World Babel: Languages and Nations in Early America
(Princeton, NJ: Princeton University Press, 1999).
—— and Norman Fiering (eds.), The Language Encounter in the Americas, 1492–1800:
A Collection of Essays (New York: Berghahn Books, 2000).
Greenberg, Joseph H., Indo-European and Its Closest Relatives: The Eurasiatic
Language Family (Stanford, CA: Stanford University Press, 2000).
Greymorning, Stephen, ‘ReXections on the Arapaho Language Project, or
When Bambi Spoke Arapaho and Other Tales of Arapaho Language Revitalization
EVorts’, in Hinton and Hale (eds.), The Green Book of Language Revitalization in
Practice, 287–97.
Grillo, R. D., Dominant Languages: Language and Hierarchy in Britain and France
(Cambridge: Cambridge University Press, 1989).
Gumperz, John J. and Dell Hymes (eds.), Directions in Sociolinguistics: The Ethnography
of Communication (New York: Holt, Rinehart and Winston, Inc., 1972).
Gunn, John, ‘Social Contexts in the History of Australian English’, Machan and
Scott (eds.), English in Its Social Contexts, 204–29.
Halliday, M. A. K., Language as Social Semiotic: The Social Interpretation of Language
and Meaning (London: Edward Arnold, 1978).
Hargreaves, Henry, ‘The WycliYte Versions’, in G. W. H. Lampe (ed.), The Cambridge
History of the Bible, ii, The West from the Fathers to the Reformation (Cambridge:
Cambridge University Press, 1969), 387–415.
Harris, Roy, The Language Machine (Ithaca, NY: Cornell University Press, 1987).
Harrison, K. David, When Languages Die: The Extinction of the World’s Languages
and the Erosion of Human Knowledge (Oxford: Oxford University Press, 2007).
Hart, John, An Orthographie (1569; rpt. Menston: Scolar Press, 1969).
Haugen, Einar, Language ConXict and Language Planning: The Case of Modern
Norwegian (Cambridge, MA: Harvard University Press, 1966).
—— ‘Language, Dialect, Nation’, in The Ecology of Language, ed. Anwar S. Dil
(Stanford, CA: Stanford University Press, 1972), 237–54.
 Works Cited 275

Hechter, Michael, Internal Colonialism: The Celtic Fringe in British National

Development (New Brunswick, NJ: Transaction Publishers, 1999).
Herriman, Michael, ‘Language Policy in Australia’, in Herriman and Barbara Burnaby
(eds.), Language Policies in English-Dominant Countries, 35–61.
—— and Barbara Burnaby (eds.), Language Policies in English-Dominant Countries:
Six Case Studies (Clevedon: Multilingual Matters, 1996).
Higham, John, Strangers in the Land: Patterns of American Nativism, 1860–1925,
2nd edn. (New Brunswick, NJ: Rutgers University Press, 1988).
Hinton, Leanne and Ken Hale (eds.), The Green Book of Language Revitalization
in Practice (San Diego, CA: Academic Press, 2001).
Hoban, Russell, Riddley Walker: A Novel (New York: Summit Books, 1980).
Hodge, Robert and Gunther Kress, Language as Ideology, 2nd edn. (London:
Routledge, 1993).
Holm, John A., ‘English in the Caribbean’, in BurchWeld (ed.), English in Britain and
Overseas: Origins and Development, 328–81.
Homer, The Odyssey, trans. Richmond Lattimore (New York: HarperCollins, 1963).
Honey, John, Does Accent Matter? The Pygmalion Factor (London: Faber and
Faber, 1989).
—— Language Is Power: The Story of Standard English and Its Enemies (London: Faber
and Faber, 1997).
Howe, John R., Language and Political Meaning in Revolutionary America (Amherst,
MA: University of Massachusetts Press, 2004).
Hudson, Anne (ed.), Selections from English WycliYte Writings (Cambridge: Cam-
bridge University Press, 1978).
—— The Premature Reformation: WycliYte Texts and Lollard History (Oxford:
Clarendon, 1988).
—— (ed.), Two WycliYte Texts, EETS 301 (Oxford: Oxford University Press, 1993).
Hudson, R. A., Sociolinguistics, 2nd edn. (Cambridge: Cambridge University Press,
1996).
Hughes, Shaun F. D., ‘Was There Ever a ‘‘Māori English’’?’, World Englishes 23 (2004),
565–84.
Hymes, Del, Foundations in Sociolinguistics: An Ethnographic Approach (Philadelphia,
PA: University of Pennsylvania Press, 1974).
Ihalainen, Ossi, ‘The Dialects of England Since 1776’, in BurchWeld (ed.), English in
Britain and Overseas: Origins and Development, 197–274.
Isidore of Seville, Etymologiarum sive Originum Libri XX, ed. W. M. Lindsay, 2 vols.,
2nd edn. (Oxford: Clarendon Press, 1962).
‘It’s English or Tongues of ‘‘Ghetto’’ Life, Gingrich Says’, Milwaukee Journal-Sentinel,
1 April 2007, 19A.
Jager, Eric, The Tempter’s Voice: Language and the Fall in Medieval Literature (Ithaca,
NY: Cornell University Press, 1993).
Jenkins, Sally, The Real All Americans: The Team that Changed a Game, a People, a
Nation (New York: Doubleday, 2007).
276 Language Anxiety

Jespersen, Otto, Language: Its Nature, Development and Origin (London: G. Allen and
Unwin, 1922).
Johnson, Samuel, The Plan of a Dictionary of the English Language (London: J. and
P. Knapton, 1747).
—— A Dictionary of the English Language (1755; rpt. New York: AMS Press, 1967).
Jones, Daniel, An English Pronouncing Dictionary (London: J. M. Dent and Sons, 1917).
Jones, Mari C., Language Obsolescence and Revitalization: Linguistic Change in Two
Contrasting Welsh Communities (New York: Oxford University Press, 1998).
Joseph, Brian D. et al. (eds.), When Languages Collide: Perspectives on Language
ConXict, Language Competition, and Language Coexistence (Columbus, OH: Ohio
State University Press, 2003).
Kachru, Braj B., ‘The Second Diaspora of English’, in Machan and Scott (eds.), English
in Its Social Contexts, 230–52.
—— ‘English in South Asia’, in BurchWeld (ed.), English in Britain and Overseas:
Origins and Development, 497–553.
—— ‘English as an Asian Language’, in Bautista (ed.), English Is an Asian Language,
1–23.
—— Asian Englishes: Beyond the Canon (Hong Kong: Hong Kong University Press,
2005).
Karttunen, Frances, ‘Interpreters Snatched from the Shore: The Successful and the
Others’, in Gray and Fiering (eds.), The Language Encounter in the Americas, 215–29.
Keller, Rudi, On Language Change: The Invisible Hand in Language, trans. Brigitte
Nerlich (London: Routledge, 1994).
Kelman, James, How Late It Was, How Late (London: Secker and Warburg, 1994).
Kennedy, Randall, Nigger: The Strange Career of a Troublesome Word (New York:
Pantheon Books, 2002).
Kilgannon, Corey, ‘Tolerance for a Racial Slur Is a Test for Potential Jurors’, New York
Times, 18 May 2006, C17.
—— ‘Attacker Guilty of Hate Crimes in Howard Beach’, New York Times, 10
June 2006, B1.
Kim, Lisa (ed.), Singapore English: A Grammatical Description (Amsterdam:
John Benjamins, 2004).
King, Jeanette, ‘Te Kōhanga Reo: Māori Language Revitalization’, in Hinton and Hale
(eds.), The Green Book of Language Revitalization in Practice, 119–28.
Knighton, Henry, Knighton’s Chronicle, 1337–1396, ed. and trans. G. H. Martin
(Oxford: Clarendon Press, 1995).
Kramer, Michael P., Imagining Language in America: From the Revolution to the Civil
War (Princeton, NJ: Princeton University Press, 1992).
Krauss, Michael E., ‘Keynote—Mass Language Extinction and Documentation:
The Race Against Time’, in Osahito Miyaoka et al. (eds.), The Vanishing Languages
of the PaciWc Rim (Oxford: Oxford University Press, 2007), 3–24.
Kynaston, Sir Francis (ed. and trans.), Amorum Troili et Creseidae Libri duo priores
Anglico-Latini (London: John LichWeld, 1635).
 Works Cited 277

Labov, William,‘On the Mechanism of Linguistic Change’, in Gumperz and Hymes

(eds.), Directions in Sociolinguistics: The Ethnography of Communication, 512–38.
—— Sociolinguistic Patterns (Philadelphia, PA: University of Pennsylvania Press,
1972).
—— ‘On the Use of the Present to Explain the Past’, in Philip Baldi and Ronald
N. Werth (eds.), Readings in Historical Phonology: Chapters in the Theory of Sound
Change (University Park, PA: Pennsylvania State University Press, 1978), 275–312.
—— Principles of Linguistic Change, i, Internal Factors (Oxford: Blackwell, 1994).
—— Principles of Linguistic Change, ii, Social Factors (Oxford: Blackwell, 2001).
Laing, Margaret (ed.), Middle English Dialectology: Essays on Some Principles and
Problems (Aberdeen: Aberdeen University Press, 1989).
Laird, Charlton, The Miracle of Language (New York: Fawcett, 1953).
Lalla, Barbara and Jean D’Costa, Language in Exile: Three Hundred Years of Jamaican
Creole (Tuscaloosa, AL: University of Alabama Press, 1990).
Langland, William, Piers Plowman: The B Version, rev. edn., ed. George Kane and
E. Talbot Donaldson (Berkeley and Los Angeles, CA: University of California Press,
1988).
Lass, Roger, On Explaining Language Change (Cambridge: Cambridge University
Press, 1980).
—— ‘Where Do Extraterrestrial Englishes Come from? Dialect, Input, and RecodiW-
cation in Transported Englishes’, in Sylvia Adamson et al. (eds.), Papers from the
5th International Conference in English Historical Linguistics (Amsterdam: John
Benjamins, 1990), 245–80.
—— Historical Linguistics and Language Change (Cambridge: Cambridge University
Press, 1997).
—— (ed.), The Cambridge History of the English Language, iii, 1476–1776 (Cambridge:
Cambridge University Press, 1999).
—— ‘Phonology and Morphology’, in Lass (ed.), 1476–1776, 56–186.
—— ‘South African English’, in Rajend Mesthrie (ed.), Language in South Africa
(Cambridge: Cambridge University Press, 2002), 104–26.
Law, Vivien, The History of Linguistics in Europe from Plato to 1600 (Cambridge:
Cambridge University Press, 2003).
Leach, Alfred, The Letter H Past, Present, and Future: A Treatise, with Rules for the
Silent H Based on Modern Usage, and Notes on Wh (London: GriYth and Farran,
1880).
Leland, Charles G., Pidgin-English Sing-Song (London: Trübner, 1876).
Leonard, Sterling Andrus, The Doctrine of Correctness in English Usage, 1700–1800
(1929; rpt. New York: Russell and Russell, 1962).
Lerer, Seth, Inventing English: A Portable History of the Language (New York: Columbia
University Press, 2007).
Lightfoot, David, The Development of Language: Acquisition, Change, and Evolution
(Malden, MA: Blackwell, 1999).
278 Language Anxiety

Lippi-Green, Rosina, English with an Accent: Language, Ideology, and Discrimination in

the United States (London: Routledge, 1997).
Locke, John, An Essay Concerning Human Understanding, ed. Peter H. Nidditch
(Oxford: Clarendon Press, 1975).
Looby, Christopher, Voicing America: Language, Literary Form, and the Origins of the
United States (Chicago, IL: University of Chicago Press, 1996).
Lusignan, Serge, Parler vulgairement: les intellectuels et la langue française aux xiii e et
xiv e siècles, 2nd edn. (Paris: J. Vrin, 1987).
Lydgate, John, Troy Book, ed. Henry Bergen, EETS es 97, 103, 106, 126 (London: Kegan
Paul, 1906–35).
Machan, Tim William, ‘Kynaston’s Troilus, Textual Criticism, and the Renaissance
Reading of Chaucer’, Exemplaria 5 (1993), 161–83.
—— ‘Language Contact in Piers Plowman’, Speculum, 69 (1994), 359–85.
—— ‘Thomas Berthelette and Gower’s Confessio’, Studies in the Age of Chaucer, 18
(1996), 143–66.
—— ‘Language and Society in Twelfth-Century England’, in Taavitsainen et al. (eds.),
Placing Middle English in Context, 43–66.
—— ‘Politics and the Middle English Language’, Studies in the Age of Chaucer,
24 (2002), 317–24.
—— English in the Middle Ages (Oxford: Oxford University Press, 2003).
—— ‘Medieval Multilingualism and Gower’s Literary Practice’, Studies in Philology,
103 (2006), 1–25.
—— (ed.), Chaucer’s ‘Boece’, Middle English Texts 38 (Heidelberg: Carl Winter, 2008).
—— and Charles T. Scott (eds.), English in Its Social Contexts: Essays in Historical
Sociolinguistics (New York: Oxford University Press, 1992).
MacMahon, Michael K. C., ‘Phonology’, in The Cambridge History of the English
Language, iv, Suzanne Romaine (ed.), 1776–1997 (Cambridge: Cambridge Univer-
sity Press, 1998), 373–535.
MacWhinney, Brian, ‘Emergent Approaches to Language’, in Bybee and Hopper
(eds.), Frequency and the Emergence of Linguistic Structure, 449–70.
‘Major EVort Is Under Way to Revive and Preserve Hawaii’s Native Tongue’, New York
Times, 15 April 2007, A18.
Manning, H. Paul, ‘The Rock Does not Understand English: Welsh and the Division
of Labor in Nineteenth-Century Gwynedd Slate Quarries’, in Joseph et al. (eds.),
When Languages Collide, 45–70.
Marshall, Jonathan, Language Change and Sociolinguistics: Rethinking Social Networks
(New York: Palgrave Macmillan, 2004).
May, Stephen, Language and Minority Rights: Ethnicity, Nationalism and the Politics
of Language (London: Longman, 2001).
Mazrui, Alamin M., English in Africa: After the Cold War (Clevedon: Multilingual
Matters, 2004).
McArthur, Tom, The English Languages (Cambridge: Cambridge University Press,
1998).
 Works Cited 279

McClure, J. Derrick, ‘English in Scotland’, in BurchWeld (ed.), English in Britain and

Overseas: Origins and Development, 23–93.
McEnery, Tony, Swearing in English: Bad Language, Purity and Power from 1586 to the
Present (London: Routledge, 2006).
McKenzie, Donald, Oral Culture, Literacy and Print in Early New Zealand: The Treaty
of Waitangi (Wellington: Victoria University Press, 1985).
McLarin, Kimberly J., ‘To Preserve Afrikaners’ Language, Mixed-Race South Africans
Join Fray’, New York Times, 28 June 1995, A4.
McWhorter, John H., Doing Our Own Thing: The Degradation of Language and Music
and Why We Should, Like, Care (New York: Gotham, 2003).
Meyers, Walter E., Aliens and Linguists: Language Study and Science Fiction (Athens,
GA: University of Georgia Press, 1980).
Michael, Ian, English Grammatical Categories and the Tradition to 1800 (Cambridge:
Cambridge University Press, 1970).
Michel, Dan, Ayenbite of Inwyt, ed. Richard Morris, EETS os 23 (London: Trübner,
1866).
Millar, Robert McColl, Language, Nation and Power: An Introduction (New York:
Palgrave, 2005).
Milroy, James, ‘Middle English Dialectology’, in Blake (ed.), 1066–1476, 156–206.
—— Linguistic Variation and Change: On the Historical Sociolinguistics of English
(Oxford: Blackwell, 1992).
—— ‘The Consequences of Standardization in Descriptive Linguistics’, in Bex and
Watts (eds.), Standard English: The Widening Debate, 16–39.
—— ‘The Legitimate Language: Giving a History to English’, in Watts and Trudgill
(eds.), Alternative Histories of English, 7–25.
—— and Lesley Milroy, Authority in Standard Language: Investigating Standard
English, 3rd edn. (London: Routledge, 1999).
Milroy, Lesley, Language and Social Networks, 2nd edn. (Oxford: Blackwell, 1987).
—— and Pieter Muysken (eds.), One Speaker, Two Languages: Cross-disciplinary
Perspectives on Code-Switching (Cambridge: Cambridge University Press, 1995).
Milton, John, Complete Prose Works, 8 vols., gen. ed. Don M. Wolfe (New Haven,
CT: Yale University Press, 1953–82).
—— The Poems of John Milton, ed. Roy Flannagan (Boston, MA: Houghton MiZin,
1998).
Mitchell, Linda C., Grammar Wars: Language as Cultural BattleWeld in 17th and
18th Century England (Aldershot: Ashgate, 2001).
Montagu, Ashley, The Anatomy of Swearing, 2nd edn. (London: Macmillan, 1973).
Morales, Ed, Living in Spanglish: The Search for Latino Identity in America (New York:
St Martin’s Press, 2002).
Morton, Herbert C., The Story of ‘Webster’s Third’: Philip Gove’s Controversial
Dictionary and Its Critics (Cambridge: Cambridge University Press, 1994).
Mufwene, Salikoko S., The Ecology of Language Evolution (Cambridge: Cambridge
University Press, 2001).
280 Language Anxiety

Mufwene, Salikoko S., ‘‘African-American English’, in Algeo (ed.), English in North

America, 291–324.
—— ‘Language Endangerment: What Have Pride and Prestige Got to Do with It?’,
in Joseph et al. (eds.), When Languages Collide, 324–45.
Mugglestone, Lynda, ‘Talking Proper’: The Rise of Accent as Social Symbol, 2nd edn.
(Oxford: Oxford University Press, 2003).
—— Lost for Words: The Hidden History of the Oxford English Dictionary (New Haven,
CT: Yale University Press, 2005).
Mühlhäusler, Peter, Pidgin & Creole Linguistics (Oxford: Blackwell, 1986).
—— Linguistic Ecology: Language Change and Linguistic Imperialism in the PaciWc
Region (London: Routledge, 1996).
Mulcaster, Richard, The First Part of the Elementarie (1582; rpt. Menston: Scolar Press,
1970).
Murray, Sir James A. H. et al., A New English Dictionary on Historical Principles,
10 vols. (Oxford: Clarendon Press, 1888–1928).
Nauerby, Tom, No Nation Is an Island: Language, Culture, and National Identity in the
Faroe Islands (Aarhus: Aarhus University Press, 1996).
Navarro, Mireya, ‘Breaking the Sound Barrier’, New York Times, 23 July 2006, 9, 1
and 10.
Nettle, Daniel, Linguistic Diversity (Oxford: Oxford University Press, 1999).
—— and Suzanne Romaine, Vanishing Voices: The Extinction of the World’s Languages
(Oxford: Oxford University Press, 2000).
Noss, R. B. (ed.), Varieties of English in Southeast Asia (Singapore: Singapore Univer-
sity Press, 1983).
Nunberg, GeoVrey, Going Nucular: Language, Politics, and Culture in Confrontational
Times (New York: PublicAVairs, 2004).
Ohaeto, Ezenwa, Chinua Achebe: A Biography (Bloomington, IN: Indiana University
Press, 1997).
Orsman, H. W. (ed.), The Dictionary of New Zealand English: A Dictionary of
New Zealandisms on Historical Principles (Auckland: Oxford University Press, 1997).
Ozolins, Uldis, The Politics of Language in Australia (Cambridge: Cambridge Univer-
sity Press, 1993).
Padley, G. A., Grammatical Theory in Western Europe, 1500–1700: Trends in Vernacular
Grammar, 2 vols. (Cambridge: Cambridge University Press, 1985 and 1988).
Page, R. I., ‘How Long Did the Scandinavian Language Survive in England? The
Epigraphical Evidence’, in Peter Clemoes and Kathleen Hughes (eds.), England
Before the Conquest: Studies in Primary Sources Presented to Dorothy Whitelock
(Cambridge: Cambridge University Press, 1971), 165–81.
Palmer, Patricia, Language and Conquest in Early Modern Ireland: English Renaissance
Literature and Elizabethan Imperial Expansion (Cambridge: Cambridge University
Press, 2001).
 Works Cited 281

Parkes, M. B. and Richard Beadle (eds.), The Poetical Works of GeoVrey Chaucer:
A Facsimile of Cambridge University Library MS Gg. 4.27, 3 vols. (Norman,
OK: Pilgrim Books, 1979–80).
Pearsall, Derek, ‘Hoccleve’s Regement of Princes: The Poetics of Royal Self-
Representation’, Speculum, 69 (1994), 386–410.
Pecock, Reginald, The Repressor of Over Much Blaming of the Clergy, ed. Churchill
Babington, Rolls Series 19, 2 vols. (London: Longman, 1860).
—— The Reule of Crysten Religioun, ed. William Cabell Greet, EETS os 171 (London:
Oxford University Press, 1927).
Pennycook, Alastair, The Cultural Politics of English as an International Language
(London: Longman, 1994).
—— English and the Discourses of Colonialism (London: Routledge, 1998).
Perry, Theresa and Lisa Delpit (eds.), The Real Ebonics Debate: Power, Language, and
the Education of African-American Children (Boston, MA: Beacon Press, 1998).
Phaswana, Nkhelebeni, ‘Contradiction or Assimilation? The South African Language
Policy, and the South African National Government’, in Sinfree Makoni et al. (eds.),
Black Linguistics: Language, Society, and Politics in Africa and the Americas (London:
Routledge, 2003), 117–31.
Phillipson, Robert, Linguistic Imperialism (Oxford: Oxford University Press, 1992).
—— English-Only Europe? Challenging Language Policy (London: Routledge, 2003).
Pinker, Stephen,‘The Game of the Name’, New York Times, 5 April 1994, A21.
Poole, Austin Lane, From Domesday Book to Magna Carta 1087–1216, 2nd edn.
(Oxford: Clarendon Press, 1955).
Poplack, Shana, Gerard von Herk, and Down Harvie, ‘ ‘‘Deformed in the Dialects’’: An
Alternative History of Non-Standard English’, in Watts and Trudgill (eds.), Alter-
native Histories of English, 87–110.
Preston, Dennis R., Perceptual Dialectology: Nonlinguists’ Views of Areal Linguistics
(Dordrecht: Foris Publicatons, 1989).
—— ‘The Story of Good and Bad English in the United States’, in Watts and Trudgill
(eds.), Alternative Histories of English, 134–51.
Price, Glanville, The Languages of Britain (London: Edward Arnold, 1984).
Prokosch, Eduard, A Comparative Germanic Grammar (Philadelphia, PA: Linguistics
Society of America, 1939).
Pryce, Malcolm, Last Tango in Aberystwyth (London: Bloomsbury, 2003).
Puttenham, George, The Arte of English Poesie (1589; rpt. Menston: Scolar Press, 1968).
Quirk, Randolph, ‘Language Variety: Nature and Art’, in Noss (ed.), Varieties of
English in Southeast Asia, 3–19.
—— ‘Language Varieties and Standard Language’, English Today: The International
Review of the English Language, 21 (1990), 3–10.
Ramson, W. S. (ed.), The Australian National Dictionary: A Dictionary of Australian-
isms on Historical Principles (Melbourne: Oxford University Press, 1988).
Ray, John, A Collection of English Words (1691; rpt. Menston: Scolar Press, 1969).
282 Language Anxiety

Read, Allen Walker, Milestones in the History of English in America, ed. Richard
W. Bailey (Durham, NC: Duke University Press, 2002).
—— ‘The Allegiance to Dictionaries in American Linguistic Attitudes’, in Milestones
in the History of English in America, ed. Bailey, 110–20.
—— ‘Amphi-Atlantic English’, in Milestones in the History of English in America, ed.
Bailey, 55–82.
—— ‘British Recognition of American Speech in the Eighteenth Century’, in
Milestones in the History of English in America, ed. Bailey, 37–54.
—— ‘Milestones in the Branching of British and American English’, in Milestones in
the History of English in America, ed. Bailey, 3–21.
Reddick, Allen, The Making of Johnson’s Dictionary, 1746–1773 (Cambridge: Cambridge
University Press, 1990).
Ricento, Thomas (ed.), Ideology, Politics and Language Policies: Focus on English
(Amsterdam: John Benjamins, 2000).
Richard of Devizes, The Chronicle of Richard of Devizes of the Time of King Richard
the First, ed. and trans. John T. Appleby (London: Thomas Nelson and Sons, 1963).
Richards, I. A., Basic English and Its Uses (London: Kegan Paul, 1943).
Richler, Howard, A Bawdy Language: How a Second-Rate Language Slept Its Way to
the Top (Toronto: Stoddart, 1999).
Rickford, John R., Dimensions of a Creole Continuum: History, Texts & Linguistic
Analysis of Guyanese Creole (Stanford, CA: Stanford University Press, 1987).
Riddy, Felicity (ed.), Regionalism in Late Medieval Manuscripts and Texts:
Essays Celebrating the Publication of ‘A Linguistic Atlas of Late Mediaeval England’
(Woodbridge: D. S. Brewer, 1991).
Ritt, Nikolaus, SelWsh Sounds and Linguistic Evolution: A Darwinian Approach to
Language Change (Cambridge: Cambridge University Press, 2004).
Rodriguez, Richard, Hunger of Memory: The Education of Richard Rodriguez:
An Autobiography (Boston, MA: D. R. Godine, 1982).
Rohter, Larry, ‘Repeal Is Likely for ‘‘English Only’’ Policy in Miami’, New York Times,
14 May 1993, A7.
—— ‘Learn English, Says Chile, Thinking Upwardly Global’, New York Times, 29
December 2004, A4.
Romaine, Suzanne, Socio-historical Linguistics: Its Status and Methodology
(Cambridge: Cambridge University Press, 1982).
—— ‘Contact with Other Languages’, in Algeo (ed.), English in North America,
154–83.
Sampson, George, English for the English: A Chapter on National Education (1921;
rpt. with introduction by Denys Thompson, Cambridge: Cambridge University
Press, 1970).
Samuels, M. L., Linguistic Evolution, with Special Reference to English (Cambridge:
Cambridge University Press, 1972).
—— ‘The Great Scandinavian Belt’, in Laing (ed.), Middle English Dialectology,
106–15.
 Works Cited 283

—— ‘Some Applications of Middle English Dialectology’, in Laing (ed.), Middle

English Dialectology, 64–80.
Sapir, Edward, Language: An Introduction to the Study of Speech (New York: Harcourt-
Brace, 1921).
Schlesinger, Jr., Arthur M., The Disuniting of America: ReXections on a Multicultural
Society, rev. edn. (New York: W. W. Norton, 1998).
Schmid, Carol L., The Politics of Language: ConXict, Identity and Cultural Pluralism
in Comparative Perspective (Oxford: Oxford University Press, 2001).
Schmidt, Sr., Ronald, Language Policy and Identity Politics in the United States
(Philadelphia, PA: Temple University Press, 2000).
Schneider, Edgar W., Postcolonial English: Varieties around the World (Cambridge:
Cambridge University Press, 2007).
Schütz, Albert J., The Voices of Eden: A History of Hawaiian Language Studies
(Honolulu, HI: University of Hawaii Press, 1994).
Senior, Jennifer, ‘Sign Language ReXects Changing Sensibilities’, New York Times, 3
January 1994, A1, 8.
Shakespeare, William, The Complete Works, ed. Alfred Harbage (London: Penguin,
1969).
Sharp, Henry (ed.), Selections from Education Records, i, 1781–1839 (Calcutta:
Superintendent Government Printing, 1920).
Shaw, George Bernard, Complete Plays with Prefaces, 6 vols. (New York: Dodd, Mead,
1962).
Sheridan, Thomas, A Course of Lectures on Elocution (1762; rpt. Menston: Scolar Press,
1968).
Silva, Penny (ed.), A Dictionary of South African English on Historical Principles
(Oxford: Oxford University Press, 1996).
Simon, John, Paradigms Lost (New York: Potter, 1980).
Simpson, David, The Politics of American English, 1776–1850 (New York: Oxford
University Press, 1986).
Skutnabb-Kangas, Tove, ‘Linguistic Diversity and Biodiversity: The Threat from Killer
Languages’, in Christian Mair (ed.), The Politics of English as a World Language: New
Horizons in Postcolonial Cultural Studies (Amsterdam: Rodopi, 2003), 31–52.
—— and Robert Phillipson (eds.), Linguistic Human Rights: Overcoming Linguistic
Discrimination (Berlin: Mouton de Gruyter, 1994).
Sledd, James and Wilma R. Ebbitt (eds.), Dictionaries and THAT Dictionary:
A Casebook on the Aims of Lexicographers and the Targets of Reviewers (Chicago,
IL: Scott, Foresman, 1962).
Smith, A. H. (ed.), Three Northumbrian Poems (London: Methuen, 1933).
Smith, Jeremy J., An Historical Study of English: Function, Form and Change (London:
Routledge, 1996).
—— ‘Standard Language in Early Middle English?’, in Taavitsainen et al. (eds.),
Placing Middle English in Context, 125–39.
284 Language Anxiety

Smith, Jeremy J., Sound Change and the History of English (Oxford: Oxford University
Press, 2007).
Smitherman, Geneva, Word from the Mother: Language and African Americans
(New York: Routledge, 2006).
Sontag, Deborah, ‘Oy Gevalt! New Yawkese An Endangered Dialect?’, New York Times,
14 February 1993, A1, 18.
St Augustine, Confessions, trans. R. S. Pine-CoYn (Harmondsworth: Penguin, 1961).
—— Tractates on the Gospel of John 1–10, trans. John W. Rettig, i of The Fathers of the
Church: A New Translation (Washington, DC: The Catholic University of America
Press, 1988).
—— De Doctrina Christiana, ed. and trans. R. P. H. Green (Oxford: Clarendon Press,
1995).
—— The City of God against the Pagans, ed. and trans. R. W. Dyson (Cambridge:
Cambridge University Press, 1998).
Starnes, De Witt T. and Gertrude E. Noyes, The English Dictionary from Cawdrey
to Johnson, 1604–1755 (Chapel Hill, NC: University of North Carolina Press, 1946).
Stavans, Ilan, Spanglish: The Making of a New American Language (New York: Rayo,
2003).
Strang, Barbara, A History of English (London: Methuen, 1970).
‘Surrender of Hostile Comanche Chiefs and Thirty-Six Warriors’, New York Times,
23 April 1875, 1.
Swanton, Michael (ed. and trans.), Anglo-Saxon Prose (London: Dent, 1975).
Sweet, Henry, A Handbook of Phonetics (Oxford: Clarendon Press, 1877).
Taavitsainen, Irma et al. (eds.), Placing Middle English in Context: Selected Papers from
the Second Middle English Conference (Berlin: Mouton de Gruyter, 2000).
Tagliabue, John, ‘In Europe, Going Global Means, Alas, English’, New York Times,
19 May 2002, A15.
—— ‘Soon, Europe Will Speak in 23 Tongues’, New York Times, 6 December 2006,
A10.
Taylor, Dennis, Hardy’s Literary Language and Victorian Philology (Oxford:
Clarendon, 1993).
Thomas, Hugh M., The English and the Normans: Ethnic Hostility, Assimilation, and
Identity 1066–c1220 (Oxford: Oxford University Press, 2003).
Thomason, Sarah Grey and Terrence Kaufman, Language Contact, Creolization,
and Genetic Linguistics (Berkeley, CA: University of California Press, 1988).
Thompson, Linda, Michael Fleming, and Michael Bryam, ‘Languages and Language
Policy in Britain’, in Herriman and Burnaby (eds.), Language Policies in English-
Dominant Countries: Six Case Studies, 99–121.
Thompson, Roger M., Filipino English and Taglish: Language Switching from Multiple
Perspectives (Amsterdam: John Benjamins, 2003).
‘Three Parties of Indian Police are After Swift Bear and His Band—An Uprising
Feared’, New York Times, 15 July 1899, 1.
 Works Cited 285

Titlestad, Peter, ‘English, the Constitution and South Africa’s Language Future’, in de
Klerk (ed.), Focus on South Africa, 163–73.
Todd, Loreto, Modern Englishes: Pidgins and Creoles (Oxford: Blackwell, 1984).
Tollefson, James W., Planning Language, Planning Inequality: Language Policy in the
Community (London: Longman, 1991).
Toon, Thomas E., ‘The Social and Political Contexts of Language Change in Anglo-
Saxon England’, in Machan and Scott (eds.), English in Its Social Contexts, 28–46.
Trudgill, Peter, The Social DiVerentiation of English in Norwich (Cambridge:
Cambridge University Press, 1974).
—— Sociolinguistic Variation and Change (Edinburgh: Edinburgh University Press,
2002).
Truss, Lynne, Eats, Shoots & Leaves: The Zero Tolerance Approach to Punctuation
(New York: Gotham, 2004).
Turner, George W., ‘English in Australia’, in BurchWeld (ed.), English in Britain and
Overseas, 277–327.
Twain, Mark, Adventures of Huckleberry Finn, ed. Henry Nash Smith
(Boston, MA: Houghton MiZin, 1958).
Unger, Harlow Giles, Noah Webster: The Life and Times of an American Patriot
(New York: John Wiley and Sons, 1998).
‘Universal Declaration of Linguistic Rights’ at [Link]
[Link].
Usk, Thomas, Testament of Love, in Chaucerian and Other Pieces, vii, Complete Works
of GeoVrey Chaucer, ed. W. W. Skeat (Oxford: Oxford University Press, 1897).
Verstegan, Richard, A Restitution of Decayed Intelligence (1605; rpt. Ilkey: Scolar Press,
1976).
Vitalis, Orderic, The Ecclesiastical History of Orderic Vitalis, ed. and trans. Marjorie
Chibnall, 6 vols. (Oxford: Clarendon, 1969–80).
Voerding, Brian, ‘Aracadia Mayor: No Más’, La Crosse Tribune, 19 August (2006), A1, 2.
Von Humboldt, Wilhelm, On Language: The Diversity of Human Language-Structure
and Its InXuence on the Mental Development of Mankind, trans. Peter Heath
(Cambridge: Cambridge University Press, 1988).
Walker, John, A Critical Pronouncing Dictionary and Expositor of the English Language
(London: Robinson and Cadell, 1791).
Walker, Ranginui, Ka Whawhai Tonu Matou: Struggle without End (Auckland:
Penguin Books, 1990).
Waller, Edmund, The Poems of Edmund Waller, ed. G. Thorn Drury (London:
Lawrence and Bullen, 1893).
Wardhaugh, Ronald, Languages in Competition: Dominance, Diversity, and Decline
(Oxford: Blackwell, 1987).
Warner, Sam L. No’eau, ‘The Movement to Revitalize Hawaiian Language and
Culture’, in Hinton and Hale (eds.), The Green Book of Language Revitalization in
Practice, 133–44.
286 Language Anxiety

Watermeyer, Susan, ‘Afrikaans English’, in de Klerk (ed.), Focus on South Africa,

99–124.
Wa Thiongo, Ngũgı̃, Decolonising the Mind: The Politics of Language in African
Literature (London: James Currey, 1986).
Watts, Richard and Peter Trudgill (eds.), Alternative Histories of English (London:
Routledge, 2002).
Waugh, Evelyn, The Complete Stories of Evelyn Waugh (Boston, MA: Little, Brown,
and Co., 1998).
Webb, Vic, Language in South Africa: The Role of Language in National Transformation,
Reconstruction, and Development (Amsterdam: John Benjamins, 2002).
Webster, Noah, Dissertations on the English Language (1789; rpt. Menston: Scolar Press,
1967).
Weinreich, Uriel, William Labov, and Marvin I. Herzog, ‘Empirical Foundations for
a Theory of Language Change’, in W. P. Lehmann and Yakov Malkiel (eds.),
Directions for Historical Linguistics: A Symposium (Austin, TX: University of Texas
Press), 95–188.
Wells, H. G., The Shape of Things to Come (New York: Macmillan, 1933).
Wenzel, Siegfried, Macaronic Sermons: Bilingualism and Preaching in Late-Medieval
England (Ann Arbor, MI: University of Michigan Press, 1994).
Weyand, Alexander M., The Saga of American Football (New York: Macmillan, 1955).
Wheeler, Robert M., Jim Thorpe: World’s Greatest Athlete (Norman, OK: University of
Oklahoma Press, 1975).
White, Hayden, The Content of the Form: Narrative Discourse and Historical Repre-
sentation (Baltimore, MD: The Johns Hopkins University Press, 1987).
Whorf, Benjamin Lee, Language, Thought, and Reality: Selected Writings, ed. John
B. Carroll (Cambridge, MA: Technology Press of Massachusetts Institute of
Technology, 1956).
Wierzbicka, Anna, English: Meaning and Culture (Oxford: Oxford University Press,
2006).
Wilgoren, Jodi, ‘Divided by a Call for a Common Language: As Immigration Rises,
a Wisconsin County Makes English OYcial’, New York Times, 19 July 2002, A8.
Wilkins, John, An Essay towards a Real Character, and a Philosophical Language (1668;
rpt. Menston: Scolar Press, 1968).
William of Malmesbury, De Gestis PontiWcum Anglorum, ed. N. E. S. A. Hamilton,
Rolls Series 52 (London: Longman, 1870).
Williams, Joseph M., ‘ ‘‘O! When Degree is Shak’d’’: Sixteenth-Century Anticipations
of Some Modern Attitudes toward Usage’, in Machan and Scott (eds.), English in
Its Social Contexts, 69–101.
Willinsky, John, Empire of Words: The Reign of the OED (Princeton, NJ: Princeton
University Press, 1994).
Wilson, Thomas, The Arte of Rhetorique (1553; rpt. Gainesville, FL: Scholars’ Facsimiles
and Reprints, 1962).
 Works Cited 287

Withers, Phillip, Aristarchus, or the Principles of Composition, 2nd edn. (London:

J. Moore, 1789).
Wogan-Browne, Jocelyn et al. (eds.), The Idea of the Vernacular: An Anthology of
Middle English Literary Theory, 1280–1520 (University Park, PA: Pennsylvania State
University Press, 1999).
Wood, Nicholas, ‘In the Old Dialect, a Balkan Region Regaining Identity’, New York
Times, 24 February 2005, A4.
Woolley, Lisa, American Voices of the Chicago Renaissance (Dekalb, IL: Northern
Illinois University Press, 2000).
Wordsworth, William, The Complete Poetical Works, ed. A. J. George (Cambridge,
MA: Houghton MiZin, 1932).
Wright, Arnold, Baboo English as ’tis Writ: Being Curiosities of Indian Journalism
(London: T. Fisher Unwin, 1891).
Wright, Joseph, The English Dialect Grammar (Oxford: H. Frowde, 1905).
Wright, Laura, ‘Macaronic Writing in a London Archive, 1380–1480’, in Matti Rissanan
et al. (eds.), History of Englishes: New Methods and Interpretations in Historical
Linguistics (Berlin: Mouton de Gruyter, 1992), 762–70.
—— Sources of London English: Medieval Thames Vocabulary (Oxford: Clarendon
Press, 1996).
—— (ed.), The Development of Standard English, 1300–1800 (Cambridge: Cambridge
University Press, 2000).
Wright, Sue, Language Policy and Language Planning: From Nationalism to Globalisa-
tion (Basingstoke: Palgrave Macmillan, 2004).
Zezima, Katie, ‘One Man’s Goal: For a Tribe to Pray in Its Own Language’, New York
Times, 15 November 2003, A15.
Zhao, Yilu, ‘Wave of Pupils Lacking English Strains Schools’, New York Times, 5
August 2002, A1, 11.
This page intentionally left blank
 Index

Aasen, Ivar 140, 172 Arabic 92

Ablaut 31 Aramaic 64, 127
Aboriginal languages 19, 51, 127, 151, 152, Arapaho 137
157 Arnold, Matthew 65, 122
Académie Française 139, 183, 196, 200 Arundel, Archbishop Thomas 147
accent 1–2, 48, 193, 212–14, 247 Asian languages 25, 95
American accent 2 ask 39–40, 48
British accent 2 Atkins, J. D. C. 155
Achebe, Chinua 67, 123, 124 Austen, Jane 247
Act of Union of England and Australia 19, 68, 151, 160, 167, 223, 225
Wales 164–5 changes in linguistic repertoire 25, 51,
Act for the English Order 165 62, 140, 167, 184
Act for the Settling of Parochial Australian English, see English, varieties
Schools 165 of
Adamic language 89–92, 96–7 Australian Language and Literacy White
Adams, John 132, 177, 178, 189 Paper 167
Adventures of Huckleberry Finn, see
Twain, Mark bachelor 46
Africa 64, 67, 71, 156, 166, 222 Bacon, Roger 211
Afrikaans 73, 173 bad language 214–16
Aidan, Bishop of Iona 65 Bailey, Nathan 197
ain’t 188–9 Bailey, Richard 207, 233
Aitchison, Jean 5 Bambara 69
Alfred the Great 4, 23, 141, 147, 149, 150, Bambi 137
179, 181, 238, 245, 257 æÆæ  93, 95
translation program 141–4, 184 Bardic Grammar 94
Algonquian 50 Barlow, Joel 70
Allestree, Richard 215 Basic English 124, 207–8
Alnwich, Bishop William 148 Bateson, F. W. 11
Amendment of Orthograpie, see Bullokar, Battle of Hastings 199
William Bede, the Venerable 52–4, 55, 65–6, 93,
Amorum Troili et Creseidae Libri Duo 113, 143
Priores Anglio-Latini, see Kynaston, Belgium 62
Sir Francis Bell, Terrel 161
Anderson, Benedict 169 Beowulf 22, 78
Andrews, E. W. 224 Berthelette, Thomas 112
Aotearoa, see New Zealand bettong 51
290 Index

Bible, The 41, 76, 82, 84, 91, 94, 109, Amendment of Orthograpie 194
140–1, 142, 144–7, 153, 154, 156, 163, Pamphlet for Grammar 196
176, 188 BurchWeld, Robert 51
Acts 2:1–2 92–3 Burgess, Anthony 124, 126
Genesis 3:1 59–60 Burke, Edmund 202
Genesis 10:31 84, 87–8, 96, 98, 105, 124, Burling, Robbins 7 n.
126–8, 209 Butler, Susan 227–30
Genesis 11:1–9 83–5, 87–8, 98, 105, 124,
126, 128–9, 209 Camden, William 200, 201
Judges 12:5–6 5 Cameron, Deborah 18, 130, 133, 204–5,
Bickerton, Derek 7 234, 246
bilingual education 3, 160–1, 168, 232, Canada 62–3, 64, 68, 160, 223, 241
249–50 Canadian English, see English, varieties
Bilingual Education Act of the United of
States 160, 161, 250 Canterbury Tales, see Chaucer, GeoVrey
bilingualism 23, 72 carbs 18
billabong 51 Cardinal Richelieu 139
Biondi, Giovanni 259 Carew, Richard 202, 209–10, 212
bird 39 Caribbean, The 64
bitch 46 Carlisle Indian Industrial School 152, 153,
Black South African English, see English, 154, 155, 260
varieties of Cather, Willa 121
blog 18, 22 Cawdrey, Robert 206
BloomWeld, Leonard 75–6 A Table Alphabeticall 196, 216
Blount, Thomas 217 Caxton, William 56, 114, 115, 191, 199, 212,
Boethius 143, 148 247
Bolton, Kingsley 72 Celtic languages 3, 65, 97, 116, 140, 246
Book of Common Prayer 150 as original language 96
Bopp, Franz 18, 60, 98 shift from 66–7, 157, 158
Bosnian 127–8, 140 chairman 45
Boucher, Jonathon 70 Chaldean 92, 180
Brazil 129 Chambre, Major Alan 71
Breton 127, 210 Chancery English, see English, varieties
Breugel, Peter 86–7 of
Bridges, Robert 15 Charles I, King of England 113
Society for Pure English 223, 232 Chaucer, GeoVrey 35, 110, 131, 212
British Council 21 n. Boece 148
British East India Company 61, 156 Canterbury Tales 78
Brooklynese, see English, varieties of ‘Reeve’s Tale’ 114–16, 118, 125
Brown, Charles Brockden 177 Troilus and Criseyde 5, 111–12
Bullock Report 232 Wife of Bath’s Tale 31
Bullokar, John 210 Chicano English, see English, varieties of
Bullokar, William 195 ‘Children of the Crucible’ 183
 Index 291

Chile 129 Dalgarno, George 97

China 140, 166 Daniel, Samuel 61–2
Chinese 21, 128, 134, 166, 253, 254 Danish 12, 13, 133
Chinese English, see English, Dante Aligheri 193, 231
varieties of De Vulgari Eloquentia 90–2
Chomsky, Noam 11, 15, 59, 98, 185, Paradiso 96
207–8, 241 Davis, JeVerson 181
Churchill, William 226 deer 5
Cicero 193 Defoe, Daniel 196, 200, 217
cly 51 Denglish 72–4
Clyne, Michael 167 dental preterite, see weak verbs
Cockney, see English, varieties of Derrida, Jacques 10
code-switching 77, 123 Descartes, Rene 10, 203
codiWcation, see standardization and descriptive grammar 35–7, 42–3, 189,
prescriptive grammar 209; also see prescriptive grammar
Coggeshall, Ralph 108 and linguistics
cognates 40–1 Devizes, Richard 107
Collingwood, R. G. 239, 241–2 diachronic change 4, 241
colonization 40–1, 65–8, 70–1, 74, 87, Dial, The 121–2, 124
222–7, 233, 247 dialectology 57–8, 220–1
communicative competence 251–2 dialect translation 55, 113
Confessio Amantis, see Gower, John dialects 1–2, 12, 37, 54–9, 105–17, 210–11,
confusio linguarum 83, 91, 126 242, 244
Cook, Captain James 61, 153, 156 as distinct from language 4 n., 29
cooliman 51 dialect-writing 118–26
Cooper, James Fenimore 176, 178, 258, Dickens, Charles 57, 117–18, 123, 124, 159,
259 234
Cooper, Robert 138, 167 Great Expectations 119
cooter 69 Nicholas Nickleby 119–20, 123
Cornish 63, 106, 127, 157, 210; see also Dicker, Susan 229–30
Celtic languages dictionaries 125, 180–1, 186–91, 233,
co-varying shifts 11, 32, 34, 40 235–6
Courtenay, Archbishop William 144 English 146–7, 196–7, 216–19
creoles 36, 52, 76, 79, 206, 225–7, in language planning 3, 25, 35, 113,
242, 243; also see interlanguages 116–17, 192, 198
and pidgins Dictionary of American Regional
Croatian 127, 140 English 54, 69
Crowley, Robert 112 Dictionary of National Biography 219
Crowley, Tony 19 n., 235, 237, 250 Dictionary of the English Language, see
Crystal, David 19–20 Johnson, Samuel
Cursor Mundi 56 Digges, Dudley 111
Curzan, Anne 44–5, 64 diglossia 106–8
Cyrillic alphabet 140, 173 discourse analysis 49
292 Index

Dissertations on the English Language, see Brooklynese 58

Webster, Noah Canadian English 54, 63, 162, 223, 228
Dixon, Robert 19 Chancery English 148, 198
double negatives 35, 48 Chicano English 72, 78, 228
Dresier, Theodore 121 Chinese English 71–2, 224
drink 31 Cockney 117, 159
‘dropped h’ 197, 234 Geordie 54, 192, 253
Dryden, John 31, 112 global English 19–21, 58–9, 64, 68,
dude 12 70–5, 128–9, 135–6, 166–7, 183, 205,
Dutch 70, 133, 137, 174 235, 243–4, 254–5
Indian English 51, 67, 224–5
eastern Europe 68 Jamaican English 71–2, 79, 223, 228
Eats, Shoots & Leaves, see Truss, Lynn London English 8, 51, 220, 221
Ebonics, see African American Maori English 159–60, 234
Vernacular English under English, Middle English 12, 22, 28, 34, 41, 42,
varieties of 48, 55–6, 58, 78, 112–13, 131–2, 134,
Echevarrı́a, Roberto González 72 199, 205, 209
Eco, Umberto 92 Midwestern English 25
Edinburgh Gaelic School Society 151 Modern English 22, 34, 41, 48, 59, 77,
education 23, 192, 204–5, 236, 249–50; 198, 199
see also bilingual education, English New Zealand English 12, 192, 197, 224,
as a second language, and language 228, 244
and education Nigerian English 51, 54
Education Act of 1870 151 non-standard English 17, 25, 118, 122,
Edward the Confessor 109 124, 209–10, 211–14, 220–1
eggs 56, 114 Old English 5, 12, 13, 22, 28, 39–40, 42,
Eisenhower, Dwight D 260 43–7, 52–4, 55, 64, 76, 78, 79, 117,
Eleanor of Aquitaine, Queen of 141–4, 198, 199, 205, 238, 239, 253
England 107 Philippine English 72, 228
Elbegdor, Tsakhia 128 postcolonial English 51, 67, 225, 245
Elements of Style 18 Scots 31–2, 139, 222
elevator 4 Singapore English 72, 253
Eliot, John 153 South African English 12, 59, 192, 223,
Elizabeth I, Queen of England 112 228, 233
Elyot, Sir Thomas 110 South Asian English 52, 72, 78, 227, 228
emergent linguistics 15, 33, 240 southern British English 4
English, varieties of southern United States English 4
African American Vernacular Standard British English 52, 60, 78, 79,
English 35, 52, 77, 192, 204, 206, 220, 207, 222, 233
221, 228, 230–1, 253 Standard English 17, 110, 113, 131, 159,
Australian English 51, 52, 59, 60, 71, 199–202, 205–7, 212, 220–1, 224,
223, 228, 233 229–30, 233, 234, 235, 243–4, 245,
Black South African English 77 247–8
 Index 293

Standard United States English 52, as prestigious language 56, 200–1

57–8, 60, 69–70, 78, 79, 177–8, 181, Canadian 63, 160
200, 222, 228, 230, 233, 244, 247–8 Fries, Charles 207
Taglish 173, 225 frisk 51
world Englishes 17, 228
English as a second language 139, 183, 248 Gaelic 63, 66, 106, 127, 133, 140, 151, 152,
English Civil War 112 165; see also Celtic languages
English Only movement 165, 171, 183, 185, Gagarin, Yuri 190
245 Gandhi, Mahatma 62
Enlightenment, The 87, 95 Garden of Eden 82, 86, 89–90, 92, 96–7,
Erasmus, Desiderius 193 143
Esperanto 24, 99–105, 125 Gaulish 64
Etruscan 64 gay 8, 134, 184, 263
etymological doublets 32 Gazophylacium Anglicanum 96
etymological fallacy 45, 258 Geikie, Revd A. Constable 223
European Charter for Regional or Geordie, see English, varieties of
Minority Languages 128, 168 Gerald of Wales 108
European Union 138 German 32, 38, 70, 133
evolution 7, 13, 15 as language of the United States 132,
137, 163–4
Faeroese 12, 13 as original language 96
father 45, 80 Germanic languages 43, 63–4, 210
Faulkner, William 118 historical 31, 32–3, 38, 39, 40–1, 52, 54,
Federalism 178–9, 180 60, 170, 205, 238–9, 253
Fichte, Johann Gottlieb 169 inXuence on Old English 63–4
Finland 140 Germanic philology 205
Finnish 140 Germany 129
First Act of Uniformity 150 Gı̄kūyū 68
First Consonant Shift 32–3, 38, 39, 41, 47, Gildas 23, 141
73, 98, 130, 170 Gill, Alexander 200, 209–10, 212
First Grammatical Treatise 94 Gingrich, Newt 187, 245
First World War 132, 163, 183, 260–1 girls 264–6
Wx 255 Glixon, David 187, 189
Flemish 63 global English, see English, varieties of
Follett, Wilson 186, 187, 188, 189, Glorious Revolution 112
217, 236 Goddard, Henry 160
France 129, 172 Golding, William 16
Franglais 72–4 Gothic 97
Franklin, Benjamin 132 Gove, Philip 187, 189, 217, 218, 235
French 15, 70, 74, 116, 137, 139, 140, 172, Gower, John 88–9, 107, 112
210, 258 Grace, Patricia 123, 124
as language of England 63, 106–9, 131, Graddol, David 21
144, 160, 169 grammar 191–2, 231–2, 246
294 Index

grammar books historiography 81–2, 93, 105, 130–3, 160,

as instruments of codiWcation 35, 113, 209, 233, 239–46
116, 157, 181, 192, 199 Augustinian 141–2
English 45, 56–7, 104, 181, 182, 194–7, Norman 106–9, 125, 160
200–2, 209–11, 232 history of English 60–1, 83, 88, 122, 131–2,
grammatical gender 43–6, 64, 73 192, 199–200, 205–7, 209–10, 222–7,
Great Vowel Shift 11, 13, 14, 31–5, 38–9, 231, 238–46
40, 46–7, 98, 243, 251 Hitler, Adolf 102–3
Greek 92, 93, 95, 97, 110, 150 Hoban, Russell 125, 126
as prestigious language 64, 157, 194, Hobbes, Thomas 202
196, 200–1, 216–17, 232, 252 Hoccleve, Thomas 110
Greenberg, Joseph 99 Hogg, William 111
Gregory IX, Pope 144 Homer 5, 6, 111, 239–40
Grimm, Jacob 32, 38, 170 Honey, John 17–18, 38, 122, 204, 235,
Grimshaw, Beatrice 226 237, 250
Gullah 206 hopefully 23
guys 12 horse 39
How Late It Was, How Late, see Kelman,
Halliday, M. A. K. 9, 252, 253 James
Hamlet, see Shakespeare, William Howe, John 175–6
Hardy, Thomas 118–19, 120, 123 hubbub 94–5
Jude the Obscure 119 Humboldt, Wilhelm von 167–70
Harold Godwinsson 109 Hume, David 70, 203, 255
Harrison, David 20 n. Hungarian 96
Hart, John 38–9, 193, 194, 195, 198, 210,
245 Icarus 105
have 197–8 ideal speaker-listener, see Chomsky,
Hawaii 153–4, 156, 224 Noam
Hawaiian 65, 153–4, 226, 245 ideology of a standard, see
Hawaiian Creole English 226 standardization
Hawaiian Pidgin English 224, 226 Ido 101
Hayakawa, S. I. 164 Igbo 67
Hebrew 63, 90–2, 95–6, 150 immigration 3, 8, 40–1, 135, 162–4, 168,
Hemingway, Ernest 42 249–50
Henry I, King of England 214 into the United States 69, 121, 132
Henry II, King of England 107 India 62, 64, 68, 80, 98, 136–7, 224, 247
Henry V, King of England 148, 168 Indian English, see English, varieties of
Henry V, see Shakespeare, William Indo-European 16, 31, 33, 43, 60, 98, 205,
Herder, Johann 169, 170, 175, 179 238–9, 254
Higgins, Henry 41, 214 Indonesia 62
Hindi 92, 185, 253 Industrial Revolution 20, 133, 169
HirschWeld, Bob 18 -ing 38, 41, 117
 Index 295

interlanguages 72–5, 79, 225–7; see also Labov, William 7, 14, 16, 40, 49, 115–16,
creoles and pidgins 117, 119, 160
Iran 140 landsmål 172
Ireland 28, 151 Langland, William 74–5, 112
Irish 28, 63, 65–6, 79, 127, 151, 152, 165; language academy 132, 177, 178, 189, 196
see also Celtic languages language and education 3, 20, 67, 80,
shift from 66, 151, 167 109, 134, 135, 136–7, 138–9, 151–2, 154,
Isidore of Seville 90 156–7, 160–1, 165, 181, 199, 204–5,
Israel 62 236, 249–50; see also bilingual
Italian 70, 201 education
language and ethnic identity 38, 80, 102,
Jamaica 71, 223 121–4, 129, 170–3, 182, 232, 245
Jamaican English, see English, language and gender 43–6, 80, 214, 215,
varieties of 264
James I, King of England 165 language and government 16, 25, 101–3,
Jamestown 61, 133 135, 138, 150–85, 248–50
Japlish 72–4 language and legislation 23, 25, 134, 135,
Jay, John 177 168, 263–4
JeVerson, Thomas 179–80, 181 language and morality 3, 37, 86, 88–93,
Jespersen, Otto 16 107, 118, 120–2, 143, 181
John of Salisbury 90 language and national identity 67,
Johnson, Dr Samuel 6, 16–17, 18, 202, 78–80, 102, 122, 132, 133, 140, 165–6,
217–18, 235 167, 168–85, 200–1, 222, 245, 248, 253
Dictionary of the English language and race 70–1, 121–2, 124, 245,
Language 180, 196, 199, 218, 255 263–4
Plan of an English Dictionary 217 language and religion 66, 137–49, 153,
Jones, Daniel 154, 163, 173, 177, 182, 186, 187, 188,
An English Pronouncing 193–4, 214–5, 219; see also Bible, The
Dictionary 197, 234 and Lollardy
Jones, Sir William 97–8, 180, 208 language and social display 50, 75–6
Judd, Laura Fish 153–4 language and social stability 121, 122, 178,
Junius, Francis 206 181, 186–90, 204–5, 234
language change and economics 21, 139,
Kachru, Braj 68, 79, 136, 227, 244 166
Kalevala 140 language change and social meaning 30,
Kelman, James 122–3, 124, 128, 178 40, 47, 50, 57–9, 116–17, 121–2, 212–13
Kendall, Thomas and Samuel Lee 157 language change, classiWcation of 47–50,
Kenya 67 76–80
King James Bible 60 language change, controlling 100–2
Kiswahili 68, 167, 253 language change, explanations of 6–16,
Knighton, Henry 145, 147, 149 77–8, 82, 254–5
Kriol 227 language change, inevitability of 50–2,
Kynaston, Sir Francis 111–13, 121, 122, 196 181, 251–5
296 Index

language change, literary representations lift 4

of 24, 37, 105–26, 159, 177, 178, 233 Lightfood, David 239–40
language change, neutrality of 17, 29 linguistics 133–4, 231–7; see also
language change, predictions dialectology, Germanic philology,
about 133–4, 254–5 grammar, and history of English
language change, regularity of 30–75, and language varieties 208–31
254–5 and standard language 191–208
language contact 9, 23, 44, 61–75, 108, historical 97–9, 104, 134–6
109, 137, 158, 160, 212, 222–3, 253 List of Banished Words 18
language death 19–20, 137, 254–5 Locke, John 87, 202–3, 208, 252
language, historical stages of 16–17, Lodge, David 255
28–9, 31, 76–9, 95–9 Logonomia Anglica, see Gill, Alexander
of English 12, 22, 24, 31–3, 34, 41, 48, Lollardy 144–9, 150, 181, 257
58–61, 112, 192, 205–7; see also London English, see English, varieties of
history of English Lönnrot, Elias 140
language, individual and collective Lord of the Flies 16
use 10–11 Lowth, Bishop Robert 198
language, origins of 14–15; see also Tower Lycidas, see Milton, John
of Babel Lydgate, John 110, 168
language, philosophy of 10–16, 48, 202–4
language planning 3, 25, 138–40 Malinké 69
language revivals 127–8 man 45
language shift 19–20, 22, 28–9, 140–1, Manchin, Joe 164
157–8, 159, 173, 184–5, 222–3, 254–5 Manning, H. Paul 173
langue, see Saussure, Ferdinand de Manx 127, 157
Lass, Roger 6, 11, 15, 17, 29, 38, 40, 41, 47, Maori 69, 78–9, 123, 127, 151, 155–60, 245
130, 240 contact with English 24, 63, 65, 185
Last of the Mohicans, see Cooper, James Maori English, see English, varieties of
Fenimore Maori Language Act 158
Latin 15, 64, 70, 74–5, 93–4, 97, 106–9, Map, Walter 108
135, 181, 199, 210 Marshall, Jonathon 54
as prestigious language 56, 110–12, 157, Massachusetts Bay Colony 153
169, 193–4, 196, 200–1, 216–17, 232 Master 213
as religious language 66, 93–4, 131, May, Stephen 252
140, 142–3, 144–9, 150, 193 Mazrui, Alamin 166
Latino Sine Flexione 99 Macaulay, Lord Thomas B. 62, 66–7, 136,
Lau vs. Nichols 161 224, 247
League of Nations 102, 125 Macdonald, Dwight 188, 217, 236, 250
Leland, Charles 71 McWhorter, John 18–19, 38, 122
Letter H Past, Present, and Future, Medulla Grammatice 216
The 234 Megapolensis, Revd John 225
liberation linguistics, see Quirk, metathesis 39–40
Randolph Methodism 140
 Index 297

Mexico 249 contact with English 19, 24, 64, 65–6,

Meyer vs. Nebraska 163–4 72, 74, 87, 90, 137, 157, 185
Meyer-Benfey, Heinrich 38 education and 152–3, 155
Michel, Dan 56 Native American Languages Act 166
Middle English, see English, Native School Code of New Zealand 157
varieties of Native Schools Act of New Zealand 157
Midwestern English, see English, natural gender 43–6
varieties of Nauerby, Tom 172
Milroy, James 262 Navajo 63, 66, 152
Milton, John 41, 202, 258 nazi 263, 266
Lycidas 11 Nazism 102–3
Paradise Lost 45, 89, 94–5, 111 Nettle, Daniel 20
Paradise Regained 111 Newspeak 208
Samson Agonistes 111 Newton, Sir Isaac 211
Miss 8, 134, 184 New York Times 187–9, 190, 259–60
Mobilian Jargon 227 New Zealand 61, 68, 127, 153, 160, 223,
Modern English, see English, varieties of 241, 245, 246, 247, 261, 264; see also
Mohawk 225 Maori
Mona Lisa 11 linguistic repertoire of 64, 69, 151,
Mongolia 128 155–60, 197
Morrison, Toni 118 New Zealand English, see English,
Morton, Herbert 189 varieties of
mother 45, 80 Nicholas Nickleby, see Dickens, Charles
Mufwene, Salikoko S. 206, 254 Nigeria 67, 68
Mugglestone, Lynda 117–18 Nigerian English, see English, varieties of
Mulcaster, Richard 200–1 nigger 184, 263–4, 266
The Elementarie 194–5 non-standard English, see English,
Müller, Max 16, 18, 60, 98 varieties of
multilingualism 6, 62–4, 106–9, non-standard language 116, 117, 208–9,
167, 170 212,
Murray, Sir James 118, 217, 219, 241 Norman Conquest 106, 109, 144
myth 82–3, 88 Northern Cities Vowel Shift 131
Norwegian 140, 172
Nadsat 124 Nynorsk 140, 172
Natick 153
National Policy on Languages of Oakland, California School Board 230–1
Australia 167 Oberly, John H. 155
National Portrait Gallery 219 Odyssey, The, see Homer
National Viewers’ and Listeners’ oYcial language 135, 158, 164–6, 173–4,
Association 215–16 183
Native American languages 3, 67, 78, 95, OYcial Languages Act of Canada 165
99, 127 Ogden, C. K. 124, 207
borrowing from 9, 50 Old Danish 52
298 Index

Old English, see English, varieties of phonological balance 14, 33 n.

Old Frisian 238, 253 pidgins 36, 54, 71, 72, 74, 76, 224, 225–7,
Old High German 39 242; see also interlanguages and
Old Norse 34 n., 39, 44, 63, 64, 73 creoles
Old Saxon 39, 238, 253 Piers Plowman, see Langland, William
Old Swedish 52 Pinker, Stephen 265–6
Once Were Warriors 261 Plantation Creole 206
optimal communication 13 postcolonial English, see English,
Oresme, Nicholas 91 varieties of
Organic Act of Hawaii 154 Postel, Guillaume 92, 102
Orosius 143 post-vocalic [r] 2, 118
Orthographie, see Hart, John Pratt, Captain Henry 153, 155
Orwell, George Presbyterianism 140
Animal Farm 243 prescriptive grammar 18, 35–7, 42–3, 189,
1984 208 209; see also bad language and
‘Politics and the English descriptive grammar
Language’ 265 Preston, Denis 57–8
Oscan 64 preterites, formation of 36–7, 52
Oswald, King of Northumbria 65–6 Primitive Germanic, see Germanic
Oxford English Dictionary 15, 118, 219, language family
236, 241–2, 258, 259 print 56, 193, 198, 234, 246
professor 39
PaciWc Rim languages 90 Prometheus 105
Pamphlet for Grammar, see Bullokar, Promptorium Parvulorum 216
William Prose Edda 94
Papua New Guinea 225, 226 Provençal 127
Paradise Lost, see Milton, John Pryce, Malcolm 123
Paradiso, see Dante Aligheri Puerto Rico 136, 151
Passamaquoddy 137–8, 140, 149 punctuation 17–18, 42, 45
Peasants’ Revolt 150 Punic 64
Pecock, Reginald 147 push-chain mechanism 8, 11, 14
Pennycook, Alastair 136 Puttenham, George 56–7, 212–13
Pentecost 92–3, 102, 106, 108 Pygmalion 41
perceptual dialectology 57–8
perfect languages 97, 99 queer 263
Perkins, Maxwell 42 Quirk, Randolph 228–30, 235,
Philippine English, see English, 245, 250
varieties of
Philippines, The 140, 173–4, 227 raccoon 50
Philips, Edward Raeti 38
The New World of English Words 217 Ragnarok 82–3
Phillipson, Robert 135–6, 235, 250 Reagan, Ronald 161
phoneme 11 Received Pronunciation 48
 Index 299

‘Reeve’s Tale’, see Chaucer, GeoVrey schmuck 69

Reformation 66, 150 science Wction 124–6
regional variation 50–60, 68–70, 77, Scot, Sir Walter 118, 173
113–17, 208–31, 242, 244, 247–8; Scots, see English, varieties of
see also dialects Second Consonant Shift 32
Restoration 112 Second Shepherds’ Play 113, 115
Richard II, King of England 107 Second World War 132, 167, 173, 189
Richards, I. A. 208 semantic amelioration 8
Rid, Samuel 216 Serbian 127, 140
Ritt, Nikolaus 7 Shakespeare, William 35, 41, 188
Robert of Gloucester 258, 264 Hamlet 11, 13
Rodriguez, Richard 152 Henry V 116, 118
Rolle, Richard 55 Twelfth Night 213
Romaine, Suzanne 20 Shaw, George Bernard 41
Romanticism 79, 167–70, 171, 175, 179, Sheridan, Richard 118
252, 254 Sheridan, Thomas
Roosevelt, Theodore 183 Course of Lectures on Elocution 213–14,
Royal American Magazine 222 234
run 39 n. Shibboleth 5
Russia 102, 129 sign language 44 n.
Russian 128, 172, 175 Silliman, Benjamin 69–70
Siman Act 163
St Augustine 99, 124, 143, 146 Simon, John 60
on Babel 84–6 Sinclair, Upton 121
on historiography 93, 104 sing 30
on language 89–91, 92, 193, 231, 233, Singapore English, see English,
245 varieties of
St Bernard 108–9 Sir Gawain and the Green Knight 132
St Gregory the Great, Pope 4, 142, slang 116–17, 188–9, 216; see also bad
143 language
SaWre, William 204 Smith, Jeremy 8, 32, 38, 48, 64
Sampson, George social diversity 3
English for the English 232 social networks 75 n.
Samuels, M. L. 77 Society for Pure English, see Bridges,
Sanskrit 97 Robert
Sapir, Edward 31 Society for the Promotion of Christian
Sapir-Whorf hypothesis 98, 261–3 Knowledge 215
Saussure, Ferdinand de 11, 98, 185, Society for the Reformation of
207–8, 241 Manners 215
Schlesinger, Jr., Arthur 129 Sockabasin, Allen 137–8, 140
schlock 69 South Africa 62, 68, 166,
Schmidt, Ronald 183 174, 223
300 Index

South African English, see English, Table of Nations 84, 88

varieties of Taglish, see English, varieties of
South America 68, 140, 222 Tagalog 140, 173–4
South Asian English, see English, tally sticks 152
varieties of Tanton, John 164
southern British English, see English, Tasman, Abel 155
varieties of teleological change 14
southern United States English, see Tennyson, Alfred Lord 86
English, varieties of Thatcher, Margaret 234
Soviet Union 102–3, 140, 170–1, 172, 175, Thiong’o, Ngũgı̃ wa 67–8
190 third person indicative inXection 4, 37,
Spanglish 72–4 73, 134
Spanish 64, 70, 72–3, 96, 136, 151, 173–4, third person personal pronouns
200–1 73, 76
as global language 21, 254 Thomas, Dylan 21
in the United States 25, 78, 133–4, 137, Thorpe, William 147–8
185, 243, 249 thou 213
Spanish-American War 136, 173 Thracian 64
spinster 46 toboggan 50
Stalin, Joseph 175 Tocqueville, Alexis de 179–80
Standard British English, see English, Todd, Loreto 227
varieties of Tok Pisin 227
Standard English, see English, varieties of Tooke, Horne 203, 208
standard language 9, 58, 116, 117, 120, Tower of Babel 129, 143, 152, 160, 162,
192, 197–200, 208–9, 212, 220, 236–7, 176, 185, 202, 204, 245, 247, 250,
246 252, 265
Standard United States English, see and Esperanto 99, 104–5
English, varieties of as cause for punishment 4, 5
standardization 23, 25, 57, 67, 169, as foundational myth 23, 24, 83–95,
192–208, 220–1, 244 180, 208, 255
standardized language 56, 198–9, 233 inXuence of 95–9, 106, 126–7, 170, 218
Stephen, King of England 109, 144 Treaty of Waitangi 158–9, 247
strong verbs 30–1 Treaty of Wedmore 141
Strunkenwhite virus 18 Trench, Richard 219
stud 46 Trevisa, John 212
Swann Report 232 Troilus and Criseyde, see Chaucer,
Sweden 140 GeoVrey
Swedish 70, 96, 133 Trudgill, Peter 7
Sweet, Henry 133 Truss, Lynn 17–18, 42, 45, 99, 122
Swift, Jonathon 217, 255 Tutonish 99
Switzerland 62 Twain, Mark 52, 57, 118, 121
synchronic variation 4, 6, 241 Adventures of Huckleberry Finn 37
syntax, violations of 34, 36 Twelfth Night, see Shakespeare, William
 Index 301

UlWlas 179 warrior 256–66

uniformitarian principle 47, 51 ‘Warriors for Christ’ 261
United Kingdom 62, 65, 68, 160, 172, 241 Watereman, William 95
United States 68, 172, 241, 246 Waugh, Evelyn 124, 126
changes in linguistic repertoire 25, 62, weak verbs 36, 52
69, 133–4, 151, 163–4, 184 Webster, Noah 177, 180–3, 184, 186, 217,
language in 132, 133, 175–84, 245 223, 247–8, 253, 257
United States Bureau of Indian Dissertations on the English
AVairs 152, 155 Language 40, 132, 133
United States Civil War 176, 178, 181 Webster’s Third International
United States Constitution 165, 168 Dictionary 186–91, 195, 204, 216, 218,
United States Declaration of 233, 235, 236
Independence 176 Wells, H. G. 124–5, 130
United States Military Academy 260 Welsh 51 n., 77, 78–9, 106, 123, 127, 140,
United States War of Independence 1, 171, 173, 210, 253; see also Celtic
176, 178–9 languages
Universal Declaration of Linguistic shift to English 24, 65, 66, 133, 151, 152,
Rights 128, 168 163–4, 185
US English 164 Welsh, Irvine 118, 122, 128, 178
Usk, Thomas 56 Welsh Language Act 165
West African languages 52
variety 4 n. which 198
Vatican II, Council of 150 White, Hayden 81–2, 84, 105, 126,
Venetic 64 233, 239
verbal hygiene, see Cameron, Deborah White, John 258, 259
Verner’s Law 33, 130 Whitehouse, Mary 215–16
Verstegan, Richard 96–7, 99, 196, 201, Whitman, Walt 176
206, 210 Whorf, Benjamin Lee 98, 204, 208;
Vidal, Sir Raimon 94 see also Sapir-Whorf hypothesis
Vikings 82–3, 141–4, 245 Wilberforce, William 137
Virgil 111, 193 Wilde, Oscar 133
Vitalis, Orderic 107–8 Wilkins, John 97, 100
Volapük 24, 99 William I, King of England 107–9
Voltaire 95 William of Malmesbury 211
vulgaria 181, 193 William of Newburgh 108
Willinsky, John 189, 219
Waitangi Tribunal 157 Wilson, Thomas 45, 195
Wales 65, 151 Withers, Philip 214
Walker, John Wittgenstein, Ludwig 10
A Critical Pronouncing Dictionary 197 Wolfe, Thomas 42
Waller, Edmund 110, 121 Wordsworth, William 259
wanna 23, 35 world Englishes, see English,
Warner, Glenn ‘Pop’ 260, 264 varieties of
302 Index

wot 117 Xhosa 35

Wright, Arnold 227
Baboo English as ‘tis Writ 224 Yiddish 69, 127
Wright, Joseph 221 Yugoslavia 140
Wulfstan, Bishop of Worcester 141
Wyclif, John 144, 145, 147, 148 Zamenhof, Lazar 99–105, 124, 126
Wyld, H. C. 206 Zulu 63, 174