Advances in Stylistics Jonathan Culpeper Mireille Ravassat Stylistics and Shakespeare S Language Transdisciplinary Approaches Continuum 2011

Stylistics and Shakespeares Language
Advances in Stylistics
Series Editor: Dan McIntyre, University of Huddersfeld, UK
Editorial Board:
Beatrix Busse, University of Berne, Switzerland
Szilvia Csbi, Etvs Lornd University, Hungary
Monika Fludernik, University of Freiburg, Germany
Lesley Jeffries, University of Huddersfeld, UK
Jean Boase-Beier, University of East Anglia, UK
Geoffrey Leech, Lancaster University, UK
Larry Stewart, College of Wooster, USA
Manuel Jobert, Jean Moulin University, Lyon 3, France.
Titles in the series:
Corpus Stylistics in Principles and Practice
Yufang Ho
D. H. Lawrence and Narrative Viewpoint
Violeta Sotirova
The Discourse of Italian Cinema and Beyond
Roberta Piazza
I. A. Richards and the Rise of Cognitive Stylistics
David West
Opposition in Discourse
Lesley Jeffries
Style in the Renaissance
Patricia Canning
The Stylistics of Chick Lit
Roco Montoro
Stylistics and Shakespeares
Language
Transdisciplinary Approaches
Editors
Mireille Ravassat
and
Jonathan Culpeper
Continuum International Publishing Group
The Tower Building 80 Maiden Lane
11 York Road Suite 704
London New York
SE1 7NX NY 10038
www.continuumbooks.com
Mireille Ravassat and Jonathan Culpeper and contributors 2011
All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical,
including photocopying, recording, or any information storage or
retrieval system, without prior permission in writing from the publishers.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
EISBN: 978-1-4411-6425-4

Library of Congress Cataloguing-in-Publication Data
A catalog record for this book is available from the Library of Congress.
Typeset by Newgen Imaging Systems Pvt Ltd, Chennai, India
Printed and bound in India
Contents
List of Illustrations vii
Notes on Contributors ix
Introduction 1
Mireille Ravassat and Jonathan Culpeper
Chapter 1: Strange deliveries: Contextualizing Shakespeares
First Citations in the OED 8
Giles Goodland
Chapter 2: Shakespeares Vocabulary: Did it Dwarf All Others? 34
Ward E. Y. Elliott and Robert J. Valenza
Chapter 3: A New Kind of Dictionary for Shakespeares Plays:
an Immodest Proposal 58
Jonathan Culpeper
Chapter 4: If I break time: Shakespearean Line Endings
on the Page and the Stage 84
Peter Kanelos
Chapter 5: Subject-Verb Inversion and Iambic Rhythm in
Shakespeares Dramatic Verse 98
Richard Ingham and Michael Ingham
Chapter 6: Shakespeares Short Pentameters and
the Rhythms of Dramatic Verse 119
Peter Groves
Chapter 7: Wholes and Holes in the Study of Shakespeares
Wordplay 139
Dirk Delabastita
vi Contents
Chapter 8: a thing inseparate/Divides more wider
than the sky and earth of Oxymoron in
Shakespeares Sonnets 165
Mireille Ravassat
Chapter 9: Rue with a difference: a Computational Stylistic
Analysis of the Rhetoric of Suicide in Hamlet 192
Thomas Anderson and Scott Crossley
Chapter 10: Shakespeares Sexual Language and Metaphor:
a Cognitive-Stylistic Approach 215
Jos L. Oncins-Martnez
Chapter 11: Cognitive Interplay: How Blending Theory and
Cognitive Science Reread Shakespeare 246
Amy Cook
Index 269
List of Illustrations
Figures
1.1 Proportion of EEBO and other provenances in
sample batch. 16
2.1 A comparison of normalized word type frequency
profles for Shakespeare and the King James Bible. 44
5.1 Syntax of subject pronoun after an initial non-subject
constituent: percentage of VSpro order overall, and
percentage of VSpro order when the pronoun was in
an even (ictic) syllable. 108
7.1 Defning wordplay at a crossroads of four continuums. 150
Tables
2.1 Three vocabulary tests: Shakespeare and eight others. 42
2.2 Types-tokens, new-to-the-group words, total inferred
vocabulary, Shakespeare and other writers (appendix). 557
3.1 Word-types and tokens in Shakespeares plays. 65
3.2 The top ten ranked-ordered collocates of good
within a fve-word span. 67
3.3 The top ranked-ordered three-word lexical bundles in
Shakespeare and other texts. 73
3.4 Rank-ordered keywords for Romeo and Juliet. 75
3.5 The semantic categories used (derived from
McArthur 1981). 77
3.6 Love comedies and tragedies: characteristic semantic
categories (rank-ordered). 77
5.1 Syllable position of subject pronouns accompanying
monosyllabic verbs. 106
viii List of Illustrations
5.2 Frequencies of VS and SV syntax with main clause
subject pronouns in inversion contexts featuring
a monosyllabic fnite verb. 106
5.3 Frequencies of inverted subject pronouns in even and
odd syllables with monosyllabic verbs in Shakespeares
dramatic verse. 107
5.4 Frequencies of main clause VSpro and SproV syntax in
inversion contexts with monosyllabic verbs, with subject
pronouns in even syllables. 107
9.1 Corpora descriptive statistics. 200
9.2 Mean, standard deviation, and F scores for LIWC
values (Hamlets versus Ophelias suicidal discourse). 201
values (Ophelias suicidal discourse versus
non-suicidal discourse). 202
values (Hamlets suicidal discourse versus
non-suicidal discourse). 204
Notes on Contributors
Thomas Anderson is Assistant Professor of English at Mississippi State
University, USA. His book, Performing Early Modern Trauma from Shakespeare
to Milton (Ashgate 2006), looks at the way historical crises such as the
Reformation, regicide and royal death are registered in early modern litera-
ture. He has published articles on Shakespeare, Marvell and Milton and has
recently co-edited with Ryan Netzley a collection of essays on John Foxes
Actes and Monuments entitled Acts of Reading: Interpretation, Reading Practices
and the Idea of the Book in John Foxes Actes and Monuments (University of
Delaware Press 2010).
Amy Cook is Assistant Professor at Indiana University, Bloomington, USA.
Her book, Shakespearean Neuroplay: Reinvigorating the Study of Dramatic Texts
and Performance through Cognitive Science (Palgrave Macmillan 2010) uses
Hamlets mirror held up to nature as a test case and provides a method-
ology for applying conceptual blending theory to Shakespeares textual
theatrics. She has published essays in Harold Blooms Modern Critical
Interpretations series, Theatre Journal, TDR, and SubStance, among others.
Scott Crossley is Assistant Professor of Applied Linguistics and ESL at
Georgia State University, USA, and he maintains a close affliation with the
Institute for Intelligent Systems at the University of Memphis, USA. His
interests include computational linguistics, corpus linguistics, discourse
processing and discourse analysis. He has published articles in genre analy-
sis, multi-dimensional analysis, discourse processing, speech act classifca-
tion, psycholinguistics, second language acquisition and text linguistics.
Jonathan Culpeper is Professor of English Language and Linguistics at
Lancaster University, UK. He is particularly interested in written (re)pre-
sentations of spoken interaction. He has published works in the area of
stylistics, especially the stylistics of drama. His most recent major publica-
tion, co-authored with Merja Kyt, is Early Modern English Dialogues: Spoken
Interaction as Writing (Cambridge University Press 2010).
x Notes on Contributors
Dirk Delabastita is Professor of English literature and literary theory at the
University of Namur, Belgium. He wrote his PhD on Shakespeares wordplay
in Hamlet and the problems of translating it (Theres a Double Tongue, 1993).
He edited Wordplay and Translation (1996) and Traductio. Essays on Punning
and Translation (1997) and co-authored a Dutch-language dictionary of
literary terms (7th edition, 2007), translated into French. He also co-edited
European Shakespeares (1993), Fictionalizing Translation and Multilingualism
(2005) and Shakespeare and European Politics (2008).
Ward E. Y. Elliott is Professor of American Political Institutions at Claremont
McKenna College, California, USA. He and his colleague, Robert J. Valenza
were co-advisors of the Claremont Shakespeare Clinic, 19871994. They
have published articles on Shakespeare authorship, notably in Shakespeare
Quarterly, Computers and the Humanities, Literary and Linguistic Computing,
among others. Ward Elliott is co-dedicatee of Brian Vickerss Shakespeare
Co-Author (OUP 2002).
Giles Goodland is Senior Editorial Researcher at the Oxford English Dictionary,
UK. He specializes in the use of databases of early modern English (such as
EEBO and ECCO) to provide information in the revision of OED entries,
helping to supervise a small team of Research Assistants in this task. He has
a D. Phil. in English Literature from Oxford University. He has had several
books of poetry published, most recently What the Things Sang (Shearsman
2009).
Peter Groves did his PhD on Shakespeares prosody at Cambridge University,
UK. He is a Lecturer in Shakespeare, Renaissance literature and poetry at
Monash University, Melbourne, where he co-edits the e-journal Versifcation:
an Electronic Journal of Literary Prosody. He has published a book and a num-
ber of articles on metre and versifcation in English, the most recent being
Shakespeares Pentameter and the End of Editing, in Shakespeare, the
Journal of the British Shakespeare Association.
Richard Ingham is a Senior Lecturer in English at Birmingham City University,
UK, where he teaches language acquisition, English grammar and the lin-
guistic history of English. His research interests are in language acquisition
and change, with special reference to English. He has published in a large
number of international refereed journals. His most frequently cited pub-
lications are on Middle English negation and pronoun use in English child
language. His current research focus is on language in medieval England.
Notes on Contributors xi
Michael Ingham is Associate Professor in the Department of English, Lingnan
University, Hong Kong, where he has been teaching English Studies since
1999. His areas of professional expertise are speech and drama, literature
and literary linguistics and performance studies, particularly Shakespeare
in performance. He is a founder member of Theatre Action, a Hong Kong
based drama group that specializes in action research on literary drama
texts.
Peter Kanelos is Assistant Professor of dramatic literature, theatre history
and dramaturgy at Loyola Universitys Department of Fine and Performing
Arts, Chicago, USA. He is the editor of Much Ado About Nothing (The
New Kittredge Shakespeare), associate editor of Twelfth Night (The New
Variorum Shakespeare) and co-editor of the forthcoming, Thunder at a
Playhouse: Essaying Shakespeare and the Early Modern Stage. He has lectured
internationally on Shakespearean drama, and is currently working on two
book projects on Shakespeare.
Jos L. Oncins-Martnez is Lecturer in English language and literature at
the University of Extremadura, Spain. He is the author of a volume on the
translations of Shakespeares Timon of Athens into Spanish and co-editor
of three collections of essays published in Spain. His academic interests
include topics such as translation, cognitive linguistics, pragmatics and
stylistics, on which he has published several articles. At the moment he is
engaged in a research project on Shakespeares phraseological language.
Mireille Ravassat is a Senior Lecturer in the Department of English Studies
at Valenciennes University, France, where she teaches English literature
mainly Shakespeare stylistics and translation from English into French.
Her main felds of interest and investigation are Shakespeares language
and stylistics. She has contributed many articles and book chapters on
this topic and regularly organizes international seminars on Shakespeare,
notably at ESSE conferences. She is currently working on a monograph on
Shakespeares style in the Sonnets.
Robert J. Valenza is the Dengler-Dykema Professor of Mathematics and
the Humanities at Claremont McKenna College, California, USA, and
author of books on linear algebra, abstract algebra and number theory.
With Ward E. Y. Elliott, he was faculty advisor to the Claremont Shakespeare
Clinic from 1987 through 1994. They have published widely on the issue of
Shakespeare authorship.
Introduction
Mireille Ravassat and Jonathan Culpeper
Even if recent times testify to a renewed surge of interest in the issue
of Shakespeares language and style, the following statement by
M. C. Bradbrook is still arresting and valid in the early twenty-frst cen-
tury: There is no question relating to Shakespeare as a writer which does
not involve his style . . . . Yet on this central problem comparatively little has
been written. It is too vast and intimidating. . . (Shakespeare Survey 1954: I)
1
.
Some four decades later, in an issue of the same journal also devoted to
Shakespeare and Language, Stephen Booth chimes in somewhat provok-
ingly: Shakespeare is our most underrated poet. He adds: The reason it is
necessary to point out Shakespeares poetic superiority to competing poets
is, I think, that we have so long, so industriously ignored the qualities in lit-
erature that drew us to it in the frst place (1997: I).
2
In other words, there
is defnitely still room for far-reaching explorations of Shakespeares multi-
farious artistry and mind-boggling inventiveness in matters of language.
Moreover we decided to edit the present volume because we felt such an
enterprise was a necessary, a timely and appropriate one, with respect to
bridging the gap between Shakespeare and language studies, while taking
into account new approaches like computational, corpus based and cogni-
tive studies. In addition, we intend this book to be a synthesis of linguistic
and literary criticism, too often alien approaches despite the in-between
development of the concept of linguistically-oriented new stylistics.
3
Here
the following statement by Geoffrey Leech and Michael Short in Style in
Fiction (2007 [1981]) is most relevant:
One major concern of stylistics is to check or validate intuitions by detailed
analysis, but stylistics is also a dialogue between literary reader and linguis-
tic observer, in which insight, not mere objectivity is the goal . . . . [l]iterary
expression is an enhancement, or a creative liberation of the resources
of language which we use from day to day. Correspondingly, stylistics
builds on linguistics, and in return, stylistics challenges our linguistic
2 Stylistics and Shakespeares Language
frameworks, reveals their defciencies, and urges us to refne them. In
this sense, stylistics is an adventure or discovery for both the critic and
the linguist.
4
Stylistics has actually been something like a revolution in the relation
between linguistic and literary studies to quote Randolph Quirk in his
foreword to Style in Fiction (x). We are happy to edit this volume at a time
when stylistic studies have shown signifcant development. The success of
the Poetics and Linguistics Association (PALA) conferences testifes to
this trend, for example, the 2006 one in Finland devoted to The State
of Stylistics
5
. In France the growth of the Socit de Stylistique Anglaise
is yet another indicator. It is also a time when going beyond boundaries
is no more a dream, as, for example, the 2008 Queens University Belfast
conference on Text and Context: Bringing Early Modern Literature and
Linguistics Together has shown. The increasing number of participants at
sessions on Shakespeare and style in which we are involved, notably at ESSE
conferences, is also very encouraging.
The idea of tackling Shakespeares language and style has really involved
a stimulating cooperation. Through it, as a literary scholar and a linguist
ourselves, we are glad to offer the following transdisciplinary studies which
demonstrate that there is new life in traditional approaches as well as
entirely new ones, charting the map of most innovative research in the joint
areas of language and literature. The chapters in the collection capture a
rich diversity of points of view and cover felds such as lexicography, versif-
cation, rhetorical analyses, cognitive and computational corpus-based stylis-
tic studies. The perspective is deliberately a broad one, the purpose being
to confront ideas and visions at the intersection of various techniques of
textual investigation.
We wish to warmly thank all the authors of this volume for their contribu-
tions and also all the team of specialists who kindly accepted to review parts
of this book and did it with close attention and expertise.
* * *
The opening chapter of this volume by Giles Goodland questions the origi-
nality of Shakespeares lexicon. Perhaps due to the prevalence of frst cita-
tions by Shakespeare in the Oxford English Dictionary, frst known as New
English Dictionary, it is a commonly held item of belief that Shakespeare
was a great coiner of words, but this has often been asserted in an impres-
sionistic and peremptory manner. Now, in the last few years it has become
Introduction 3
possible to address this issue with greater confdence since a large body
of Early Modern printed material has become available for searching on
Early English Books Online and similar databases. In this chapter, Giles
Goodland focuses on some sections of the dictionary and traces to what
extent Shakespeares antedatings in the OED are true frst uses, and to what
extent they are a product of the prominence and availability of Shakespeares
texts at the time at which the dictionary was originally written. His demon-
stration paves the way for a useful reassessment of Shakespeares lexical
creativity.
Chapter 2 by Ward Elliott and Robert Valenza also tackles this specifc
issue. Indeed bardolatry lives on, notably in the still-common notion that
Shakespeares vocabulary dwarfed all others. The authors debunk this
myth by analysing Renaissance texts and using three tests of vocabulary
size and richness: type/token ratios, well-known measures of variety, new-to-the-
group words, a test measuring distinctiveness and inferred vocabulary, a new
test based on the top half of inferred frequency curves. All three tests say
that Shakespeares vocabulary was larger and richer than John Fletchers,
smaller and poorer than Miltons, and about equal to those of fve other
leading Elizabethan playwrights, and to those of most college-educated
people today. The authors demonstrate the problem has not been overesti-
mation of Shakespeare, but underestimation of everyone else. In that, their
research most usefully complements Chapter 1.
Jonathan Culpepers study concludes the lexicography section of this vol-
ume. The language of Shakespeares plays has received substantial treatment
in various dictionaries, glossaries, lexica and concordances. However,
the classic works are written in the philological tradition that characterized
the Oxford English Dictionary. This chapter explores how modern principles
and techniques developed in Corpus Linguistics can be deployed in the
creation of a radically new kind of dictionary. In particular, this involves a
focus on usage and frequency. A further innovation is that the proposed
dictionary will be comparative, making both internal comparisons (e.g.
female characters compared with male) and external comparisons (e.g.
Shakespeares usage compared with that of contemporary plays and other
genres). The bulk of this chapter is made up of case studies, involving dis-
cussion of the words horrid, good, ah and and, multiword units, and
linguistic profles for characters and plays. Through these, the aim is to
demonstrate the characteristics of the dictionary and raise pertinent issues,
including, for example, how many and what kind of words to include in
the dictionary, whether the dictionary should include only words (and how
they should be defned), how word-senses should be distinguished, how
stylistic and social meanings should be captured, and what approach to
grammar should be taken. This study is not narrowly focused on dictionary
design, but taps into key stylistic issues in Shakespeares plays, including
characterization, genre and theme.
The three following chapters deal with another important topic, namely
Shakespeares verse. Peter Kanelos, a professor and editor of Shakespeare,
with long-standing experience of working with actors and stage directors,
probes into one of the hallmarks of Shakespeares mature verse, namely the
issue of his increasingly enjambed style. Indeed, the more Shakespeares
career progressed, the more his line patterning became irregular, thereby
refecting his more complex vision of the universe. Relying on his dramatur-
gical background, Peter Kanelos focuses on the issue whether Shakespeare
advocated a fowing diction unimpeded by any pause in the phrasing,
or conversely meant to mimic the metrical arrangement by uttering the
enjambed lines in succession. By getting to the heart of this question, he
sheds light on the development of Shakespeares poetics.
In Chapter 5, Richard and Michael Ingham point out the fact that subject-
verb inversion was still common as a stylistic device in late sixteenth-century
English and they investigate here what interaction there may have been
between ictic stress placement (i.e. the stress is placed on an even-numbered
syllable) and the use of syntactic inversion in Shakespeares verse drama.
The frequencies of subject-verb inversion with respect to the placement
of ictic stress in blank verse pentameters are compared in 15 non-comedy
plays from Titus Andronicus to Cymbeline. Shakespeares practice changed
very markedly with respect to ictic stress and subject pronoun inversion
between the earlier and later plays. In the later plays, in cases when iambic
metre places ictic stress on a pronoun, the syntax strongly tends to avoid
inversion. Their fndings notably highlight that the decline in ictic stress
placement refects an independent stylistic choice on the authors part and
is not simply a by-product of a decline in inversion. Their study of syntax
and metre is relevant to debates about the value of iambic metre to mod-
ern-day actors speaking Shakespearean verse. Awareness of stylistic change
in versifcation over the course of Shakespeares career should enhance
understanding and informed choice among the wider discourse commu-
nity of those who perform Shakespeares language.
In Chapter 6, Peter Groves scrutinizes Shakespeares innovative use of
lacunae (empty slots in the metrical codifed pattern of iambic pentam-
eter). Lacunae represent either silent beats or silent off-beats, produced
in our negotiation as performers between written line and known pho-
nological and metrical constraints. The silent off-beat comes itself in two
Introduction 5
distinct favours: the jolt, which occurs between intonational phrases, and
which has the effect of emphasizing discontinuity, and in mimetic terms of
suggesting surprise, alarm, anger, urgency and so on; and the much rarer
drag, which occurs within a phrase and has the effect of locally slowing
down the tempo and forcing pitch-accent emphasis on the syllable that pre-
cedes it. The silent beat or rest, on the other hand, functions as a way of
cueing gesture and action in the theatre, and in addition has a number
of interesting deictic uses, marking and drawing attention to features of
the interaction between characters. The chapter explores the ontology and
epistemology of lacunae how as readers we recognize them, and as per-
formers produce them and investigates in detail the sorts of aural and
experiential effects they produce in performance.
In the following chapter, Dirk Delabastita presents a comprehensive, fex-
ible and radically historical model of Shakespeares wordplay. A systematic
set of critical distinctions is proposed that needs to be considered in the dis-
cussion of puns. In this way, various types of wordplay can be distinguished
from various types of non-wordplay. Quite crucially, it is argued that these
are gradual and very much context-sensitive distinctions rather than radical
ones. Moreover, by permitting the description of diachronic shifts along the
defnitions various continuums, such distinctions can be viewed from a his-
torical angle, which takes into account factors such as phonetic and seman-
tic evolution but also the historicity of interpretation (after all, it is our
interpretative and discursive strategies as spectators, readers and critics that
make or mar the puns that could potentially be attributed to Shakespeares
texts). The model is original inasmuch as it tries to transcend the limitations
inherent in the narrowly linguistic or literary approaches that have dom-
inated the feld until now: it reconciles the need for maximum descriptive
rigour with an acknowledgement of the hermeneutic indeterminacies and
the role of interpretative strategies and interpretative communities inevita-
bly involved in the identifcation and understanding of wordplay. Examples
are taken from a range of Shakespeare plays and the chapter enters into
a dialogue with wordplay critics as wide apart as the Victorian F. A. Bather
and our contemporary Patricia Parker.
In Chapter 8, Mireille Ravassat focuses on the use of oxymoron in
Shakespeares Sonnets. In a contextualized perspective, she offers a compre-
hensive and systematic study of this present absent fgure of speech and
thought ruling the various intertwined plots of this poetic collection. She
argues that a unifed perception of difference within similarity and similar-
ity within difference is the doubly chiastic and oxymoric pattern generally
defning Shakespeares bifocal vision in his dramatic and poetic production.
She further demonstrates that in the Sonnets, which are at the crossroads of
the baroque vein of ontological vacillation and of a modernized vision of
the concept of psychomachia, oxymoron violates its ethos of reconciliation
in favour of unresolved division. Far from an ideal of oxymoron as heav-
enly mingle (Antony and Cleopatra 1.5.62), the fgure of the Crosse-couple, in
the Sonnets, turns out to spawn compounds strange (76.4) and incongru-
ous lexical and semantic creations. Ultimately, systematically undoing what
it does, oxymoron maps out a discordant pattern of dissolution in fusion
that signals the birth of the modern self. This study reassesses oxymoron as
a very rich and complex stylistic device in Shakespeare which, by resorting
to clashes of opposites and the explosion of semantic categories, leads the
reader very deep into the human psyche.
In Chapter 9, Thomas Anderson and Scott Crossley examine the rhetoric
of suicide in Hamlet and demonstrate how computational stylistic methods
enrich literary approaches to the play. Using the computational tool LIWC
(Pennebaker et al. 2001), they conduct a lexico-semantic and corpus analy-
sis of the play to examine the semantic prosodies present in the suicidal
and non-suicidal discourses of Hamlet and Ophelia. A stylistic and literary
interpretation of the play demonstrates that the lexical variation between
the characters supports the interpretation that the two suicides are dis-
crete events with different cultural valences. Statistical analyses demonstrate
that while Ophelias apparent suicide registers the rhetoric of religious sin
that offers a narrative of appropriate punishment by Renaissance standards,
Hamlets suicidal rhetoric highlights the evolving cultural meaning of sui-
cide, removing it from its religious narrative and placing it within an emerg-
ing discourse of affect and societal disengagement. The representation of
Ophelias suicide includes the threat of legal and theological condemna-
tion while the rhetoric of Hamlets death embodies a new cultural under-
standing of suicide that avoids the stain of criminality and sin. In Ophelias
words, the two suicidal characters rue with a difference (4.5.178).
In Chapter 10, Jos L. Oncins-Martnez proposes a cognitive stylistic
approach to Shakespeares sexual language. He offers a panoramic survey
of the feld and, more specifcally, he sets out to explore the metaphors that
Shakespeare uses for the expression of sexual matters in his plays and poems
and their stylistic import. As eminent critics have demonstrated, much of
Shakespeares imagery is informed by this particular topic, but since the
1980s, this notion of imagery has been pointed out as lacking terminolog-
ical rigour. It has been notably reassessed within Conceptual Metaphor
Theory (CMT), already fruitfully used, since Lakoff and Johnsons (1980)
infuential work, for the analysis of Shakespeares style on the one hand and
Introduction 7
of sexual language in non-literary contexts on the other hand, but not for
exploring Shakespeares sexual language itself, as proposed here by Jos L.
Oncins-Martnez, with a specifc focus on the sex is war source domain.
In the fnal chapter, Amy Cook argues that cognitive linguistics
Conceptual Blending Theory (CBT) in particular illuminates the dynamic
interplay between language, cognition, and the body in ways that impact on
practical and theoretical issues in performance. The key endeavour here
is to explain how a meaning comes about, not simply what the meaning is.
She neatly demonstrates this with reference to Laura Bohannans (1995)
article Shakespeare in the bush accounting for how diverse interpreta-
tions of Shakespeares Hamlet arise. She goes on to unpack the frst sen-
tence of Richard III, using CBT as a tool in her analysis. Finally, she turns to
the embodied interplay between actor, character and spectator. The brain
can re-write the actors sense of self, so that the distinction between their
own sense of self and that of the character becomes blurred. This can even
apply to the actors sense of their own body, and Amy Cook backs this up
with reference to cognitive research on phantom limbs (i.e. limbs which are
missing but the brain believes still to be there in some respect).
Notes
1
Fifty years of the criticism of Shakespeares style: a retrospect in: Allardyce Nicoll
(ed.), Shakespeare Survey 7 Style and Language. Cambridge University Press, 1954,
I11.
2
Shakespeares Language and the Language of Shakespeares Time, in: Stanley
Wells (ed.), Shakespeare Survey 50 Shakespeare and Language. Cambridge University
Press, 1997, I17.
3
A term coined by Roger Fowler and used in a collection of essays he edited: Style
and Structure in Literature: Essays in the New Stylistics, Blackwell, 1975.
4
Second edition. Longman, introduction, 45.
5
See the proceedings of the conference: The State of Stylistics, 2008, edited by Greg
Watson. Amsterdam and New York: Rodopi.
Chapter 1
Strange deliveries: Contextualizing
Shakespeares First Citations
in the OED
Giles Goodland
1. Introduction
This chapter examines the use that the Oxford English Dictionary has made
of Shakespeares works, referring in particular to new electronic databases
that offer a wider and more easily searchable sample of published English
than was available to James Murray and his team. The large number of
quotations in which Shakespeare is the frst recorded user, has in the past
led to some criticism of the dictionary from the discipline of historical
linguistics; while, in the feld of literary studies, it has led to the occasional
but persistent impression that Shakespeare either was an extremely prolifc
coiner of words, or that he possessed an unusually large vocabulary
1
.
Historical linguists on the one hand, and literary specialists, on the
other, can be characterized as demanding different kinds of accuracy from
the OED: an accurate refection of English as it developed, versus a full and
explicatory treatment of a number of writers deemed to be great or signif-
cant. These demands can be diffcult to reconcile because a full treatment
of a single author, including atypical or rare uses of language, distorts the
overall picture of language in that writers period. Shakespeare is the most
obvious, but not the only instance. In the past, it has been hard to see
how this irreconcilability can be squared. Databases such as Early English
Books Online should help to redress this imbalance, both by antedating
and contextualizing many of Shakespeares frst citations, and by isolating
a number of instances in which Shakespeare was demonstrably using lan-
guage in an atypical way.
In certain cases, words used by Shakespeare have neither been ante-
dated nor post-dated. I will argue that these probable hapax legomena in
the language present a special case of vocabulary, explicable within a
Shakespeares First Citations in the OED 9
specifc literary context, but largely inappropriate for historical-linguistic
analysis, because they are not recorded diachronically. Their only mean-
ing resides in their single known use. This distinction should beneft
both those who use the OED as a resource in historical linguistics, and
the community of OED users who seek clarifcation of specifc literary
senses and uses.
In this chapter, some of the terminology will need explanation, since
much of it is specifc to lexicography, and even to the OED. A lemma is
any lexical item, such as a word or compound, defned or listed in a dic-
tionary. An antedating is a recorded instance of the use of a word earlier
than the previous frst use recorded in the NED. A nonce-word is a ( usually
rare) word used for the nonce, that is, according to the demand of a
specifc occasion, such as Coleridges Mammonolatry for the worship of
Mammon. A slip is a small sheet of paper on which an illustrative quota-
tion is written, containing also the necessary bibliographical details and
the lemma which the quotation illustrates. A quotation paragraph is the
section of an OED entry in which quotations for a lemma are given, in
chronological order. A coinage is a word that is formed deliberately by a
known person, usually with some attribution to prove it.
2
A frst citation,
following the use of Jrgen Schfer (1980), is the earliest citation in an
NED quotation paragraph. In this chapter also I shall use the phrase
hapax legomenon (shortened to hapax) in a slightly unconventional way, to
mean a lemma of which only one contextual instance is recorded within
a historical period.
A further comment is necessary on the way in which the OED is referred
to. When describing the original aims and the frst publication of the
dictionary, up to its frst full publication in 1928, I use NED for New
Oxford Dictionary. In 1933 the full set was republished with a Supplement,
under the title Oxford English Dictionary (OED). In the 1970s and 1980s,
four more volumes of Supplements were produced, containing new or
previously unrecorded lemmas or senses, and these were integrated into
the earlier OED in a new edition in 1989. This edition is what is now
referred to as the OED. It was not a full revision of the NED, but the
old text with the supplementary material added, so it does not affect
the data in this chapter, since attempts were not made to antedate NED
entries. In the year 2000, publication started of what is now called OED
Online. Sections of the OED are thoroughly researched and revised, and
these are published online at quarterly intervals. Work started at the
letter M, and now proceeds both alphabetically, and, more recently, in
out-of-sequence batches.
2. Shakespeare in the New English Dictionary
It is perhaps unfortunate timing that Murray and his team started work
on the NED before historical linguistics was a developed science, while by
the time of publication of the last fascicle of the NED, many of its prem-
ises would have seemed questionable from a linguistically-based point of
view. Murrays Preface to Volume I (1888), in describing the original plan
of the dictionary, states how it was resolved to extract . . . typical quotations
for the use of words, from all the great English writers of all ages. To the
linguist, it is the word great that might give pause. If it is only the great
writers who are being quoted, it may well seem doubtful that the dictionary
really can, as claimed,
furnish an adequate account of the meaning, origin, and history of
English words now in general use, or known to have been in use at any
time during the last seven hundred years [illustrated] by a series of quota-
tions ranging from the frst known occurrence of the word to the latest.
(Preface, ivv)
Considering the limited resources available to nineteenth and early
twentieth-century lexicographers, however, the choice of the great writ-
ers must have seemed reasonable, for at least two reasons. First, the texts
of great writers were more readily available. In the case of Shakespeare,
Alexander Schmidts 2-volume Shakespeare Lexicon (18741875) provided a
template for the treatment of Shakespeares vocabulary, with each head-
word, however obscure, provided with a defnition. Likewise, Murray and
his team would have been able to consult concordances of the 1611 Bible,
and the works of Milton. Resources such as these made it hard to ignore
even the most atypical words from canonical texts.
Secondly, great writers were (then as now) the ones that are more
likely to continue to be read: there was a cultural imperative to cover the
canonical writers. Failure to record a diffcult or obscure word used by
Shakespeare or Milton would be less excusable than the omission of a
word used by a less well-known writer. In fact, obscurities by Shakespeare
positively called for explication, whereas obscurities in minor or non-lit-
erary works tended to remain invisible because there was no necessity to
include them. The OED archive contains a long sequence of slips that are
marked as Not In: that is, after some procedure of evaluation, it has been
decided that the lemma that the slips illustrate are not important or com-
mon enough to merit inclusion in the OED. To a certain extent this is an
inevitable tendency, given the need to flter out noise in order to make
sense of the apparent chaos of new and unrecorded words. Even up to the
1980s, Robert Burchfeld (1992), editor of the OED Supplements, advocated
the inclusion of all vocabulary from certain twentieth-century canonical
authors.
However, compared to the other canonical texts, Milton and the Bible,
Shakespeare still illustrates a proportionally higher number of frst cita-
tions. This is because Milton and the Bible tended to use a more restricted
set of linguistic contexts than Shakespeare. As a playwright, Shakespeare
employed representations of speech as well as literary language. Scenes in
his plays (as with other playwrights of the period) moved rapidly from the
discourse of kings to that of commoners, with a high representation of clas-
ses, types and trades between. Indeed, linguistic register often provided a
shorthand that signalled location, class and character to the audience. In
addition, characters onstage manipulated forms and conventions of lan-
guage according to specifc situations, which for intrinsic reasons of stage-
craft are more various in a play than in, for instance, a narrative poem. The
nature of the dramatic action often demanded rapid changes of context.
Choice of vocabulary in plays could be a marker of social station, dialect
or personality. Different characters were often distinguished by idiolect.
All of these factors contribute to a greater linguistic variety than in other
types of text.
There are also instances (which I explore more fully in section 4), in
which Shakespeare does appear to have been deliberately coining, or, as
his character Berowne puts it, creating fre-new words (Loves Labours Lost
1.1.176). This tends to be a more subdued activity in the other canonical
texts which the NED used while building up its quotation paragraphs. Other,
non-canonical literary writers such as Thomas Nashe or Gabriel Harvey, do
seem to have coined words with an energy as great, or greater than, that
of Shakespeare. However, deliberate coinage of words in literary texts is very
hard to measure and quantify. Dictionaries have no way of assuming inten-
tionality on the part of a writer, so tend to ignore the issue. For instance, it
is not uncommon for a writer to state that they are coining a word, when in
fact earlier evidence of this word can be found. In a literary text such unme-
diated statements of intent would be rare (a character in a play, such as Don
Armado or Berowne, may declare that a word is new, but for the audience it
would be unimportant if this were true or not). In this chapter I will follow
OED practice, and not make assumptions about intentionality.
There are of course cases in English in which words have been deliber-
ately invented, and this conscious invention is on record as a frst citation,
but in general it is impossible to infer with certainty if OED Onlines frst
citations are cases of a word being invented for the frst time. Even if we
could be certain that we are seeing evidence of the frst written occurrence
of a word, this would not in itself be evidence of coinage. Often, perhaps
predominantly, a word exists in speech, or in ephemeral writings such as
letters and journals, before it is published. OED Online follows NED in not
claiming that frst citations are the actual frst occurrence of a word.
As ever, there are also complicating factors involved. Unlike in the cases
of Milton and the Authorized Version, Shakespeare did not (at least in
most cases) supervise the publication of his own works; hence, there was
more room for transmission errors and compositorial misreadings or sup-
positions than in other bodies of canonical texts. Even without making
any assumptions about Shakespeares greatness, there are clearly factors
some extrinsic to him tending to place him as a high-vocabulary writer.
The NEDs high proportion of Shakespeare frst citations was already
noted by the historical linguist Otto Jespersen in 1909:
In turning over the pages of the New English Dictionary . . . one is struck
by the frequency with which Shakespeares name is found affxed to the
earliest quotation for words or meanings. In many cases this is no doubt
due to the fact that Shakespeares vocabulary has been registered with
greater care. (234)
The case was exposed more particularly after the publication of Jrgen
Schfers (1980) Documentation in the OED. Before the age of databases or
complex computer technology, using statistical analysis based on concor-
dances and the research of a number of graduate students, Schfer demon-
strated the extent to which Shakespeare was over-represented in the OED
compared to similar writers, both in Shakespeares period (Nashe) and in
earlier periods (Malory and Wyatt). Schfer did not choose to question the
central role of literary texts in the OED. His chosen counter-examples were
all literary writers (but not playwrights); in the case of Nashe, a unique,
pre-Joycean one even more prone to neologizing and linguistic playfulness
than Shakespeare. A non-literary text would have signifcantly changed
the statistical picture that Schfer developed; but his work remains as sig-
nifcant independent evidence that Shakespeare was relied on dispropor-
tionately, compared to other writers of the period, in the NED. Essentially,
Schfer showed that Murray and his team allowed themselves to stray from
the procedure outlined in the NED Preface to extract typical quotations for
the use of words. For Shakespeare, and a small number of other writers,
atypical quotations were used to illustrate atypical lemmas. Because of this,
a statistical account of language in the period of Shakespeare, basing itself
on the un-reedited OEDs data, runs the risk of being skewed.
The obvious answer to these criticisms would be a utilitarian one, based
on the profles of those who use the OED as a source of defnitions and
factual information: as in Murrays time, and perhaps more so, accurate
information on most, if not all, of Shakespeares words is required by lit-
erary scholars. Shakespeare, in this argument, was, and remains, a special
case, in which a coverage which includes atypical uses and hapax would
seem essential.
In some cases there is a gap of 120 years or more between the entries
edited for the NED, and the text as re-edited for OED Online. Literary schol-
arship, infuenced by varying modes of historicism, has changed. It does
not just want explication of what Shakespeare meant by particular words.
The fact that almost all of Shakespeares language is presented somewhere
in the OED becomes a point of interest in itself. As editing on OED Online
progresses, a more modulated answer becomes possible; it is not that there
was too much Shakespeare, it is that there was too little of other texts and
writers. Previously unrecorded lemmas from this period are being drafted
as part of the OED revision process. In addition, added information sup-
plied by OED Online in the form of extra quotations, fuller defnitions,
more detailed etymologies and clearer labelling help us to contextualize
the Shakespeare frst citations.
3. Shakespeare in OED Online
In the last few years, electronic corpora and websites such as Literature Online
(LION) have become readily and immediately available to lexicographers
and researchers. With fully searchable databases, one of the most diffcult
aspects of lexicography in the era of Murray was resolved: examples of any
lemma could be retrieved without human error, especially without the ten-
dency to miss a signifcant word. Working on paper, only a small amount
from any text would have been converted into slips, often only the more
visible and obvious uses, at the expense of apparently less signifcant early
uses. Full searchability removed this tendency to miss signifcant evidence:
each text suddenly had a greater ease of use, and the same evidence-value
as any lexicon or concordance.
Literature Online, as its name suggests, is a source of literary texts, so part
of the same general feld of discourse that the OED covers with Shakespeare,
Milton and its many other literary sources. However, it includes a large
number of writers not generally considered to be canonical. A further and
more signifcant change in the nature of the searchable texts has come
more recently with Early English Books Online (EEBO). A considerable
amount of Early Modern material is viewable:
this incomparable collection now contains about 100,000 of over 125,000
titles listed in Pollard & Redgraves Short-Title Catalogue (14751640) and
Wings Short-Title Catalogue (16411700) and their revised editions, as well
as the Thomason Tracts (16401661) collection and the Early English Books
Tract Supplement.
and a sizable proportion of this is searchable:
To accompany the citations and page images, a separate initiative, the
Text Creation Partnership (TCP), is in the process of creating SGML cod-
ing for the full text of 25,000 EEBO works, so users can search the full
ASCII text of the documents and view both the text and the correspond-
ing original page images.
(EEBO website information, accessed May 2010)
EEBO is still unfnished, but at last it has become thinkable that all
printed Early Modern English-language texts can be searched. There are
two important factors in this: the frst is the sheer size of the EEBO data-
base, dwarfng any other resource from the period. The second, no less
signifcant, is the nature of the EEBO texts. They are not restricted to the
literary register, as was the earlier Literature Online. Because of this, large
numbers of texts that were never considered for the original NED reading
programme because of their recondite, turgid or inaccessible nature are
now instantly searchable. It is worth remembering that the NED Reading
Programme was a voluntary endeavour, appealing to the general pub-
lic, and thus limiting itself to books that were in some degree interest-
ing or appealing to its unpaid readers. In contrast, recent antedatings of
Shakespeare on OED Online that were found by means of EEBO include
sources such as a 1550 translation of Thucydides History of the Peloponnesian
War (PRACTISANT n.), A Treatise concerning Statutes (PRECURSE n.), a
1612 commentary on an Epistle of St. Paul (PREPAREDLY adv.), a 1541
defence of the marriage of priests (PRODIGIOUSLY adv.), a 1549 transla-
tion of Erasmus Praise of Folly (PROMETHEUS n.) and a 1566 book with
the title Answeare to M. J. Fekenham (RATIFIER n.). In particular, works
of theological controversy and debate have become a fruitful source of
antedatings, perhaps because they were an unattractive proposition for a
method of research that involved the close perusal of the full text. In short,
electronic searchability removes the task of having to read a text in order to
extract lexical information from it.
The graph and data below show that, in the sample of the edited portion
of OED Online, in a large proportion of cases the Shakespearian frst citations
can be antedated and contextualized. However, there are also a signifcant
number of lemmas that have not been antedated. This large remaining sub-
strate of frst citations might appear to justify the criticism of some historical
linguists that Shakespeare is over-represented in the OED. I will contextualize
some of these senses in greater detail below, but it might be worth initially
clarifying the nature of this material, with reference to the feld of academia
in which historical linguistics and literary studies meet: that of poetics.
The Russian Formalist and linguist Viktor Shklovsky argued that one
of the distinguishing features of literature is atypicality. Literature, he
argued, is in fact at its most typical when it is being unconventional in its
use of language.
3
This bold argument was taken up and modifed by the
Prague School of linguists. Jakobson and Mukarovsky, among others, dis-
tinguished literary discourse from ordinary language because it contained
or consisted of types of foregrounding that functioned by producing effects
of estrangement on its audience, thus giving literature its aesthetic quality.
Linguists of the Prague School linked this effect to that of markedness in
general linguistics: having a positive content, against a (more expected)
unmarked form. They argued that language in literature will tend in general
to have a higher content of marked than unmarked forms than non-literary
language:
The novelty, e.g. of poetic language in contrast to spoken language, the
patterning imposed by metre, the tension of plot, all devices . . . for the
aim of art, which . . . is conceived as a shock to our ordinary indifference,
as a heightening of awareness, as making strange. (Wellek: 8)
Shakespeares poetry and drama are literary, and hence distinguished
from functional or purely communicative or non-literary discourses by
effects of foregrounding; most relevantly for us, in atypical use of lan-
guage. In a non-literary text, language is a medium with which to express
some purpose or argument with clarity. In many literary texts, words, the
medium of expression, are foregrounded.
4. Graph and data
For this chapter I have chosen to focus on the NED Shakespeare frst cita-
tions in the alphabetical sequence from P to Ra-. These sections of the NED
were originally published between 1903 and 1904.
The corresponding OED Online sections were edited and published
online in the period just before, and during the frst uses of the EEBO
databases, from March 2005 to June 2008 (see Figure 1.1). Other elec-
tronic databases, notably Literature Online, had been in use for some
time. In addition, OED revision involved the searching of OED itself, the
Middle English Dictionary, material from the OEDs in-house library, the
OED (15)
12%
SL (3)
2%
EEBO (15)
12%
HC (1)
1%
HRP (3)
2%
LIB (1)
1%
LION (6)
5%
MED (8)
7%
MICH (6)
5%
NOT ANTEDATED (64)
53%
Figure 1.1 Proportion of EEBO and other provenances in sample batch. Paren-
thetical values represent associated raw frequencies. SL: slips; HC: historical corpus,
an old database of early modern material; HRP: Historical Reading Program; LIB:
from the OED library; LION: from Literature Online; MED: Middle English Dic-
tionary; MICH: slips from University of Michigans abandoned Early Modern
dictionary project.
OEDs electronic historical corpus of quotations, and the use of slips from
its own collection of quotations, as well as from the Michigan collection
that it inherited from an abandoned project to assemble an Early Modern
English Dictionary. All of these sources contributed to the antedating of
Shakespeare frst citations in this period. An ensemble of sources and data-
bases continues to be most useful; to cover not just printed works, but also
diaries, notebooks and manuscripts (when printed later), and unprinted
archival material. Thus provenances such as HRP (for Historical Reading
Program) will continue to be signifcant for OED revision in the foresee-
able future.
More recent tranches of OED Online as they are published online are
highly likely to show a higher proportion of Shakespeare antedatings,
because (in contrast to the sample discussed in the following sections)
it does not include a period before the use of EEBO in research and
editing.
5. Interpreting the data
5.1 A full list of the Shakespeare NED frst citations
in the range P-RA: (117)
Those antedated are in italics. Those citations which have been re-analy-
sed by OED Online are in square brackets.
pageant v.; pageantry n.; paiocke (pajock) n.; palate v.; pale-faced a.; palliament n.;
palmy a.; pander v.; parkward adv.; parling a.; partner v.; passado n.; paternal a.;
pauser n.; peaking a.; pebbled a.; pedant n.; pedantical a.
1
; peeping n.
2
; pellet v.; pelt-
ing n.; pendulous a.; peregrinate a.; perplex v.; persistency n.; persistive a.; person-
ating n.; perusal n.; petition v.; phantasime n.; Phoebe n.; phraseless a.; picked-hatch a.;
pig-nut n.; pilcher n.; pioned a.; pious a.; pip n.
2
; [placcate (now placate n.)];
plantage n.; pleached a.; please-man n.; plighter n.; plodder n.
2
; plodding n.; plumpy
a.; poniard n.; pooh int.; poppering n.; portage n.
2
; portcullis v.; posied a.; poster
n.
1
; pouncet-box n.; prabble n.; practisant n.; preceptial a.; precipit n.; precurrer n.;
precurse a.; predecease v.; predict n.; preformed a.; premeditated a.; [prenzie a.];
preparedly adv.; presented a.; press n.
2
; prevailment n.; preyful a.; pribble n.;
priceless a.; primogenitive n.; primy a.; printless a.; prison-gate n.; probal a.;
prodigiously adv.; proftless a.; prologue v.; Promethean a.; Prometheus n.; promising a.;
prompture n.; proof a.; propertied a.; property v.; prophetic a.; proposer n.; protester n.;
protesting n.; protractive a.; published a.; pudency n.; pugging a.; puke v.; [pulpiter n.];
pulsidge n.; pupil age n.; puppy-dog n.; purr n.
3
; push n.
3
; push-pin n.; qualify-
ing a.; quarrelsome a.; quartering a.; quatch n.
2
(now adj.); queen v.; questant n.;
questrist n.; quillet n.
2
; radiance n.; rancorous a.; ransomless a.; rant v.; ranting a.;
rat-catcher n.; rated a.; ratifer n.
NEDs placcate has been re-analysed by OED Online as sense 3 of
PLACATE n., which is attested earlier. Prenzie is now explained by OED
Online as a compositorial misreading of precise. Pulpiter was the NEDs now-
rejected emendation of Iupiter; it remains as a small-type note, since
PULPITER is attested as a lemma from 1681. Thus the crude statistics
are, out of 112 lemmas, 2 are excluded. Fifty-six have been antedated, and
54 remain unantedated.
5.1.1 Antedated Shakespeare frst citations, P-RA: (41)
First dates refer to the NEDs dating, usually based on records of frst perfor-
mance. Second dates refer to OED Onlines more conservative dating, based
on publication dates. The third date refers to the year of the antedating
quotation. In each case, OED Online can be consulted for more details.
pageant v. 1606/16091606
pale-faced a. 1592/15931570
palmy a. 1602/16041440
paternal a., 1605/1608a1450
peaking a. 1598/a16161595
pedant n. 1588/1598a1586
perplex v. 1595/a16161477
perusal n. 1600/1604?1589
Phoebe n. 1590/1600a1393
pious a. 1602/1604c1450
placcate (now placate n.) 1588/
15981567
please-man n. 1588/15981570
plodder n.
2
1588/15981584
poniard n. 1588/16001533
pooh int. 1602/16041593
poster n.
1
1605/a16161538
practisant n. 1591/a16161550
precurse a. 1602/1604a1591
premeditated a. 1590/16001583
press n.
2
1596/1598?1592
prison-gate n. 1590/16001560
probal a. 1604/a16161439
prodigiously adv. 1595/a16161541
proftless a. 1599/16001574
Prometheus n. 1588/15941549
promising a. 1601/a16161594
prophetic a. 1595a1616c1484
proposer n. 1602/16041566
protester n. 1601/a16161591
protesting n. 15991582
protractive a. 1606/16091596
published a. 1605/1608a1400
qualifying a.1606/16091582
quarrelsome a. 1596/a16161582
quillet n.
2
1588/15981576
radiance n. 1601/1608a1593
rancorous a. 1590/1597?1517
ransomless a. 1588/1594a1420
rat-catcher n. 1592/15971565
rated a. 1595/15981487
ratifer n. 1602/16041566.
It should be noted here that each antedating represents a slight equal-
izing in the over-representation of Shakespeare noted by Schfer and
others. Each antedated entry also represents an added contextualization
of a lemma that was previously a Shakespeare frst citation, and hence this
is signifcant information for the Shakespeare scholar. For instance, in the
case of PROTESTER n., NED had a citation from Julius Caesar as the only
evidence for this to mean one who makes a protestation or solemn affr-
mation, which OED Online is now able to redefne as A person who makes
an avowal or solemn declaration, esp. a person who makes a protestation
of love. OED Online now has six quotations for this sense, ranging from
1591 to 1742. On certain occasions, these antedatings expose a possible
borrowing or infuence. For instance, at PLODDER n.
2
, OEDs antedating
is interesting because of the similarity of the quotations. This from Loves
Labours Lost (1.1.86):
Small haue continuall plodders euer wonne,
Saue base aucthoritie from others Bookes.
Ironically (given the meaning of this sentence), John Lyly (Sapho and
Phao i. iii. sig. B
v
[1584]) is now shown to have used this word in 1584, in a
sentence with a similar meaning, and the same last word:
Wee silly soules are onely plodders at Ergo, whose wittes are claspt Vppe
with our bookes.
Ergo (Latin for therefore) is here presumably referring to scholastic
use of Logic.
5.1.2 Lemmas antedated due to re-dating of NED quotes: (16)
palliament n. 1588/15941593
pedantical a.
1
1588/15981592
peeping n.
2
1593/1594
(equidated with Nashe)1593
personating n. 1607/a16161615
petition v. 1607/a16161607
picked-hatch a. 1598/16021598
pip n.
2
1596/a16161604
prabble n. 1598/a16161603
preparedly adv. 1606/a16161612
prevailment n. 1590/16001599
pribble n. 1598/a16161603
Promethean a. 1588/15981594
proof a. 1592/15971583
puke v. 1600/a16161601
rant v. 1598/16041602
ranting a. 1598/a16161609.
This is a more problematic set of data.
4
In these cases, it was possible
that the word was used earlier by Shakespeare, but the only documen-
tary evidence supporting this refers to the date of performance of the
play (except for PEEPING n.
2
, in which the text of Shakespeares Lucrece
was re-dated from 1593 to 1594 to follow the dating of the English Short
Title Catalogue). In some cases, as with PUKE v., the entry has been
restructured on the basis of this redating. The proximity of the two
dates suggests in general that as a playwright, Shakespeare may have
been using words that were common in speech, but had not yet made it
into print.
The antedated Shakespeare frst citations are thus 57 out of the
117 lemmas, or 49 per cent.
6. Un-antedated Shakespeare frst citations
6.1 Followed by non-infuenced later citations within 200 years: (38)
pageantry n.
palate v.
pander v.
parling a.
passado n.
pauser n.
pebbled a.
pelting n.
pendulous a.
persistency n.
persistive a.
pig-nut n.
pilcher n.
pioned a.
plodding n.
plumpy a.
poppering n.
portcullis v.
posied a.
preceptial a.
predecease v.
preformed a.
presented a.
preyful a.
priceless a.
primy a.
printless a.
prologue v.
prompture n.
propertied a.
property v.
pudency n.
pupil age n.
puppy-dog n.
purr n.
3
push-pin n.
quartering a.
queen v.
In these cases, Shakespeare remains the recorded frst user of a word;
but later evidence recorded in OED Online shows that the word may have
had some currency at a similar time, or not long after. In these cases it is
impossible to say confdently if these words were coined or frst used in
Shakespeares texts; but the evidence would suggest that many of them
were probably current, at least in speech, at that time.
Often, the information that the Shakespeare frst citation for a word can-
not be antedated is itself interesting. When Caliban in The Tempest says
I with my long nayles will digge thee pig-nuts (2.2.167), OED Online has
no evidence that this word had not until then been used in print. This
says something about the register that Caliban is using: a regional or idi-
omatic speech far removed from the literary language of the courtiers
in The Tempest. OED Onlines following quotation, from 1693, describes
The Roots . . commonly calld Kepper-Nuts, Pignuts and Gernuts . . in the
North.
Colloquiality can perhaps explain the use of puppi-dogges instead of
puppies:
Heres a large mouth . .
That spits forth death, and mountaines, rockes, and seas,
Talkes as familiarly of roaring Lyons,
As maids of thirteene do of puppi-dogges.
(King John 2.1.4614)
Here the suggestion is that this is the kind of word a maid of thirteen
might use. Similarly, push-pin in this list is a childhood game. Then, as
now, colloquial and childhood words have a greater tendency to exist in
speech before they are used in print.
In some cases, reference to the OED Online etymology will provide
useful information. For instance, POPPERING n., a type of pear, is frst
recorded in Romeo and Juliet. The term comes from the name of a town
in Belgium that was famous for its pears. It seems to be just chance that
Shakespeare is mentioning this in English for the frst recorded time, as a
modern writer might mention a new type of soft drink, perhaps. PASSADO,
a term in fencing, from the same play, is thought to have come from Italian.
Again, Shakespeare was unlikely to have been the frst person to have bor-
rowed this term: more likely, the word was in speech at the time.
Other items in this list derive from Latin, and a distinction might use-
fully be made between Latin-derived terms (such as PENDULOUS a. and
PUDENCY n.), and those formed from English stems and affxes. A more
detailed chapter than this might distinguish between Shakespeares for-
eign, Latin and English-derived frst citations.
In other cases, these words show a distinctive literary provenance. For
instance PLUMPY a., a synonym for plump, occurs in Antony and Cleopatra
(2.7.111) in song, where plumpy fts the partly trochaic metre in a way
that plump (or other synonyms) would not: Come thou Monarch of the
Vine, / Plumpie Bacchus, with pinke eyne. Here eyne is a grammatical
archaism, suggesting a context for the unusual Plumpie.
There are other cases in which exigencies of metre, stress or rhyme may
have determined the choice of an unusual word: primy instead of prime,
perhaps to avoid two strong stresses in A Violet in the youth of primy
nature, Forward, not permanent, sweete, not lasting. (Hamlet 1.3.7) (but
producing a hendecasyllabic line); although (as with plump and plumpy),
primy and prime are of course not necessarily synonymous, and the NED
entry for the -y
1
suffx has an interesting note: In the 15th cent., if not
earlier, certain monosyllabic adjs. were extended by means of this suffx,
app. with the design of giving them a more adjectival appearance . . . In this
application the suffx has not infrequently come to express much the same
notion as -ish.
These words are also recorded as being used by other writers, so it is pos-
sible that they had an extra-literary provenance, or were part of a fund of
literary words that could be called upon at will (in a time before English
dictionaries, it is important to remember that a sense of what was verbally
possible would have been more fuid). The situation with many of these
words is probably that, even though Shakespeare may well be the frst user
of a word, he was not neologizing.
6.2 Cases in which a Shakespeare frst citation is either followed by
a gap in citations of more than 200 years, or in which all subsequent
citations refer to the Shakespeare quote in some way: (10)
paiocke (pajock) n.
parkward adv.
partner v.
pellet v.
peregrinate a.
phraseless a.
plantage n.
pleached a.
plighter n.
pouncet-box n.
These are in effect Shakespeare hapax, but such is the unique canonical
status of Shakespeare that his unusual usage has inspired later imitators.
6.3 Shakespearian hapax (lemmas not diachronically attested): (12)
phantasime n.
portage n.
2
precipit n.
precurrer n.
predict n.
primogenitive n.
pugging a.
pulsidge n.
push n.
3
quatch n.
2
(now adj.)
questant n.
questrist n.
It is categories 6.2 and 6.3 that are of most interest: although smaller
than the previous categories, they are statistically signifcant, and repay
study both from a literary and linguistic point of view. In a sense (and this
is from someone whose job it is to do this) antedating an OED headword by
using EEBO can be a banal activity; there is a probability that many, if not
most headwords can be antedated. But as Schfer notes, any antedating
simply opens up the possibility of a further antedating from a yet earlier
source. The OED quotation paragraphs, although it is not their intention,
tend to reinforce an illusory sense of the primacy of the written, and espe-
cially the printed word. The OEDs aim is to give the frst recorded instance
of a usage; for many hundreds of years, this could only have been through
the medium of writing. The NED, we should remember, did not claim that
its frst citations were frst uses.
An antedating is more primarily interesting for the information that it
conveys, which would include contextual data about the meaning or reg-
ister of the word; dating changes that enable a change in the structure of
an OED entry; or information about etymology (for instance, if the earlier
quote turns out to be from a translation). It is in this sense, of conveying
information, that not being able to antedate a word can also be signifcant.
These words in sections 6.2 and 6.3 can be shown to be explicable in
terms of either a conscious neologistic poetics occurring at certain times,
and for certain contextual reasons, within Shakespeares work; or else may
be cases of either errors at some point in the transmission of the text, or
else editorial over-interpretation.
Of the 22 words in 6.2 and 6.3 above, some are
6.3.1 Possible transmission errors
POUNCET BOX n. from 1 Henry IV (1.3.37):
Twixt his fnger and his thumbe he helde
A pouncet boxe [1623 Pouncet-box], which euer and anon
He gaue his nose, and tookt away againe.
The etymology section of OED Online cannot say with certainty where
this frst element, pouncet, originates and states that it may be a misprint
for pounced box.
PRECIPIT n. from Henry VIII (5.1.140):
Go too, You take a Precepit for no leape of danger,
And woe your owne destruction.
is, as noted in OED Onlines etymology, probably a transmission error for
PRECIPICE n. (OED Online also mentions the French precipite, meaning
precipice).
PUGGING a. occurs in The Winters Tale (4.3.7) in a song by Autolycus:
The white sheete bleaching on the hedge,
. . .
Doth set my pugging tooth an edge,
For a quart of Ale is a dish for a King.
OED Online defnes this as Of uncertain meaning; perhaps: that pulls
or tugs, thieving ; the etymology section notes that early comments some-
times interpreted it as a transmission error for prigging. With words like
this, there is not enough context to defne it with certainty.
6.3.2 Possible editorial over-interpretation
In cases where the apparent hapax is a later homonym of an established
word, editorial over-interpretation may be suspected. In the two cases in
our sample, we have PORTAGE n.
2
, treated in section 6.3.5, and PUSH n.
3

from Much Ado about Nothing (5.1.38):
There was neuer yet Philosopher,
That could endure the tooth-ake patiently,
How euer they haue writ the stile of gods,
And made a push at chance and sufferance.
OED gives this a separate homonym and relates it to the interjection
push (similar to PISH int.). This sounds persuasive (as editorial interpre-
tations should; otherwise they would not have survived); but it is asking us
to believe that Shakespeare wanted us not to think of the common noun
(from the even commoner verb) that is PUSH n.
1
In this case, this begs the
question, why did Shakespeare not use the more common form, pish?
Shewmaker (1996) simply defnes make a push at as to defy or scorn.
The remaining terms in the list can be accounted for largely in terms of a
poetic, rather than a referential meaning. These are the words that sit in
the dictionary with least happiness, because their semantic meaning is less
important than the role they play in the literary text.
6.3.3 Humorous, malapropistic or comically pedantic nonce-formations
These have two functions: as a source of humour, and as idiolect, a marker
of character. These uses tend to be confned to a proportionally small
number of characters and plays. In our sample, they are largely from his
earlier comedies: two from Loves Labours Lost, one from As You Like It, one
from The Merry Wives of Windsor and one from 2 Henry IV. Shakespeare
here may be drawing on the humorous rhetorical trope of bomphiologia,
which Sherry describes as verborum bombus when small & trifying thyn-
ges are set out wyth great grasyng words (Treat. Schemes & Tropes [1555]
sig. D.vii.), or perhaps cacozelia or mala affectatio:
Mala affectatio, euyll affectacion or leude folowyng, when the wytte lack-
eth iudgement, and fondlye folowyng a good maner of speaking, runne
into a faute, as . . . we fall into a vaine bablynge, or laboryng to be brief,
wax bare & drye. Also if we shuld saye: a phrase of building, or an audi-
ence of shepe, as a certen homely felow dyd.
PARKWARD adv. from The Merry Wives of Windsor (3.1.5):
Marry Sir, the pittie-ward, the Parke-ward: euery way: olde Windsor way,
and euery way but the Towne-way.
Here the -ward suffx is clearly being employed by the character Simple
in nonce-fashion.
PEREGRINATE a. from Loves Labours Lost (5.1.14):
Ped: He is too picked, to spruce, too affected,
to od as it were, too peregrinat as I may call it.
Curat: A most singuler and choyce Epithat.
This in context is clearly neologistic: so much so that after Holofernes
(Pedant in the Quartos directions) says this, Nathaniel (Curate) in the
stage-directions is instructed to draw out his table-book and write it down.
OED Online derives it from classical Latin peregrinatus.
PHANTASIM occurs only in Loves Labours Lost, but twice, 4.1.98 and
5.1.18. It is clearly related to the foreignness of the braggart Don Armado:
This Armado is a Spaniard that keepes here in court,
A Phantasime a Monarcho, and one that makes sport
To the Prince and his Booke-mates.
. . .
I abhorre such phanatticall phantasims, such
insociable and poynt deuise companions.
This second citation is again by Holofernes, immediately after his speech
in which he used the word peregrinate, above. The word is deliberately
strange because the thing it is describing is strange. OED Onlines etymol-
ogy is from the Italian fantasima.
PULSIDGE in 2 Henry IV (2.4.21), is a humorous blunder for pulse, and
hence is idiolectal. This is Mistress Quickly in a comedic passage:
You are in an excellent good temperalitie. Your pulsidge beates as extraor-
dinarily as heart would desire.
QUATCH a. from Alls Well that Ends Well (2.2.17) is apparently a humor-
ous variant (though earlier) of quat or squat, in the phrase quatch-
buttock:
A Barbers chaire, that fts all buttockes, the pin buttocke, the quatch-
buttocke [etc.].
6.3.4 Possible metrical determinants
A number from this list may be determined by exigencies of metre,
although in all of them there must be other considerations also at play, as
I shall attempt to uncover.
QUESTANT n. from Alls Well that Ends Well (2.1.16) is from the digni-
fed language of the King. OED Online cross-references this to the def-
nition A person or hound who quests (in various senses). The est of
questant chimes with the est of bravest, while also echoing the e sound
in wed:
You come
Not to wooe honour, but to wed it, when
The bravest questant shrinkes.
6.3.5 Possible literary estrangement effects
The last category, occurring mainly but not exclusively in the later plays,
seems to be one in which the effect is one of strangeness, of rhetorical
fourish occurring at moments of high dramatic tension, highlighting for
the audience a state of unease in character or situation:
PAJOCK n. (NED PAIOCKE) from Hamlet (3.2.272):
For thou doost know oh Damon deere
This Realme dismantled was
Of Ioue himselfe, And now raignes heere
A very very paiock.
This is glossed as a possible transmission error, perhaps meaning pea-
cock, although it occurs in a speech where Hamlet is in a heightened,
perhaps crazed, state, so may show the disjointedness of his thought. David
Crystal, among others, suggests that the meaning is savage. It follows a
previous rhyming quatrain, and Shewmaker (1996) suggests that [T]he
expected end-rhyme would have been ass (388).
PELLET v. from A Lovers Complaint (in Sonnets sig. K
v
) which the
OED defnes in a small-type note as to send or supply in the form of
pellets:
Laundring the silken fgures in the brine,
That seasoned woe had pelleted in teares.
Since the silken fgures on a napkin are the object of the verb, this
is a possible transmission error, or perhaps a suggestive and metrically
demanded elongation of the verb pelt. It is hard to see how a silken fgure
could be sent or supplied in the form of pellets. The sense could also be:
to hit or pelt with pellets. In Antony and Cleopatra (3.13.163) there is a sim-
ilar phrase: By the discandying of this pelletted storm, Lie Graveless.
PHRASELESS a. from A Lovers Complaint (Sonnets sig. L) has a gloss in
OED Online apparently: that there is no phrase to describe . Apart from
the semantic sense and the metre, there may be a conscious use of asso-
nance and alliteration in repeated a and s sounds:
Oh then aduance (of yours) that phraseles hand,
Whose white weighes downe the airy scale of praise.
PLANTAGE n. in Troilus and Cressida (3.2.173) occurs in a passage where
Troilus is describing the language that future swains will want in order to
describe him; the neologized word (perhaps a metrical lengthening of plants)
reinforces the mock-heroic clichs that become ludicrous through repetition:
As true as steele, as plantage to the moone.
As sunne to day: as turtle to her mate.
PLEACHED a. occurs twice in Much Ado about Nothing (1.2. 316 and
3.1.7), and also occurs in a different sense in Antony and Cleopatra (4.15.73).
OED defnes as Interlaced, intertwined, tangled, but it may in fact mean
something more like, as David Crystal suggests, bound together. It derives
from the much earlier verb pleach; in all occasions in poetic-sounding
verse passages:
The prince and Count Claudio walking in a thicke pleached
alley in mine orchard
(Much Ado about Nothing 1.2.8)
Bid her steale into the pleached bowere,
Where hony-suckles ripened by the sunne,
Forbid the sunne to enter. (3.1.79)
and from Antony and Cleopatra:
Wouldst thou . . see Thy Master thus with pleacht Armes,
bending downe His corrigible necke? (4.14.73)
PLIGHTER n. occurs in Antony and Cleopatra (3.13.127), again in a pas-
sage of heightened verse. OED Online defnes this as a person who makes
a pledge or promise, and gives the etymology as from PLIGHT v.
1
; it is
tempting, but not inevitable, to see a pun with PLIGHT n.
1
, or indeed
with an earlier sense of PLIGHT v.
1
, in the sense of putting someone in
danger:
My play-fellow, your hand; this Kingly Seale, And plighter of high hearts.
PORTAGE n.
2
occurs in Henry V (3.1.10) in the famous once more unto
the breech speech:
Lend the Eye a terrible aspect:
Let it pry through the portage of the Head,
Like the Brasse Cannon.
OED Online defnes this as Provision of ports or portholes; fg. in quot.
This certainly fts the sense, but we have to remember that if we accept
this, we have to also accept that Shakespeare is coining a new and fgu-
rative sense, when he could mean the existing sense of PORTAGE n.
1
, An
amount of space or weight on board a ship allowed to a mariner for his own
cargo in lieu of wages, in the sense that Shakespeare also used in Pericles
(3.1.356):
Thy losse is more then can
Thy portage quit, with all thou canst fnd heere.
It may also be a substitution for porthole, or indeed a transcription
error for the more obvious portals. More important here than the specifc
dictionary defnition, is the heightened rhetorical context that the unusual
word implies.
PRECURRER n. from Phoenix (in R. Chester Loves Martyr 170); is perhaps
easier in diction and sounds less Latinate than precursor (although OED
Online gives it a Latin etymology): Thou shriking harbinger, Foule precur-
rer of the fend.
PREDICT n. from Sonnet 14 (sig. B3
v
); it occurs in some lines with a
clipped, abbreviated tone, with no words longer than two syllables:
Nor can I fortune to breefe mynuits tell;
Pointing to each his thunder, raine and winde,
Or say with Princes if it shal go wel
By oft predict that I in heauen fnde.
PRIMOGENITIVE n. from Troilus and Cressida (1.3.106) is cross-referenced
in OED Online to PRIMOGENITURE, sense 3, The right of the frstborn child
of a family, esp. a son, to succeed or inherit property or title to the exclusion
of other claimants. There is an earlier word in English, PRIMOGENIT, a frst-
born child. It occurs in a long rhetorical speech by Ulysses, in which he asks
what happens when Degree is shakd: the long and legalistic Latinate words
perhaps suggest that there is something over-determined in his pleading:
How could Communities,
Degrees in Schooles, and Brother-hoods in Cities, . .
The primogenitiue [1609 primogenitie], and due of Byrth,
Prerogatiue of Age, . .
(But by Degree) stand in Authentique place?
QUESTRIST n. is from King Lear (3.7.15). This is perhaps an idiolectal
example: the Steward who speaks these words has a tendency to insincere
long-windedness; it occurs just after the Plucke out his eyes scene:
Thirtie of his Knights
Hot questrits [1623 Questrists] after him, met him at gate.
7. Conclusion
If we exclude the hapax from the percentage, the percentage of antedat-
ings change. 57 antedated citations out of 95 genuine words gives us
a rate of 60 per cent of Shakespeare frst citations antedated. But the
fgures also tell us that 19 per cent of Shakespeares lemmas in the sam-
ple batch were probably coinages, which seems like quite a high rate.
The category of strange or genuinely atypical words brings us back to
the feld of poetics: these are all words peculiarly hard to defne from
a linguistic point of view, but effective and unsettling aesthetically. The
literary theorist Michael Riffaterre (1983) has something very pertinent
to say here:
Literary neologisms are profoundly different from neologisms in every-
day language. The latter are coined to express a new referent or signi-
fed. They are, frst and foremost, the bearers of a new meaning, and
are not necessarily perceived to be unusual forms. A literary neologism,
by contrast, is always perceived as an anomaly, and is employed pre-
cisely because of that anomaly, sometimes even without regard for its
meaning. (iv. 62)
Riffaterre had in mind more modernist modes of literature. Neologism
has been an engine of poetry in many types of modernism, and of course
had progenitors in nonsense poetry. But Riffaterre is perhaps not saying
anything that would have surprised Shakespeare or his contemporaries.
The grammarian Richard Mulcaster, writing in 1582, wanted to give some
historical justifcation for the increasing use of new words in his period.
He wrote new occasions brought furth new words, as either more cunning
made waie to more terms, or as strange deuises did seke strange deliueries
(1st Pt. Elementarie xxii. 154). Mulcaster is making the same distinction as
Riffaterre, between ordinary neologisms, called for as occasion requires
(much like the OEDs nonce-words), and the alienating or unsettling
devices that called for expression with strange deliveries: the same that
occur, at certain points, in literature.
If we consider the Shakespeare hapax to be essentially sui generis, lexi-
cally speaking, making sense only within the language game of a poem
or play, and set apart from communicative discourse, we can exclude
them from the statistics of general Shakespeare frst citations. In this
case, the number of OED Online Shakespeare frst citations falls to a
much more reasonable 41 per cent of NED Shakespeare frst citations,
and one that promises to decrease still further as EEBO becomes more
widely searchable.
This is a useful distinction in that it along with the many antedatings
of Shakespeare tabled above helps us to square the circle between the
demands of historical linguistics upon the OED, and those of philologi-
cal or literary studies. By excluding the Shakespeare hapax from their sta-
tistics (for instance by ignoring OED Online senses or headwords labelled
as rare), the ground is evened. Shakespeare is made more equal with the
other writers of his period, such as Nashe, whose atypical uses were often
silently ignored, or not noticed, by the NED editors. At the same time, the
isolation of Shakespeares rare uses is of signifcance for literary scholars,
since it shows those passages where he was aiming for effects of strangeness
and extremity; where the action of choosing a word was replaced by the act
of creating one.
The fact that a healthy number of frst citations, despite the efforts of
the OED Online team, still remain unantedated, may still perhaps continue
the impression that Shakespeare had a large vocabulary. We can now look
at this in a different light. If you look closely at the list of unantedated
Shakespeare quotations, and for each of them, ask the question, under
what exigency did Shakespeare choose to use this word, one realizes that
the often-posed question, how many words did Shakespeare coin?, is largely
false. Like other poets and playwrights of the period, he was simply using
language to create effects. In a time period before practical, or even
remotely comprehensive dictionaries, it would not have seemed like a good
question at the time, to ask if a word was in the language.
Perhaps, in the Early Modern period, we should look at writers as not so
much possessing a vocabulary, but having an ability to generate one. No
one would advance Finnegans Wake as evidence that Joyce possessed a large
vocabulary. It shows that he was able to generate a vocabulary. The high
number of hapax in our sample shows a remarkable linguistic creativity.
Any argument about Shakespeares exceptionality, however, would need
to compare these fgures with the number of hapax in other writers of the
same genre and period. Clearly, no other writer shares exactly the same
background and characteristics as Shakespeare. But there are a few poet-
playwrights for the period for whom comparison might be more enlighten-
ing than obscuring.
It may be that Shakespeare (or any other author whose verbal creativ-
ity we wish to analyse) was simply using a word that was already known in
common speech, and it is simply chance that, after searching many data-
bases, this word remains the frst that can be cited. It may be a nonce-
word, formed for the occasion, perhaps because of the demands of metre
or rhyme. Or it may be a strange word. These are the ones that can now
with the help of EEBO, or as mediated by OED Online be shown to be
hapax. Isolating this last category will, I believe, be a useful tool in the con-
tinued analysis of Shakespeares vocabulary.
Notes
1
The most recent I am aware of is in Lerer: [Shakespeare] coined nearly six
thousand new words (129); [Shakespeare] coined words and phrases at a rate
unmatched by any previous or subsequent author (135). In a more popular for-
mat, see also McQuain and Malless.
2
See Berg (1993) for more detail on these, and a much wider glossary.
3
See Shklovsky (1998), especially Chapter 1, Art as Device.
4
This issue is discussed in more detail in Durkin (2002: 68).
References
All Shakespeare citations in this chapter are given as cited by OED
Online. NED refers to the New English Dictionary fascicles published in
19041905.
Berg, Donna Lee 1993. A Guide to the Oxford English Dictionary. Oxford and New
York: Clarendon Press.
Burchfeld, Robert 1992. Points of View: Aspects of Present-Day English. Oxford and
New York: Oxford University Press.
Durkin, Philip 2002. Changing Documentation in the Third edition of the Oxford
English Dictionary: Sixteenth-century vocabulary as a test case in: Teresa Fanego,
Beln Mndez-Naya and Elena Seoane (eds), Sounds, Words, Texts and Change.
Amsterdam: John Benjamins, 6581.
Jespersen, Otto 1909. A Modern English Grammar on Historical Principles. Heidelberg:
Carl Winter.
Lerer, Seth 2007. Inventing English: A Portable History of the Language. New York and
Chichester, West Sussex: Columbia University Press.
McQuain, Jeffrey and Malless, Stanley 1998. Coined by Shakespeare: Words and Meanings
First Penned by the Bard. Springfeld, MA: Merriam-Webster.
Murray, James 1888. New English Dictionary. Oxford: Oxford University Press.
Riffaterre, Michael 1983. Text Production. New York: Columbia University Press.
Schfer, Jrgen 1980. Documentation in the O.E.D.: Shakespeare and Nashe as Test Cases.
Oxford: Clarendon Press.
Schmidt, Alexander 18741875. Shakespeare Lexicon and Quotation Dictionary (Vols 1
and 2). Mineola, N.Y. Dover Publications.
Shewmaker, Eugene 1996. Shakespeares Language: A Glossary of Unfamiliar Words in
His Plays and Poems. New York: Facts on File.
Shklovsky, Viktor 1998. Theory of Prose, trans. Benjamin Sher. Urbana-Champaign:
University of Illinois.
Wellek, Ren 1969. The Literary Theory and Aesthetics of the Prague School. Ann Arbor:
University of Michigan.
Chapter 2
Shakespeares Vocabulary: Did it Dwarf
All Others?
Ward E. Y. Elliott and Robert J. Valenza
1. Size
Did Shakespeares vocabulary dwarf all others?
1
Many respected scholars
have said that it did and does. F. Max Mller, the great Victorian philologist
and Sanskrit scholar, said it thus:
Shakespeare, who displayed a greater variety of expression than prob-
ably any writer in any language, produced all his plays with about 15,000
words. Miltons works are built up with 8,000; and the Old Testament says
all that it has to say with 5,642 words.
By contrast, he added, English country labourers of the day supposedly
had not 300 words in their vocabulary; a well-educated person in England,
who has been at a public school and at the university, who reads his Bible,
his Shakespeare, and the Times, . . . seldom uses more than about 3,000 or
4,000 words in actual conversation; . . . and eloquent speakers may rise to a
command of 10,000 (Mller 1862: 2667, 1891: 3779). Mllers magnum
opus, The Science of Language (from which these numbers are taken), is still
in print. He was widely followed in subsequent centuries; and he is still
widely cited.
Notable twentieth-century supporters of the Mller thesis were Ernest
Weekley, who published fve editions of The Romance of Words and ten other
widely read books on words and names; Louis Marder (1962), long-time
editor of the Shakespeare Newsletter and Robert McCrum, William Cran
and Robert MacNeil, authors of the companion text to the 1986 Public
Broadcasting System series, The Story of English, now in its third, revised edi-
tion (2002). In 1928 Weekley wrote: Of Shakespeare it may be said without
fear of exaggeration that his contribution to our phraseology is ten times
Shakespeares Vocabulary: Did it Dwarf All Others? 35
greater than that of any writer to any language in the history of the world
(Weekley 1952 [1928]: 55).
Alfred Hart, the leading mid-20th-century authority on Shakespeares use
of new words, though less sweeping than Mller and the others, thought
that Shakespeares vocabulary and coinages far outstripped those of his
early contemporaries, Marlowe, Greene, Peele, Kyd, and Chapman. In 1942
he wrote:
[S]hakespeare was a lifelong and insatiable word-collector. . . .
Shakespeares unmistakable sign-manual in a play is the presence of
plenty of words peculiar to it alone. . . . [A]n author who is credited by
the compilers of the Oxford English Dictionary with being the frst user of
about 3,200 words . . . has verbal riches compelling the employment of
superlatives in describing them. . . . Shakespeares vocabulary was fuller,
more varied and expressive of fner shades of meaning than that of any
of his early contemporaries . . . . [Marlowes (and Chapmans)] inferiority
in vocabulary is manifest. (Hart 1942: 22, 28, 4501)
In 1986 McCrum et al. said this to a Public Broadcasting System audience
of millions: Shakespeare had one of the largest vocabularies of any English
writer, some 30,000 words. (Estimates of an educated persons vocabu-
lary today vary, but it is probably about half this, 15,000) (2002 [1986]:
102). The Mller thesis marches on in this century in McCrums third edi-
tion, and others heartily concur: The average educated man today . . . has
a working vocabulary of less than half that of Shakespeare(Bragg 2003:
135); Take a look at Shakespeares enormous vocabulary . . . 27,780 are dif-
ferent words. The average person . . . has a recognition vocabulary of about
5,000 words. Some of the greatest writers may have twice this capability
(Ravi 2003).
It should not be forgotten, when one is thinking of superlatives for
Shakespeares vocabulary, that English itself is said to have the richest vocab-
ulary in the world today, with at least a million words, depending on what
gets counted (Crystal 1995: 119),
2
and growing. According to McCrum et
al., English is fve to ten times richer in words than German or French (2002:
10). Shakespeare lived and wrote during the languages most glorious and
formative years, years when its vocabulary was growing explosively at rates
unmatched before or since (Nevalainen 1999: 3389), years when, driven
by patriotism, Protestantism and the mass markets both permitted and
demanded by the printing press, it grew from a collection of base, rude, rus-
tic, spoken local dialects to replace Latin as the default language of national
literary discourse (Hussey 1992 [1982]: ch. 2, McDonald 2001: ch. 1).
It wasnt just that Shakespeare had the most superlative English vocabulary
of his day if, in fact, he did. His day was the most superlative period of
vocabulary growth for English, and English vocabulary today leaves all oth-
ers in the dust. How much more superlative could you get?
Everyone in Shakespeares day was adding furiously to the language, and
judging from listings in the Oxford English Dictionary and the Chronological
English Dictionary, Shakespeares new-word coinages seem to have led all the
rest, even after many corrections for extravagant overcounts in the early
twentieth century. We argue in Part 2 that Shakespeares coinages are still
probably overcounted by a factor of at least two. For now, it is enough to note
that coinages were an important component of the persistent belief that
Shakespeares vocabulary dwarfed all others in size, variety and creativity.
Shakespeares supposedly outsized vocabulary and coinage rates are fre-
quently invoked in authorship controversies. Harts studies of Shakespeares
new-word usage fgured prominently in MacDonald Jacksons and Kenneth
Muirs landmark ascriptions of A Lovers Complaint to Shakespeare, helping
explain why A Lovers Complaints abundance of words new to Shakespeare
strengthened, rather than contradicted the Shakespeare ascription ( Jackson
1965: 819, Muir 1973 [1964]). The powerfully documented Jackson and
Muir studies overturned the consensus of the day that A Lovers Complaint
was not Shakespeares; thanks to their work, the consensus today (which
we and others now think is probably wrong, see Elliott and Valenza 1997
and 2004, Vickers 2007) is that Shakespeare did write A Lovers Complaint.
Moreover, if mainstream scholars savoured the idea of Shakespeares ver-
bal pre-eminence with their morning coffee, anti-Stratfordian insurgents
devoured it for breakfast, lunch and dinner, providing, as it seemingly did,
compelling evidence that Shakespeare was far too learned to have been
the lowly Stratford grain-dealer. Never, said Oxfordian doyen Charlton
Ogburn Jr. after a long, exclamatory discourse on Shakespeares learn-
ing, citing Mller, Marder and others, has such verbal prodigality as
Shakespeares been approached (Ogburn 1984: 2912).
To be sure, there was at least one notable oldtime Mller sceptic, not
so much of the richness of Shakespeares vocabulary as of the supposed
poverty of others vocabulary. The sceptic was Otto Jespersen (18601943).
His classic Growth and Structure of the English Language was frst published
in 1905 and last revised by him in 1938. In three tightly-packed pages on
the subject, Jespersen conceded to Shakespeare his customarily estimated
vocabulary of 20,000+ words, but he thought that Mllers notion that a
farm-labourer uses only 300 words was obviously wrong. So, it seems,
was his notion that educated persons used not more than 4,00010,000
words. Two-year-olds with doting, documenting parents were known to use
4891,121 words; Swedish peasants seemed to have more than 26,000 words;
and enterprising college professors E. S. Holden and E. H. Babbitt, using
randomly selected pages from a dictionary, estimated that most of their stu-
dents recognized a little below 60,000 words (Jespersen 1982 [1938/1905]:
2001). Jespersen added:
These statements are easily reconciled with the ascription of 20,000 words
to Shakespeare. For it must be remembered that in the case of each of us
there is a great difference between the words known . . . and the words actu-
ally used in conversation. . . . If Milton as a poet uses only 8,000 as against
Shakespeares 20,000 words, this is a natural consequence of the narrower
range of his subjects, and it is easy to prove that his vocabulary really
contained many more than the 8,000 words found in a Concordance of
his poetical works. We have only to take any page from his prose writ-
ings, and we shall meet with a great many words not in the Concordance
(202, listing 21 such words on a single page of Miltons Areopagitica).
The greatness of Shakespeares mind is therefore not shown by the fact
that he was acquainted with 20,000 words, but by the fact that he wrote
about so great a variety of subjects . . . that he needed this number of words
in his writings.(ibid.)
Jespersens cautionary book went through ten editions and must have
been read and admired by thousands of people and you can still fnd
his cautions refected and updated in the writings of at least two modern
English-language encyclopaedists, Tom McArthur (1992: 1092), and David
Crystal (1995: 123, 2004: 317). But Jespersen (unlike Mller) was out of
print for decades; not enough people have paid attention to McArthur and
Crystal; and it would seem from the continued prevalence of Mllers view
that the old-line bardolatry still holds sway.
But is it right? Our short answer, based on our analysis of the Claremont
Archive of Renaissance Texts and on type-token fgures for other writers
graciously supplied to us by Professor David Hoover, of New York University,
is that Mller and company were right that Shakespeare had a big vocabu-
lary, but wrong in supposing that it was bigger or better than other writers
vocabularies, either in his own day or since. Much of Shakespeares appar-
ent pre-eminence over others is due to the greater accessibility of his writ-
ing. He wrote more than most others and was better recorded, catalogued
and anthologized. The people who wrote the Oxford English Dictionary could
get to him in ways that they could not get to other writers. Mller, for exam-
ple, could choose between three Shakespeare concordances, notably Mary
Cowden Clarkes Complete Concordance to Shakespeare (1881 [1845]), which
had been 16 years in the making. His estimates for works with lexicons
Shakespeare, the Bible and parts of Milton were in the right ballpark;
but, where he lacked a lexicon, as with farm labourers, he was way off; and,
where, as with Milton, he had a lexicon based on a much shorter, less com-
prehensive corpus than Shakespeares, he had no way to correct for it.
What is true of Mller was largely true of his uncomputerized succes-
sors, even the best of them, Hart, who was well aware of the limitations of
handmade lexicons but could only partially compensate for them. It takes
a lexicon, an inventory of word types, to count someones total manifest
vocabulary; handmade lexicons took years to produce and were unavailable
for most other writers; and, even when available, they offered no easy way
to correct for differing corpus size, which is crucial for vocabulary compari-
sons since, other things equal, larger corpora yield larger vocabularies of
distinct words.
The most authoritative modern Shakespeare lexicons are Marvin Spevacks
(1968, 1973), a long one and a short one, both meticulously compiled on
a computer from the meticulously edited Riverside Shakespeare (Evans et al.
1997 [1974]) though it would be surprising if even the Riverside preserved
every countable distinction between one archaic spelling and another
burthen/burden; murther/murder; fadom/fathom; vild/vile; Dolphin/
Dauphin; crocadile/crocodile; wreck/wrack, and so on. Every Shakespeare
edition, in a sense, is a translation from Early Modern to contemporary
English, and variations among different editors commonizing zeal produce
defendable variations among the countable different words to be found in
each ones canon (Taylor 1989: 254, 316). Spevack machine-counted 29,066
different words types from a Riverside Shakespeare canon of 884,647
total words tokens (Spevack 1973: v). His is considered the best available
count of Shakespeares manifest vocabulary, taken from words Shakespeare
actually recorded, but even his count has limitations typical of machine
counts, our own included. It is utterly reliant on the editors orthographi-
cal judgement; it includes some words supplied by editors, and it includes
words from parts of the canon maybe as much as a tenth of it which were
not written by Shakespeare solo but by some combination of Shakespeare
and a co-author (see Elliott and Valenza 1996, Vickers 2002b, Wells 1987).
Most important, it counts every different word infection found, not just
each words root (also called its lemma, or headword). Horse and horses
are two words by this convention, not two variants of one dictionary head-
word.
3
Both of these conventions take a large view of Shakespeares vocabu-
lary, especially using an infected lexicon, which is much larger maybe
two-thirds larger than the lemmatized vocabularies used by hand-counters
like Hart.
4
If the two-thirds correction factor is good, Shakespeares lemma-
tized, manifest vocabulary is closer to 1718,000 than to 29,000 words (Hart
1943a = 17,677; Scheler 1982: 89 = 17,750; WordHoard = 18,090
5
) close
enough, given the many arguable ways of counting words, to make Mllers
estimate of 15,000 look remarkably accurate.
On the other hand, Spevacks count, like ours, understates Shakespeares
vocabulary in three ways, two small, one large. The frst small one is mul-
tiple meanings of homographs, different words with the same spelling. Our
Websters New Collegiate Dictionary (which was new in 1950) lists eight mean-
ings for spring as an intransitive verb, eight more for spring as a transitive
verb, nine for spring as a noun, and three for spring as an adjective, 28
different possible meanings in all, and up to 28 possible different words to
the context-sensitive hand counter but only one word to most computers,
including Spevacks and ours. Correcting for homographs, if you could do
it, would modestly increase the total, probably by at least the 700+ words
listed in Spevack (1968, vol. IV, App. D). The second small understatement
is linked word phrases like come on or grown up, lexical units which are
functional equivalents of a single word and mean something distinguish-
able from come and grown, yet normally dont get caught and counted
by computers.
The largest understatement of any kind of vocabulary count, machine
or hand, is the one touched on by Jespersen, that they dont measure how
many words an author knows his latent vocabulary but only those he has
used and recorded his manifest vocabulary. Some writers use the term rec-
ognition vocabulary or passive vocabulary as near-synonyms for latent,
and use vocabulary or active vocabulary as near-synonyms for manifest.
Latent /passive /recognition vocabularies are always larger than manifest/
active/use vocabularies. Works which cover many different subjects, as
Jespersen noted, should show larger vocabularies than works with fewer sub-
jects. And large corpora (other things equal) should normally show larger
vocabularies than small ones because each new token (token = countable
word, including every repetition) added to a given corpus offers a chance
of adding a new type (type = distinct word type, doesnt count repetitions)
as well, by transferring the type from the authors latent vocabulary to his
manifest vocabulary. Hart well understood this last point and, with the help
of concordances, calculated Shakespeares rates of introducing once-used,
or new words for Shakespeares poems and a few plays. Such new-to-
Shakespeare words amounted to 10 per cent of the total word types in King
Lear and appeared about once every ten lines of text in Lear and Hamlet,
once every twenty lines in earlier plays (Hart 1943b: 12831, summarized
in Jackson 1965: 112).
Hart did his best to control for date and text length, but he was badly ham-
pered by the limited concordancing available to him, which was skimpy for
everybody but Shakespeare, and which could not easily control for corpus
or sample size. Like Mller and everyone else between the nineteenth cen-
tury and the computer era, Hart had the nineteenth-century deconstruct-
ing genius for breaking down Shakespeares and others writing into their
constituent components words, grammar, metrics and so on (see Taylor
1989: 193) but he couldnt apply it as freely and systematically as you
can with computers today. All these stalwart old stylometricians were road-
bound and oxcart-borne, amazingly resourceful and profcient in using the
slow, judgemental tools they had, and hugely better than their predecessors
who lacked even roads (lexicons) and oxcarts, but gravely limited, nonethe-
less, by the technology of their time.
Computers and sizeable machine-readable text archives changed this
situation profoundly and, in 1976, made possible an astonishingly sophis-
ticated, landmark estimate of Shakespeares total recognition vocabulary.
However, the landmark appeared in Biometrika, a distinguished statistical
journal unfamiliar to most number-shy Shakespeare and language schol-
ars. It was a landmark all but invisible to regular Shakespeare scholars but
of surpassing visibility and importance to us. Even without their own text
archive, and using only the then-new Spevack concordance, plus a clever
methodology frst propounded by Sir Ronald Fisher in the 1940s, and mas-
sive crunching with an IBM 360, master statisticians Brad Efron and Ronald
Thisted, were able to make a wholesale extrapolation of Shakespeares new-
word productivity. They concluded that Shakespeare, had he doubled his
historic output of 884,647 tokens, would have added 11,460 new types, plus
or minus 150, to his known 31,564. Even more astonishingly, they showed
that Shakespeares latent vocabulary had to be at least 35,000, besides the
known 31,564, for a total vocabulary of at least 66,000 infected words
(Efron and Thisted 1976: 435). If David Crystal is right in supposing that
the English language contained about 150,000 lemmas in Shakespeares day
(Crystal 2004: 317), and if our rough estimate of Shakespeares infected-to-
lemma ratio of 1.67:1 is close to correct, it means that Shakespeare knew
about 40,000 lemmas, more than a quarter of words then in the English
language.
6
Like Fishers original study of how to forecast the discovery of new but-
terfy species from known past discovery rates (1943), the Efron-Thisted
study is still seen as a triumph of statistical ingenuity, a landmark demon-
stration of how accurately you can guess the unknown from the known,
with the right techniques. No one has tried more elaborately than we have
to retest Efrons and Thisteds main techniques, assumptions and conclu-
sions from the ground up. We found some limits for microlevel applications
(below) but concluded that their fndings for Shakespeare in the large are
still sound and well-deserving of their landmark status among statisticians
(Valenza 1990). This chapter is, in part, an effort to make it a landmark for
language and literature scholars as well.
A decade after their unheralded breakthrough, responding to Gary
Taylors ascription of Shall I Die? to Shakespeare, Thisted and Efron
attempted to apply the same methods they had applied wholesale in 1976
at retail to 500-word Shakespeare and non-Shakespeare samples compa-
rable in length to Shall I Die? They concluded that Shall I Die? had enough
Shakespeare-new words to be Shakespeares, but other sample poems by
Marlowe, Jonson and Donne did not (Thisted and Efron 1987).
7
Though it was the second of the two Thisted-Efron articles (1987) that
started the Shakespeare Clinics work, it may well be the frst (1976) which
fnishes it. Working from the second article, we found that new words
were one of the Clinics best tests but not in the way that the Mller thesis
would have led us to expect. Shakespeare did not invent and use new words
so furiously as to dwarf his peers; quite the contrary. After we calculated
Shakespeares normal, expected new-word ranges, using our Intellex soft-
ware to isolate 1,500-word blocks from the rest of the Shakespeare corpus
and to make some necessary corrections for type-token ratios, we found
(unsurprisingly, since we drew our Shakespeare profles to keep false neg-
atives at or below 5 per cent) that none of our 3,000-word blocks from
Shakespeares poems, and only 5 per cent of those from his plays, exceeded
Shakespeares expected new-word range. The surprise was that 64 per cent
of such blocks by other poets, and 56 per cent of such blocks by other
playwrights exceeded Shakespeares normal expected range of words new to
him. In other words, most of other peoples poems and plays have more
Shakespeare-new words than 95 per cent of Shakespeares. Also surpris-
ingly, not a single other-authored block fell below Shakespeares expected
new-word range.
How could this be, when everyone knew from Mller and company
that Shakespeares vocabulary and new-word coinage rates dwarfed every-
one elses? Our frst thought was that the other writers must have different
inventories of favourite words, so different as to overbalance Shakespeares
supposedly overwhelming advantage in vocabulary richness. Our second
thought was to wonder whether Shakespeares supposed advantage was as
overwhelming as everyone believed. Both of these thoughts are probably
true. The second one is what prompted us to write this chapter. We didnt and
dont doubt that he knew 29,000 words; they were all there in the 884,000-
word canon. Or that he coined hundreds, if not thousands of words. But did
it actually set him above his peers and posterity? Only if the traditional low
estimates for other writers were accurate. We believe they were not.
To get a better estimate of how Shakespeares vocabulary compared with
pertinent others, we frst controlled for size, taking large blocks of about
40,000 words from Shakespeare, eight of his contemporaries and Milton.
We gave them three tests, one traditional, two novel and doable only with
our own program, Intellex, to test raw variety, distinctiveness and total
inferred vocabulary. Table 2.1 and Table 2.2 (appendix), from which it is
drawn, show the results: no matter which test you use, Shakespeare comes
out toward the middle of the pack. If anyone towered over the others, it was
Milton and perhaps Spenser, not Shakespeare.
Test One, Types to Tokens, is a well-known but imperfect one. It measures
raw variety. Milton used an average of 6,500 different words (or types)
in each of the two 40,000-word (= 40,000-token) halves of Paradise Lost.
Shakespeare used an average of 5,470 words per block in four 40,000-word
blocks of plays and one of poetry. Types-to-tokens was too variable, overlap-
ping and sensitive to sample size for us to have used it as a reliable author
identifer for the Shakespeare Clinic, but, with sample size very large and
well controlled, it says clearly enough that, in terms of raw variety of words
used, Milton was at the top, Fletcher at the bottom, and Shakespeare and
the others in between.
Table 2.1 Three vocabulary tests: Shakespeare and eight others
Author: number of blocks Types New words Inferred vocab.
Milton: 2 6,500 1,691 131,953
Shakespeare: 5 5,470 905 62,569
6 others: 9 5,223 839 66,132
Fletcher: 2 4,444 430 49,983
Three vocabulary tests, Shakespeare, seven contemporaries and Milton, 18 40,000-word blocks
averaged for each author or author-group in each category. Milton leads the group; Fletcher
trails; Shakespeare and six others Jonson, Dekker, Marlowe, Middleton, Chapman and
Greene fall in between. All these tests measure infected, not root vocabularies. Types and
New Words are per block, averaged.
Test Two is New-to-the-Group Words, employed for the frst time in this chap-
ter. If types-tokens measures variety, New-to-the-Group Words measure dis-
tinctiveness. Using Intellex, we compiled a baseline lexicon from a corpus of
600,000 words, 160,000 by Shakespeare, the rest by seven other authors. The
600,000 tokens yielded 28,747 different types, 12,265 of them new words
found in only one of the ffteen 40,000-word blocks. Shakespeare, with
27 per cent of the groups tokens, contributed 31 per cent of the groups
new words, a bit more than his share, but far less per block than Milton,
who was, again, at the top, with Fletcher at the bottom, and Shakespeare
and others in between.
8
Test Three, Total Virtual Inferred Vocabulary (TVIV), is our newest and
most ambitious. From earlier work (Valenza 1990), we had two pieces of
solid-looking ground, to which we now add a new way of inferring from
them a writers total vocabulary, from a much smaller baseline and with
much less elaborate computation than it took Thisted and Efron to do
similar calculations for all of Shakespeare in 1976. The frst piece of solid
ground was Thisted and Efrons own bootstrap calculations, which, using
two completely different methodologies, came up with the same low-range
estimate for Shakespeares latent vocabulary of 35,000 words. This, with
his manifest vocabulary of 31,000 words, added up to a well-supportable
estimated total vocabulary of at least 66,000 words or perhaps something
closer to 64,000, if added to Shakespeares corrected manifest vocabulary
of 29,000 words per Spevacks 1974 count. Either would do for our pur-
poses, but, for consistency with Thisted-Efrons earlier article, we use the
original higher one, of 66,000 words. The second piece of solid ground
was Valenzas own massive recalculations of Thisted-Efrons data in 1990
for all of Shakespeare and all of the King James Bible. Both of these large
corpora, though differing greatly in provenance, genre, subject matter
and vocabulary richness, turned out to have normalized word frequency
curves which tracked each other almost exactly across the entire frequency
range (Figure 2.1).
If two corpora as different from each other as Shakespeare and the Bible
track each other that closely, you should be able to infer a plausible com-
parable frequency curve for other poets and playwrights. From that, you
should be able to infer their total vocabularies simply from the number
of word types they have to go through to get through half the tokens in
their corpus or sample. Another way of putting this is that (1) if an author
has a big vocabulary of different word types, an analyst should go through
more types before exhausting all the authors tokens; and (2) if different
authors word-frequency curves do track one another, the analyst should
also go through more types before exhausting half of each authors tokens,
and this adjustment makes the counting task much easier. If you accept the
Thisted-Efron fgure of 66,000 for Shakespeares total inferred vocabulary
as reasonably solid, as most do, and suppose that most word frequency
curves track one another, as is certainly so of Shakespeare and the King
James Bible and could well be so for others, you can then make a plausible
guess about a given authors total inferred vocabulary simply from count-
ing the types it takes to get through half the tokens and applying a correc-
tion factor to place the blocks frequency curve relative to Shakespeares
total vocabulary of 66,000 words. See Table 2.1, last column; in Table 2.2
(appendix) the 18 blocks by nine authors are ranked by TVIV.
The correction factors are such that taking 77 types to go through
20,000 tokens would put a given blocks TVIV just about at the level of
Shakespeares, 66,000. Higher or lower type counts at the halfway mark
would put a blocks TVIV proportionally higher or lower, as can be seen
by comparing Milton1, with 127 types at the 20,000-token halfway mark,
implying a TVIV of 140,000 types, and Fletcher1, with only 63 types at the
halfway mark, implying a TVIV of 49,000 types. The extra V, for virtual,
is our way of underscoring the speculative nature of our assumptions that
word frequency curves track each other.
Figure 2.1 A comparison of normalized word type frequency profles for
Shakespeare and the King James Bible. Shakespeare introduces new types at
roughly twice the rate at which they occur in the Bible across the entire profle.
Source: Valenza, 1990: 32.
1.00
0.50
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
0 10 20 30 40 50 60 70 80 90 100
Shakespeare
King James Bible
Do these tests make any sense? Our frst move was to scan Table 2.1 and
Table 2.2 (appendix) for consistencies and discrepancies. The consisten-
cies are many: both Milton samples cluster at the top by all three tests. Both
Fletcher samples cluster at the bottom by all three tests. Everyone else falls
in between by all three tests. Shakespeare himself, with fve blocks mea-
sured, shows the greatest range of inter-block discrepancy, with three blocks
of tragedies and romances very tightly clustered around a TVIV midpoint
of 60,000 or so, but with two outliers, poems at 84,000 and early come-
dies at 49,000. Neither of the two Shakespeare outliers is shockingly distant
from Shakespeares mean, the Thisted-Efron-estimated TVIV of 66,000: the
poems are +36 per cent, the comedies 26 per cent. Neither the long out-
lier nor the short one clashes with common-sense expectations, that poems
would be richer than plays, and tragedies richer than comedies an expec-
tation which is also confrmed by the other two tests.
The mean of all fve Shakespeare blocks tested is 62,569 words, just 5 per
cent lower than the 66,000 calculated for the entire canon by Efron and
Thisted (1976), and only 2 per cent lower if Shakespeares actual vocabu-
lary is 64,000 per Spevacks later, lower estimate. Most two-block samples
by the remaining authors, three out of the four remaining pairs, are sepa-
rated by no more than 5,000 words ( 7 per cent) from the two-sample
midpoints.
In short, though the three tests measure richness in very different ways
one (types-tokens) with no reference at all to frequency curves, one (new
words) based solely on the bottom 1 per cent of the frequency curve (least
frequent), and one (TVIV) based solely on the top, most frequent, half
they are remarkably consistent with one another and with prior work by
Thisted and Efron and in remarkable agreement that Mller and com-
pany were wrong about other writers vocabularies relative to Shakespeares.
Once you remove the gross biases of corpus size from the calculations, it
becomes clear that, if anyones vocabulary dwarfed others in size, it was
Miltons, and maybe Spensers, not Shakespeares. If so, we would conclude
that Shakespeares pre-eminence was much less a matter of how many words
he knew, than of how he used them.
The spreadsheet also has a scattering of further comparative statistics on
other writers, including a few willing members of the Claremont McKenna
College faculty, to the same general effect: the frst fve colleagues we could
get to give us recognition percentages of three sample pages from the dic-
tionary all turned out, after extrapolation, to have vocabularies fve or ten
times larger than the 510,000 words confdently and wrongly assigned
to educated people by Mller and his modern followers. It also turns out,
not surprisingly, that we were not the frst to think of this simple, obvious,
dictionary-sampling methodology, which was equally available to Mller
and the others, but apparently never tried by them. Jespersen and Crystal
both mention such tests, and with similar outcomes, that is, crediting run-
of-the-mill college-educated moderns with Shakespeare-sized or larger
infected vocabularies of 5080,000 words (Crystal 1995: 123, Jespersen
1982 [1938/1905]: 201).
TVIVs project even larger infected vocabularies for other people than
Shakespeare, including two moderns well known to us who are not other-
wise to be compared with Shakespeare: 132,000 words for Spenser, 96,000
for Melville, 81,000 for the Iliad, almost 100,000 for Elliott and Valenzas
manuscript on Shakespeare, and 177,000 for Elliotts alone on population
policy. Jespersen and Crystal were in no position to test peoples TVIVs the
way we have, but our results are entirely consonant with two of Jespersens
views, that modern vocabularies equal or exceed Shakespeares, and that
vocabulary richness should increase with the variety of subjects addressed
(Jespersen 1982 [1938/1905]: 202); hence the much greater TVIV for
Elliotts wide- ranging population manuscript than for our more tightly-
focused Shakespeare book.
People who write about Shakespeare often use much fancier language
than Shakespeare did himself. One need only open a late twentieth-century
Shakespeare journal to fnd words far more abstruse than Shakespeares
fetishisation, commodifcation, poststructuralism, inferred virtual vocabulary and
enclitic microphrases, for example. Some of these inkhorn polysyllables
are our own, and there is a place for them, but they are all far, far from
Shakespeare. Shakespeare managed to do some of his best work with short,
concrete words like seething brains, or cool reason, artfully combined:
. . . I never may believe
These antic fables, nor these fairy toys.
Lovers and madmen have such seething brains,
Such shaping fantasies, that apprehend
More than cool reason ever comprehends.
(A Midsummer Nights Dream 5.1.26)
The result would be the same with any set of Shakespeares most famous
lines Friends, Romans, countrymen, All the worlds a stage, To-morrow,
and to-morrow, and to-morrow, and many others.
In terms of sheer numbers of words known, and by several different
measures, many of us can match or exceed Shakespeare unsurprisingly,
perhaps, given the vast expansion of the language and the whole human
enterprise since Shakespeares time. So could a number of his own con-
temporaries who drank from the same Castalian fountain of words that
Shakespeare did. Quantitatively, Mller and company were completely
wrong. Qualitatively, in terms not of how many words Shakespeare knew,
but of how well he used them, we would suppose that there is still plenty of
room for bardolatry. Shakespeare learned early how to strike deep, not with
an outsize inventory of long, inkhorn words, but with a par-for-the-course
inventory, mostly of plain words, surpassingly well chosen and put together.
Its not too soon, or too late, for the rest of us to learn it too.
2. Shakespeares word coinages
What about the other Shakespeare commonplace, that he invented more
new words than anyone else? That, too, could well be a myth, but we do
not yet have the evidence that would settle the question. Some of todays
authorities still rely on old, extravagant overcounts of Shakespeares new-
word coinages. Seth Lerer tells us, for example, that Shakespeare coined
nearly six thousand new words at a rate unmatched by any previous or sub-
sequent author (2007: 129, 135). Most other modern estimates are much
lower, clustering around 1,700,
9
still high enough for Shakespeare to sur-
pass all others. But, as Giles Goodlands Strange deliveries, in this volume,
Chapter 1, shows, current estimates could still be twice too high, and there
is little evidence that the defationary process begun in Strange deliveries
has run its full course.
Bryan Garner, using the Chronological English Dictionary, counted 10,302
loanword neologisms added to the English language between 1580 and
1619, a 39-year span (1982: 151). Shakespeares writing life, which ran from
about 1590 to 1613, encompassed almost 60 per cent of this period, good
for, say, 6,000 of the whole periods new, mostly Latinate loanwords. Garner
thought that Shakespeare contributed just over 600 of them, about a
tenth. Latinate loanwords are only a fraction of all newly coined words, but
Garners one-tenth estimate for loanwords both matches and cites estimates
by Joseph T. Shipley (1977: 28), and is based on tables in Jrgen Schfer
(1973: 20620) that Shakespeare was the frst user of well over 1,700 words
one new word in every ten. Shakespeare, Shipley concluded, was the
greatest word-maker of them all (cited, Garner 1982: 153). David Crystal,
having scanned the electronic OED for coinages, and subtracted nonsense
words like gratillity but not parallel words used by others at about the
same time found 2,035 lexeme coinages for Shakespeare, about a tenth
of Shakespeares 20,000 or so total headwords, and many more than Nashe
(around 800 coinages), Spenser (c. 500), Sidney (c. 400) or Marston (c. 200);
(Crystal 2004: 3278).
Following Crystal and Goodland, we would guess that the 2,000+ coined
words estimate is probably still too high for Shakespeares actual contri-
butions of coined words relative to others, for some of the same reasons
we would guess that his overall vocabulary is still overestimated relative
to others: Shakespeares words were well recorded and catalogued, and
readily accessible to the compilers of the original Oxford English Dictionary;
other peoples words were not (Crystal 2004, Jespersen 1982 [1938/1905]:
211, Schfer 1980). His frst-uses would routinely be dated from the frst
recorded mention of the play whose later publication turned out to contain
it. Lesser writers new words were generally dated from frst publication, not
frst mention. The OED2 now dates everyones frst-uses from frst publica-
tion, not frst-mention, and, as Goodland has shown, fxing that accounting
bias has cost Shakespeare many frst uses.
Shakespeare gets credit for the frst recorded mention of words like
sblood and Newgate, but its hardly likely that he originated these terms.
He also gets credit for putting together composite and compound words
like unreal, worthless, well-read or worm-hole, whose constituent elements were
not his invention (Crystal 2004: 325). Others likewise get credit for such
composites, but one wonders whether Shakespeare doesnt proft dispro-
portionately from such accounting. He gets more credit for malapropisms,
nonce-words and proper names than, say, Thomas Nashe. Shakespeare had
many such words, and it seems odd for us to count them as coinages if
Shakespeare did not intend them as such. Fewer of his new words were
overlooked in the OED; and more parallel citations of new words from
the same year are credited to him than to others (Goodland 2011, Schfer
1980: ch. 2). According to Crystal, up to 644 headwords used by others
within 25 years of Shakespeares frst use could be described as parallel
citations. If we divide these half and half between Shakespeare and others,
it gives him 1,713 coinages, about the going rate today for Shakespeare
coinage estimates (Crystal 2004: 326).
Further whittling of the total will be done as more of other peoples
writings gets digitized, and as co-authored parts of the canon get more
clearly distinguished from Shakespeares parts. Both of these are happen-
ing apace see Goodlands Strange deliveries, Chapter 1 in the present
volume, and Vickers 2002a and 2002b. If someone else wrote A Lovers
Complaint or the frst two acts of Pericles, Shakespeare should not get the
credit for the coinages these contain. Moreover, over 900 of Shakespeares
coinages were words like adoptious, which never caught on with others
and hence, as coinages, are more like private IOUs or confederate money
than like actual current coin of the realm (Crystal 2004: 326; see the dis-
cussion of nonsense and nonce-words above). When George W. Bush
comes up with Bushisms like subsidation, analyzation, hopefuller, more few
and explorationists, we suppose that he is struggling to follow accepted
rules of word formation (see Crystal 2004: 3145) but has gotten in over
his head. Everyone sniffs at such gaffes, and no one praises them as addi-
tions to the language (http://slate.msn.com/id/76886/). If Bush gave
us words like insultment, omittance, opulency, revengive, thoughten or casted,
these would likewise be gathered and laughed at as Bushisms. But it was
not Bush who gave us the second set, it was Shakespeare and his gaffes
are hailed as brilliant landmarks of linguistic daring, fresh evidence of
his peerless mastery of the language, 24-carat coinages for Shakespeare
that would be dismissed as pot-metal if they came from someone else. It
is probably true that the lines between obvious gaffes and permissible cre-
ative license are clearer now than they were in Shakespeares time,
10
but
a coinage still in circulation today, like acutely, carries more weight than
one like insultment that never caught on and would be a gaffe or a mala-
propism if it came from anyone else. Again, we suspect that Shakespeares
coinage list would be considerably shorter if he were held to the same
standard as others.
Counting coinages has most of the problems of counting other words
making due allowance for corpus size, latency, infections, multiple mean-
ings and so on plus special problems of its own, which are harder to
do with computers and have been less explored: sorting out what counts
as a word, whether it is one persons real coinage, and not anothers,
and deciding whether it has caught on. The bigger, better and more
searchable the corpus of available comparison texts, the more and better
such sorting can be done. David Crystal has pushed this kind of anal-
ysis farther than most, but it is still in its earliest stages (Crystal 2004:
31729). We do suspect from a Jrgen Schfer estimate (1980: ch. 4), that
Shakespeares rate of innovation units per thousand words of corpus
is lower than Nashes, but we dont know yet how it compares with, say,
Jonson, whose total corpus and inferred vocabulary seem similar in size
to Shakespeares, nor with Milton, whose inferred vocabulary looks much
larger than Shakespeares, but who wrote too late to get in on the Early-
Modern orgy of word-creation. After Crystals initial whittling down (but
not Schfers or Goodlands), Shakespeare still retained an impressive
residue of core coinages which exceeded what would be left of Nashe
and the others similarly whittled (Crystal 2004: 327) but not nearly as
other-dwarfngly as one would have thought from reading Mller, Marder
or Shipley. Schfers (1980) computations of authorial innovation rates
per thousand published words would put Shakespeare behind Nashe, just
as our three different estimates of vocabulary richness per 40,000-word
block put him behind Milton and Spenser. Over the years, as analysis has
grown more carefully focused, and the impacts of bardolatry better con-
trolled, Shakespeares coinage edge over his peers has not grown, but
steadily shrunk.
Giles Goodland shows us that much more shrinkage can be expected.
He started with what looks like a baseline of 122 words beginning with P,
Q and Ra-, about 7 per cent of Shakespeares estimated 1,700 coinages,
and re-examined them against just 40 months of then newly-recorded
Renaissance e-texts. From the baseline he subtracted 41 words antedated
by the new texts (about one a month); 16 words redated by rules for
the frst time applied equally to Shakespeare and others; and 17 words
that looked like intended nonce-words, not coinages. These reductions
cut the baseline by 60 per cent, with no allowance for further antedat-
ings or for the three surviving Shakespeare coinages from A Lovers
Complaint, which Shakespeare may not have written posied, pellet,
and phraseless. If all of Shakespeares estimated 1,700-word coinages
were also deflated by a comparable 60 per cent, it would leave only
680 coinages, with an overall ongoing shrinkage of about 14 words a
month. That would be less than Crystals 2004 estimate for Nashe (2004:
3278) and consistent with Schfers conclusion that Nashe had higher
rates of innovation than Shakespeare (1980: ch. 2). We would expect
Nashes coinages, like Shakespeares, to suffer some attrition from recent
and future additions to Early-Modern e-text databases but, following
Schfer, would be surprised if his coinage counts turned out to be as
grossly infated as Shakespeares.
In fairness to Shakespeare, again following Crystal, we should note that
our simple word-counting does not give Shakespeare (or others) due credit
for what could be hundreds of cases where he (or they) added a new sense
or meaning to an already-known word (Crystal 2004: 31821). It is possible
that Shakespeare would do better than most by this hard-to-count, qual-
itative standard and might yet be found to surpass the others even after
further massive defation but this is a different way of stating the theme
of this chapter: that Shakespeares genius is not so well measured by the
number of words he used or coined, as by the way he used them.
Notes
1
Our thanks to David Crystal, David Hoover, MacDonald Jackson, and three anony-
mous readers for helping us try to answer this question. And our congratulations
to Hugh Craig, of the University of Newcastle, Australia, for his impressive study,
Shakespeares Vocabulary: Myth and Reality, forthcoming in the Shakespeare
Quarterly, which we have had a chance to read in manuscript. He arrived indepen-
dently at conclusions very similar to ours: Shakespeare is in fact no different from
his contemporaries in the number of different words he uses.
2
Some encyclopaedists say that English could have as many as a billion words, count-
ing proper names, abbreviations, and scientifc terms (McArthur 1992: 1091).
3
Horse can also be a morpheme, the smallest linguistic unit of a generated or com-
pounded headword (like horsemanship) that has semantic meaning. Horsemanship
has three morphemes, horse, -man, and -ship. In principle, a dictionary stripped of
compounded headwords found in the language and reduced to nothing but mor-
phemes would be even shorter than conventional dictionaries. But we know of no
one who has tried to de-compound all of Shakespeares words into morphemes,
and we have not tried it ourselves.
4
Harts handcounting also permitted him to eliminate proper nouns from his word
counts, a task which is harder for machine counts.
5
http://wordhoard.northwestern.edu/userman/index.html.
6
Efron and Thisteds estimate had a problem which we consider a small one: it
seems to be based on a preliminary 1968 Spevack estimate of 31,654 Shakespeare
word types, and not on their fnal 1974 estimate of 29,066 types. No one can
remember what caused the difference between the two estimates. Possibly some
of the poems from The Passionate Pilgrim were assigned to Shakespeare in the frst
estimate but not the second. Given the variety of ways a word or a corpus could
legitimately be defned, and the fact that both estimates are in the same ballpark
with generations of prior estimates, we consider the 7 per cent difference between
the two estimates of manifest vocabulary of minor importance compared to the
major breakthrough that the two statisticians made in inferring a large latent
vocabulary from a smaller manifest one of either size.
7
After many years of crunching, we (and most others, such as Foster 1987, Pendle-
ton 1989, and Vickers 2002a) doubt that Shall I Die? is Shakespeares; but Thisted
and Efron, in fact, had only claimed that it was a could-be for Shakespeare, not a
must-be. We also doubt that Thisted-Efron tests are valid for samples as short as
500 words. But we consider these tests well-validated for samples of 1,000 words or
more and copiously validated for the entire Shakespeare corpus of 884,000 words.
Thisted-Efrons 1987 article was the primary inspiration for our Claremont Shake-
speare Clinic, a series of student teams using computers to test the works of 37
Shakespeare Claimants, along with 30-odd poems and plays of the Shakespeare
Apocrypha and Dubitanda, for Shakespeare authorship (Elliott and Valenza 1996).
8
One could argue that Miltons Paradise Lost had a different subject-matter vocab-
ulary from those of the playwrights tested, and would therefore be expected to
add more to the groups new-word inventory than a sample with the same subject-
matter as the others or that Shakespeare, who contributed at least a quarter
of the groups tokens, might have contributed fewer words per block that were

new to himself. Both could be so, but probably not enough so to bring Shake-
speare to Miltons level. Shakespeares own poems score higher on all three of
our tests than his tested tragedies, which, in turn, score higher than his tested
comedies. But even his poems score below Miltons, and also below two Dekker
plays. Further testing with other combinations of texts could explore these pos-
sibilities, but we believe that our comparison of large, standardized blocks from
a variety of different authors is far more indicative of their relative vocabularies
than the traditional alternative of comparing whole corpora of vastly different
sizes.
9
Schfer (1973: 20620; Shipley (1977: 28); Garner (1982: 151); Crystal (2004:
326); Hitchings (2008: 124); McQuain (1998 viii [1,500]).
10
It is also probably true that the Shakespeare list of Bushisms if read in context,
could be discounted in several ways. Revengive KL 2.1.48, is found in two Lear
Quartos, but the Folio has it as reuenging. Omittance AY 5.5.133; insultment
Cym., 5.5.141; opulency Tim., 5.1.387; and perhaps even casted H5 4.1.23, are spo-
ken by characters of whom Shakespeare was making fun for trying to speak beyond
their abilities: respectively, Phoebe, Cloten, Poet, and Henry V. It is doubtful that
Shakespeare himself intended them as serious words, yet lexicographers count his
Bushisms but not Bushs, at full value. It seems to us a double standard.
References
All Shakespeare references are to Evans, G. Blakemore and J. J. M. Tobin
(eds), (1997 [1974]). The Riverside Shakespeare. Boston: Houghton Miffin
Company.
Bragg, Melvyn 2003. The Adventure of English. New York: Arcade Publishing.
Clarke, Mary Cowden 1881 [1845]. The Complete Concordance to Shakespeare: Being
a Verbal Index to All the Passages in the Dramatic Works of the Poet, Revised Edition.
London: Bickers & Son.
Crystal, David 1995. The Cambridge Encyclopaedia of the English Language. Cambridge
University Press.
Crystal, David 2004. The Stories of English. Woodstock and New York: The Overlook
Press.
Efron, Brad and Ronald Thisted 1976. Estimating the number of unseen species:
how many words did Shakespeare know?, Biometrika 63: 43547.
Elliott, Ward E. Y. and Robert J. Valenza 1996. And then there were none: winnow-
ing the Shakespeare claimants, Computers and the Humanities 30: 191245.
Elliott, Ward E. Y. and Robert J. Valenza 1997. Glass slippers and seven-league
boots: C-prompted doubts about ascribing A Funeral Elegy and A Lovers Complaint
to Shakespeare, Shakespeare Quarterly 48: 177207.
Elliott, Ward E. Y. and Robert J. Valenza 2004. Did Shakespeare write A Lovers
Complaint? The Jackson ascription revisited in: Brian Boyd, (ed.), Words That
Count: Essays on Early Modern Authorship in Honor of MacDonald P. Jackson. University
of Delaware Press, 11740.
Finkenstaedt, Thomas, Ernst Leisi, and Dieter Wolff 1970. A Chronological English
Dictionary Listing 80,000 Words in Order of their Earliest Known Occurrence.
Heidelberg: C. Winter, Universittsverlag.
Fisher, Ronald A. et al. 1943. The relation between the number of species and the
number of individuals in a random sample of an animal population, Journal of
Animal Ecology 12: 4258.
Foster, Donald W. 1987. Shall I die? post mortem: defning Shakespeare,
Shakespeare Quarterly 38 (Spring): 5877.
Foster, Donald W. 1989. Elegy by W. S.: A Study in Attribution. Newark, Delaware:
University of Delaware Press; London and Toronto: Associated University Presses.
Garner, Bryan A. 1982. Shakespeares latinate neologisms, Shakespeare Studies 15:
14970.
Goodland, Giles 2011. Strange deliveries: Contextualizing Shakespeares First
Citations in the OED, in this volume, Chapter 1.
Hart, Alfred 1942. Stolne and Surreptitious Copies, a Comparative Study of Shakespeares
Bad Quartos. Melbourne and London: Melbourne University Press.
Hart, Alfred 1943a. Growth of Shakespeares vocabulary, Review of English Studies
19: 24254.
Hart, Alfred 1943b. Vocabularies of Shakespeares plays, Review of English Studies
19: 12840.
Hitchings, Henry 2008. The Secret Life of Words: How English Became English. New York:
Farrar, Straus, and Genoux.
Hussey, Stanley S. 1992 [1982]. The Literary Language of Shakespeare. Harlow, Essex:
Longman Group Limited.
Jackson, MacDonald P. 1965. Shakespeares A Lovers Complaint: Its Date and Authenticity.
Auckland: University of Auckland.
Jespersen, Otto 1982 [1938/1905]. Growth and Structure of the English Language.
Chicago: The University of Chicago Press.
Lerer, Seth 2007. Inventing English: A Portable History of the Language. New York:
Columbia University Press.
Marder, Louis 1962. His Exits and Entrances: The Story of Shakespeares Reputation.
Philadelphia and New York: J.B. Lippincott Company.
McArthur, Tom (ed.) 1992 [1902]. The Oxford Companion to the English Language.
Oxford and New York: Oxford University Press.
McCrum, Robert, Robert MacNeil and William Cran 2002 [1986]. The Story of
English. New York: Penguin Books.
McDonald, Russ 2001. Shakespeare and the Arts of Language. Oxford: Oxford University
Press.
McQuain, Jeffrey and Stanley Malless 1998. Coined by Shakespeare: Words and Meanings
First Penned by the Bard. Springfeld, MA: Merriam-Webster.
Muir, Kenneth 1973 [1964]. A Lovers Complaint: a reconsideration in:
Shakespeare the Professional and Related Studies. London: Heinemann, 20419.
Mller, F. Max 1862. Lectures on the Science of Language. New York: Charles
Scribner.
Mller, F. Max 1891. The Science of Language, Founded on Lectures Delivered at the Royal
Institution in 1861 and 1863. New York: Charles Scribners Sons.
Nevalainen, Terttu 1999. Early Modern English Lexis and Semantics, in: Roger
Lass (ed.), The Cambridge History of the English Language, 1476 to 1776. Cambridge:
Cambridge University Press, volume 3, 771.
Ogburn, Charlton, Jr. 1984. The Mysterious William Shakespeare: The Myth and the
Reality. New York: Dodd, Mead, & Company.
Pendleton, Thomas A. 1989. The Non-Shakespearean Language of Shall I Die? ,
Review of English Studies 40 (August): 32351.
Ravi, V. S. 2003. No doubts about his genius, The Hindu, 31 August 2003.
Schfer, Jrgen 1973. Shakespeares Stil: Germanisches und Romanisches Vocabular.
Frankfurt am Main: Athenaeum Verlag.
Schfer, Jrgen 1980. Documentation in the O.E.D.: Shakespeare and Nashe as Test Cases.
Oxford: Clarendon Press.
Scheler, Manfred 1982. Shakespeares Englisch: Eine sprachwissenschaftliche Einfhrung.
Berlin: Schmidt.
Shipley, Joseph T. 1977. In Praise of English: The Growth and Use of Language. New
York: Times Books.
Spevack, Marvin1968. A Complete and Systematic Concordance to the Works of Shakespeare.
Hildesheim: Georg Olms.
Spevack, Marvin 1973. The Harvard Concordance to Shakespeare. Cambridge, MA:
Belknap Press of Harvard University Press.
Taylor, Gary 1989. Reinventing Shakespeare: A Cultural History from the Restoration to the
Present. New York: Weidenfeld & Nicholson.
Thisted, Ronald and Brad Efron 1987. Did Shakespeare write a newly-discovered
poem?, Biometrika 74(3): 44555.
Valenza, Robert J. 1990. Are the Thisted-Efron authorship tests valid?, Computers
and the Humanities 25: 2746.
Vickers, Brian 2002a. Counterfeiting Shakespeare: Evidence, Authorship, and John
Fords Funerall Elegye. Cambridge, UK, and New York: Cambridge University
Press.
Vickers, Brian 2002b. Shakespeare, Co-author: A Historical Study of Five Collaborative
Plays. Oxford and New York: Oxford University Press.
Vickers, Brian 2007. Shakespeare, A Lovers Complaint, and John Davies of Hereford.
Cambridge and New York: Cambridge University Press.
Weekley, Ernest 1952 [1928]. The English Language. London: Andre Deutsch
Limited.
Wells, Stanley W., Gary Taylor, John Jowett, and William Montgomery 1987. William
Shakespeare: a Textual Companion. Oxford and New York: Clarendon Press, Oxford
University Press.
Name Includes
40,000 words
each from
Tokens Types NW 1/2
tokens
@
Total Virtual
Inferred
Vocab.
MILTON1 Paradise Lost, 1-6 40,000 6,671 1,790 127 140,415
MILTON2 Paradise Lost, 7-12 38,886 6,328 1,591 117 123,490
DEKKER* Whore of Babylon,
Honest Whore
40,000 5,742 1,093 92 85,567
SHPOEMS* Venus, Lucrece,
Sonnets
40,000 6,054 1,147 91 84,180
MARLOWE1* Tamburlaine, I & II,
DF16
40,000 5,444 936 83 73,449
JONSON1* Sejanus, Volpone 40,000 5,452 576 81 70,867
GREENE* Alphonus, Friar
Bacon, James IV
40,000 5,042 667 79 68,324
MIDDLE1* Phoenix, Michael-
mas, Chaste Maid
40,000 4,761 883 77 65,822
MIDDLE2* Witch, Hengist,
Women Beware
Women
40,000 5,006 833 76 64,586
JONSON2* Alchemist,
Bartholomew Fair
40,000 5,857 1,076 74 62,144
SHTEMPWT* Tempest, Winter's
Tale
40,000 5,841 964 73 60,938
SHCORCYM* Coriolanus,
Cymbeline
40,000 5,341 879 72 59,742
SHOTHLR* Othello, Lear 40,000 5,221 865 71 58,556
MARLOWE2* Jew of Malta, Edward
II, Massacre
40,000 4,892 720 66 52,777
CHAPMAN* Gentleman Usher,
Bussy D'Ambois
40,000 4,811 767 65 51,652
FLETCHER2* Barnavelt, Island
Princess
40,000 4,460 477 64 50,536
SH3COMS Shrew, TGV, Comedy
of Errors
40,000 4,892 672 63 49,430
FLETCHER1* Woman's Prize,
Valentinian
40,000 4,427 382 63 49,430
Continued
Appendix
Table 2.2 Types-tokens, new-to-the-group words, total inferred vocabulary,
Shakespeare and other writers
By all three measures, Shakespeares poems had richer vocabularies than
his tragedies and romances, which, in turn, were richer than his early come-
dies. On average, Shakespeares writings were about as rich as those of most
of his contemporaries tested, and richer than Fletchers, but not as rich as
Miltons in Paradise Lost or Spensers in The Faerie Queene, Book 1, or the last
part of Moby Dick, or two modern non-fction texts by Elliott and Valenza.
Name Includes
40,000 words
each from
Tokens Types NW 1/2
tokens
@
Total Virtual
Inferred
Vocab.
Milton
Average
6,500 1,691 122 131,953
Shakespeare
Average
5,470 905 74 62,569
6 Other
Playwrights
Average
5,223 839 77 66,132
Fletcher
Average
4,444 430 64 49,983
* = in baseline composite lexicon for eight authors Sh NW 3,855 =31%
All NW 12,265
18 blocks by 9 authors, 15 blocks by 8 authors in baseline lexicon; 600,000 words, 4 blocks by Shakespeare,
160k words = 27%
Total composite lexicon: 28,747 words, 12,265 of them NW (43.1%)
All calculations done with Intellex
Spenser Faerie Queene, Book
1, old spelling
40,000 6,235 122 131,827
Molire L'Avare, Don Juan, in
French
39,253 5,003 69 56,215
Elliott Politics of Popula-
tion, 40k words
40,000 7,067 147 177,275
Elliott &
Valenza
Shakespeare Book
40k words
40,000 5,810 102 99,984
Melville Moby Dick Last
40k words
40,000 7,056 99 95,553
Iliad 3-40 Iliad Block 3 First
40k words
40,000 3,985 89 81,437
Odyssey 2-40 Odyssey Block 2 First
40k words
40,000 4,023 58 44,053
David Hoover Type-Token fgures, all done with TACT
Paradise Lost 1 40,000 7,005
Paradise Lost 2 40,000 6,860
Sh. Poems Sonnets, Venus,
Lucrece
40,000 6,098
Moby Dick 1 40,000 6,554
Moby Dick 2 40,000 7,339
Moby Dick 3 40,000 6,855
Moby Dick 4 40,000 7,171
Moby Dick 5 40,000 6,937
Lewis,
Main St. 1
40,000 7,285
Lewis,
Main St. 2
40,000 7,119
Continued
Name Includes
40,000 words
each from
Tokens Types NW 1/2
tokens
@
Total Virtual
Inferred
Vocab.
Lewis,
Main St. 3
40,000 6,996
Lewis,
Main St. 4
40,000 6,381
Lewis, 367K Babbitt, Main St.,
Our Mr. Wrenn
367,609 24,105
Melville, 612 K Moby Dick, Pierre,
Redburn, Benito
C., Omoo
612,647 31,979
Other large corpora
Shakespeare
884K entire
canon per
Spevack
884,647 29,066
Homer Iliad & Odyssey,
Samuel Butler
Translation
269,779 13,041
King James
Bible
Entire King James
Version
789,955 12,803
Milton, all Para Lost; Para
Regained, Samson;
young
117,693 14,899
Milton, Para
Lost
Paradise Lost 80,779 11,244
Sh. Average, 6x40,000word blocks 62,569
Thisted-Efron estimate of Shakespeare's Total Vocabulary, minimum 66,000
Chapter 3
A New Kind of Dictionary for Shakespeares
Plays: an Immodest Proposal
1
Jonathan Culpeper
1. Introduction
The best-known classic Shakespearean dictionary is probably Charles T.
Onionss Glossary (1986 [1911]), written in the philological tradition that
characterized the Oxford English Dictionary (OED), and providing pithy def-
initions and illustrative quotations.
2
The proposed dictionary of the lan-
guage of Shakespeares plays is analogous to more recent developments in
dictionaries of general English, and, more specifcally, the departure from
the philological tradition brought about by the Collins Cobuild Dictionary of
the English Language (Sinclair 1987). The Collins Cobuild is a corpus-based
dictionary. This implies both a particular methodology for revealing mean-
ings and a particular theoretical approach to meaning, as we shall see in
this chapter. In particular, there is a strong empirical emphasis. There is less
reliance on the vagaries and biases of editors, and a greater focus on the
evidence of usage. The question of what does X mean? is pursued through
another question: how is X used? To answer the how question, corpus
approaches deploy the whole gamut of computational techniques, in order
to reveal patterns of usage in context. This inevitably involves matters of
frequency. Frequency is not in fact as alien as it might seem to the literary
critical ear. Any textual analysis that identifes a pattern implicitly involves
frequency, as a pattern is the (full or partial) repetition of elements. In fact,
the proposed dictionary goes beyond what one might fnd in the Collins
Cobuild in a number of ways. Crucially, an additional feature proposed for
the dictionary that makes it like no other is that it aims to be comparative.
3

Saying that X word occurs Y times in Shakespeares plays and that it has
W and Z senses is less informative than contrasting those facts with those
of his contemporaries (and not just writers of literary texts but writers of
various text-types, including records of spoken interaction). In this way, we
can reveal not just the denotative or conceptual meanings of words but
A New Kind of Dictionary for Shakespeares Plays 59
also their stylistic, discoursal and pragmatic values in the general language
of the period. Similarly, the plan for the dictionary is that it should also
conduct internal comparisons, taking account of the distribution of items
over internal genres (e.g. comedy, tragedy, history, particular characters,
particular plays) and social categories (e.g. gender, role). Of course, what
is revealed through these internal comparisons can be further pursued
through external comparisons. For example, having identifed that X is
typical of women in Shakespeare, one could examine whether X is typical
of women in plays by other contemporary playwrights, in real life trial
proceedings, and so on.
In this chapter, I will deploy a number of case studies to show how tech-
niques developed in corpus linguistics can be used to produce this new
kind of dictionary based on usage and frequency. The case studies below
are chosen to illustrate particular issues relating to the dictionary; each case
study is not complete in itself. Moreover, my objective in elaborating these
case studies is not simply the design of a dictionary, but to show how light
can be thrown on some general issues in stylistics, such as characterization
and theme, as well as style in Shakespeare.
2. Labels and contents of current general
Shakespearean dictionaries
I refer to general Shakespearean dictionaries in order to exclude
dictionaries focusing on specific registers, such as legal, military or
informal language (see the Athlone Shakespeare Dictionary Series).
However, even with this exclusion, identifying what might count as a
general Shakespearean dictionary is far from easy. We find various labels
for books with contents characteristic at least to some degree of dic-
tionaries, notably, dictionary, glossary, lexicon and word-book. To
these one might wish to add concordances, in recognition of the fact
that such works contain a complete word list and (statistical) informa-
tion about those words aspects that might characterize a dictionary.
Moreover, what these works contain varies greatly. It is possible to identify
three groups. One is strongly linguistic in content, typically containing
information about the existence of a word-form, as well as its mean-
ing (conveyed with a brief definition) and illustrative quotation(s) and
part-of-speech (e.g. Crystal and Crystal 2002, Foster 1908, Onions 1986
[1911], Schmidt 1971 [1902]). Another group is strongly non-linguistic
in content, typically containing play summaries (largely plot focused),
character descriptions, cultural information and biographical informa-
tion (e.g. Boyce 1996, Wells 1998). Note that, although non-linguistic,
both of the examples cited are entitled Dictionary of Shakespeare.
The final group is strongly focused on (frequency of) occurrence informa-
tion, typically containing an index of all words (plus textual location)
and the frequency of word-forms (absolute and relative), (e.g. Spevack
19681980, Howard-Hill 19691972). There is a little slippage between
these groups for example, Schmidt (1971 [1902]) contains a com-
plete index of words and Crystal and Crystal (2002) was constructed
with frequency information in mind but in the main they are sepa-
rate. My proposal involves bringing together the three areas in a more
comprehensive and systematic fashion. This will clearly involve a broad
scope. Consequently, the label Dictionary of Shakespeare may not be
the best. A better alternative might be Encyclopaedia of Shakespeares
Language.
4
3. General Shakespearean dictionaries and present-day
English language dictionaries compared
The majority of present-day dictionaries of English contain pronunciation
information (typically, a broad phonetic transcription with an indication
of syllable stress). Doing the same for Shakespeare would require a signif-
cant research programme, and there would be thorny issues, such as whose
accent to represent. Consequently, this is not currently part of the dictio-
nary proposal. Many present-day dictionaries contain spelling variants, and
the OED, of course, excels in this respect. Shakespearean dictionaries do
not note more than the occasional spelling variant. Perhaps spelling vari-
ants are assumed not to be part of the real Shakespeare, given that they are
produced by compositors and printers. Nevertheless, spellings are the prism
through which we receive Shakespeare, and Shakespearean texts repre-
sent a source of information about spelling in the Early Modern period.
Moreover, quantifying spelling variation would be relatively easy to do with
the computational methodology supporting the proposed dictionary (see
section 12). Other differences in content include the fact that corpus-based
dictionaries of present-day English, notably, the Collins Cobuild Dictionary,
include defnitions that are more contextualized and information about
multiword units, as I will illustrate in sections 4 and 9.
Perhaps even more signifcant than differences in the kinds of infor-
mation that might be included are differences in policies for including
or excluding words and for prioritizing meanings. Shakespearean dictio-
naries, notably, Foster (1908) and Onions (1986 [1911]), as well as the
even more recent corpus-informed dictionaries such as Crystal and Crystal
(2002), tend to include only those words considered diffcult or hard for
readers. In contrast, corpus-based dictionaries typically include all the words
in the corpus (though that may not, in fact, be the best thing to do for a
Shakespearean dictionary; see section 5). Furthermore, present-day dictio-
naries, particularly corpus-based dictionaries, take a different approach to
the way meanings are prioritized within particular entries. Dictionaries in
the philological tradition exemplifed by the OED (e.g. Foster 1908 and
Onions 1986 [1911]) take etymology as a guiding principle. This is most
obviously refected in the way that (1) word defnitions gravitate towards
etymological meanings, and (2) the organization of the senses of poly-
semous words is based on etymological priority (i.e. the earliest sense is
listed frst). In contrast, corpus-based dictionaries capture meanings based
on usage in context, and organize those meanings according to frequency
(usually the most frequent is placed frst).
4. Towards a contextualized defnition: the case of horrid
The OED gives three senses for the word horrid: (1) bristling, shaggy,
rough, (2) causing horror or aversion; revolting to sight, hearing or con-
templation; terrible, dreadful, frightful; abominable, detestable and (3)
colloq. in weakened sense. Offensive, disagreeable, detested; very bad or
objectionable. Noted in NED as especially frequent as a feminine term of
strong aversion (here, and in all quotations from dictionaries in this chap-
ter, accompanying quotations are generally excluded for brevity). The
frst sense corresponds with that of the Latin term horridus from which
the English word is derived, and, judging from the illustrative quotations,
was still current in Shakespeares period. The second sense, and one that
is contemporary with Shakespeare, is a metonymic development of the
frst, and the fnal sense is apparently a weakened development of the
second. The fact that the frst quotation given to illustrate the second
sense is from Shakespeare should alert us to a major problem in using
the OED to interpret Shakespeare the problem of circularity, given that
Shakespeare plays such a large role in determining the entries in the OED
for the period in question. The third sense developed after Shakespeare.
Note that the OED does at least supply a modicum of stylistic information,
noting that the third sense is colloquial, and very occasionally some social
information, here noting that the third sense is especially frequent as a
feminine term.
5
Turning to three Shakespearean dictionaries, we fnd the following
defnitions:
Foster (1908): (1) Awful, hideous, horrible. (2) Terrifc.
6
(3) Horrifed,
affrighted.
Onions (1986 [1911]): No entry.
Crystal and Crystal (2002): Horrifying, frightful, terrifying.
Fosters (1908) frst defnition seems to shade into the third sense given
in the OED. This is odd because the frst citation date for that sense given
in the OED is 1666. The single illustrative quotation given by Foster is from
Macbeth: If good, why do I yield to that suggestion / Whose horrid image
doth unfx my hair. This quotation includes a classic reaction to fear the
unfxing of hair. This does not necessarily support the sense given by Foster,
which need not involve fear, just as in the third sense given in the OED. In
fact, the usage here falls within the scope of the OEDs second sense, as
indeed do Fosters second and third defnitions.
Note that the strongly overlapping array of synonyms given in the def-
nitions do little to pin down the sense of horrid in Shakespeare. What
is being described as horrid? Who is using this word? In what circum-
stances are they using it? Is Shakespeare using it in a way that his contem-
poraries would not? And so on. We can look at a computer concordance
(a list of the occurrences of the word along with their local co-text) and
the distribution of a word, in order to answer such questions. Here is
the entire concordance of horrid (the head noun to which it refers is
underlined):
7
Appear in forms more horrid, yet my duty, As doth a Rock
Up Sword; and know thou a more horrid hent
8
: When he is drunk asleep
And cleave the general ear with horrid speech, Make mad the guilty
heard and seen, Recounts most horrid sights seen by the watch
shall break his wind With fear and horrid fight.
I will meditate the while upon some horrid message for a Challenge.
Not in the legions of horrid hell, can come a devil more damned
Proper deformity seems not in the fend So horrid as in woman.
And what a beard of the generals cut and a horrid suit of the camp
Presented then unto the gazing moon So many horrid ghosts.
Crammed with distressful bread; Never sees horrid night, the child of hell
all the sparks of nature, To quit this horrid act. Out treacherous
Such sheets of fre, such bursts of horrid thunder, Such groans of
couriers of the air, Shall blow the horrid deed in every eye,
Of thy dear husband, than that horrid act Of the divorce held make
I yield to that suggestion, Whose horrid image doth unfx my hair
It can be seen that horrid is used to describe acts, sights and sounds, but not
just any such things most have a strong supernatural or unnatural connection.
This seems to have been overlooked in all dictionary defnitions, despite the
fact that it is quite obvious in the concordance. We can deepen our under-
standing of the word by considering its distribution both within Shakespeare
and without. Putting the results together, a dictionary entry might be as fol-
lows (All = all Shakespeares plays, T = tragedies, C = comedies, H = histories,
M = male speakers, F = female speakers, Pla = other EModE plays, Fic =
EModE prose fction, Tr = EModE trial proceedings, Ha = EModE hand-
books in dialogue form, Sc = EModE scholarly works; the fgures in brackets
are normalized per 100,000 words):
HORRID. Something that is horrid causes fear; typically, it refers to
supernatural or unnatural acts, sights and sounds. Distribution: All = 16
(1.8); T = 10 (3.9), C = 2 (0.6), H = 4 (1.5); M = 14 (1.9), F = 2 (1.4).
Comparisons: Pla = 187 (0.17), Fic = 0, Tr = 0, Ha = 0, Sc = 1 (0.14). E.g.
Whose horrid Image doth vnfxe my Heire, I wil meditate the while
vpon some horrid message for a Challenge.
9
The above is no more than an indication as to the direction a dictio-
nary entry might take. Note that the frst sentence offers a contextualized
defnition of the type used in the Collins Cobuild, rather than a handful of
synonyms. However, going beyond the Collins Cobuild, the fgures follow-
ing offer a broader discoursal contextualization. They give some indication
as to the social and stylistic meanings the word might have acquired on
account of being to some degree contextually bound (Leech 1981: 145;
see also Enkvist 1964: 2935). Focusing on the more meaningful normal-
ized fgures, one might note, for example, that the word horrid, which
appears much more densely in tragedies than either histories or comedies,
is used slightly more frequently by male characters compared with female;
and that Shakespeare uses it considerably more than his contemporary
playwrights did, and also that it is most characteristic of Early Modern plays
and, surprisingly, scholarly literature.
However, this particular example is severely hampered by frequency limi-
tations: the strongest fnding revealed by the fgures simply being that hor-
rid is rare. I will focus on frequency limitations in the next section. Here, I
will briefy indicate four ways in which the above entry could be improved:
The defnition was derived from collocational information and some of
this information could have been included in the entry (see section 6).
Sociolinguistic information could be enriched by the inclusion of other
sociological variables (e.g. status, age) and also comparative data (e.g.
addressing questions such as: is X word associated with male or high
status speakers in Shakespeare specifcally or is this a more general fea-
ture of Early Modern English?).
A statistical measure could be employed in order to indicate whether
differences in distribution are signifcant.
The presentation of information could be improved (e.g. the use of
graphs, or a verbal description instead of fgures).
5. Frequency limitations
A corpus-based dictionary typically includes all words in the corpus.
However, this presents two problems: (1) how to treat rare or infrequent
words, and, from the more practical point of view of publication, (2) how to
ft all the words into one volume.
10
As is clear from the sample entry of hor-
rid above, low frequency words lead one to the mere conclusion that they
are low frequency, as the more robust and informative distribution patterns
fail to materialize.
11
The corpus-based methodology is not best suited to
investigating low frequency words (cf. Biber et al. 1998: 30; Meyer 2002: 15);
instead, we need to look towards alternative methodologies, such as the
philological approach that already underpins most current Shakespearean
dictionaries. A partial solution to these problems is simply to adopt a fre-
quency cut-off point such that words below a certain frequency are not con-
sidered for inclusion in the dictionary. But what would be the implications
of such a cut-off point for the coverage of Shakespeares vocabulary?
Onions (1986 [1911]) supposedly covers some 3,000 words, according to
Crystal and Crystal (2002: Introduction), who also claim to include 21,263
entries under 13,626 headwords in their own volume. Table 3.1 displays the
consequences of various cut-off points. For example, the frst row consid-
ers words (word-types) which are repeated at least 100 times (they have at
least 100 tokens). As can be seen from the following columns, 998 words
( word-types) are repeated at least 100 times (have more than 100 tokens),
and, if these 998 word-types and all their repeats are added up, they account
for 706,974 of the words that make up Shakespeares plays.
As the bottom row shows, there is a total of 899,092 word-tokens in
Shakespeare and 24,842 different word-types (in other words, a smaller
number of different words are repeated a number of times to make up
the total vocabulary of Shakespeares plays). Horrid occurs 16 times. If
we only consider for the proposed dictionary word-types that occur more
than 16 times, then, potentially, we would only need to have 4,652 different
entries in our dictionary, and yet we would still cover most of the language
of Shakespeares plays: 835,925 word-tokens. However, I pointed out above
that 16 occurrences are too few for our purposes. If a cut-off point of, say,
50 word-tokens for any word entry were imposed, resulting in a potential
and certainly more manageable 1,564 word-form entries, then that still
would account for the vast bulk of the words in Shakespeare (761,472 out of
899,092 word-forms). Note that 7,753 word-types occur less than 10 times,
accounting for a mere 37,260 word-tokens. Yet it is precisely here that the
current Shakespearean dictionaries tend to focus, as these rare items tend
to be considered hard. In my view, there is no justifcation for exclud-
ing more frequently occurring vocabulary items. From a linguistic perspec-
tive, we know that all words change meaning: even the most frequent of
items have incurred shifts of meaning which present-day readers must take
on board. From a literary perspective, we should beware of letting more
unusual vocabulary distract our attention from the more usual. As John F.
Burrows (1987: 1) eloquently puts it: It is a truth not generally acknowl-
edged that, in most discussions of works of English fction, we proceed as if
a third, two-ffths, a half of our material were not really there.
Table 3.1 Word-types and tokens in Shakespeares plays
How many word-types (different
words) are accounted for
How many word-tokens (instances
of words) are accounted for
Word-types with more
than 100 tokens
998 706,974
than 50 tokens
1,564 761,472
than 16 tokens
4,652 835,925
Word-types with less
than 10 tokens
7,753 37,260
All instances of
all word-types
24,842 899,092
6. Polysemy and collocates: the case of good
Current Shakespearean dictionaries give defnitions for the word good (as
an adjective) such as these (illustrative quotations are excluded):
Foster (1908): (1) Not bad, worthy of praise; (2) Fit, adapted; (3)
Trustworthy, genuine; (4) Kind, benevolent; (5) Proper, right; (6)
Substantial, safe, solvent, able to fulfl engagements; (7) Real, serious;
(8) Favourable, propitious; (9) Abundant, rich; (10) Skilful, clever; (11)
Adequate. (Notes phrases and compounds).
Onions (1986 [1911]): (1) Conventional epithet to titles of high rank; (2)
Comely; (3) Financially sound, (hence) wealthy, substantial. (Notes quasi-
adverbial usage, e.g. good easy man, and phrases and compounds).
Crystal and Crystal (2002): (1) [intensifying use] real, genuine (love no
man in good earnest); (2) kind, benevolent, generous; (3) kind, friendly,
sympathetic; (4) amenable, tractable, manageable; (5) honest, virtuous,
honourable; (6) seasonable, appropriate, proper; (7) just, right, com-
mendable; (8) intended, right, proper; (9) high-ranking, highborn, dis-
tinguished; (10) rich, wealthy, substantial. (Notes phrases and compounds).
Lists of synonyms in some cases overlapping do not always provide the
reader with assistance in discriminating the various senses. For example,
in Fosters (1908) defnitions, how does genuine in sense 3 differ from
real in sense 7? Similarly, ft in sense 2 can uncomfortably overlap with
proper, right of sense 5. Onionss (1986 [1911]) defnitions are fairly dis-
crete, while in contrast Crystal and Crystal (2002) seem to have gone for a
deliberate policy of overlap (note that kind, proper and right appear in
more than one defnition), perhaps indicating that indeed senses do over-
lap. We might also note that each dictionary orders the senses in a different
way, and that some, rather worryingly, contain senses that others do not
(note, for example, Onionss frst sense).
A simple technique in corpus linguistics for investigating the meaning of a
word is to examine a concordance and note the words with which the word
in question co-occurs, something which we have already demonstrated with
the word horrid. It is the collocates of a word the company it keeps (cf.
J. R. Firth 1957) that may help distinguish different senses (see, for example,
Partington 1998: 3346). Frequent collocating words to the right of good
include: (my) good friend(s)/sir/Lord/master/man/Lady/Madam/etc.,
good old man/friend/etc., good morrow/night/even, (in) good faith,
good will/wish(es), good god(s), good luck / hap, good news/report/
words, good now, and (in) good time. Even without further elaboration,
seeing such collocations helps make accessible distinct senses, and so they
should be included within a dictionary entry. Also, the frequency of such
collocations can feed into the ordering of senses within the entry. However,
with a dizzying 2,711 instances constituting a concordance of good, the
human can only identify some collocational patterns, and cannot accurately
assess the strength of those patterns and thus come to a principled decision
about which to include in the dictionary. One possible solution is to calculate
the statistical likelihood with which particular words and good co-occur to
form a collocation. Using z-scores, a statistical measure, the top ten ranked-
ordered collocates fve words to the left and right of good are: morrow, Lord,
my, do, sir, good, your, have, be and you, as can be seen in Table 3.2.
12
These collocational patterns point to sentences like the following (con-
structed) example: Good morrow, my good lord, you have . . . This evi-
dence clearly underpins Onionss (1986 [1911]) frst sense, a sense that is
not clearly represented in the other dictionaries, and underscores the role
of good as a politeness marker. Such an investigation could be extended in
three ways: (1) collocational patterns (and ones not limited to single word
collocates) can be identifed with other statistical procedures (including
the methodology in section 9), (2) collocational patterns in Shakespeare
can be compared with collocational patterns in other Early Modern texts
(e.g. is Shakespeare peculiar in his usage of good as a politeness marker?)
and (3) grammatical relations can be explored via collocations (e.g. as is
transparent for the concordance of horrid the items immediately to the
right are nearly always nouns something that confrms the status of hor-
rid as a typical adjective in English).
Table 3.2 The top ten ranked-ordered collocates of
good within a fve-word span
Collocate (5/5) Frequency Z-score
Morrow 6 18.0
Lord 11 4.8
My 21 2.8
Do 8 2.5
Sir 6 2.1
Good 6 2.1
Your 11 2.0
Have 9 1.5
Be 9 1.1
You 15 0.7
7. The inclusion of pragmatic/discoursal
words: the case of ah
Interjections, onomatopoeic sounds, hesitation phenomena, discourse
markers and so on have received scant attention in Shakespearean dictio-
naries (of course, this is not true of specialist dictionaries, notably Blake
2004). For example, there is no entry for ah in Foster (1908), Onions
(1986 [1911]) or Crystal and Crystal (2002). The issue is whether such
items are considered words, and that depends on your defnition of a word.
Corpus linguistics favours an orthographic defnition, such as a string of
uninterrupted non-punctuation characters with white space or punctuation
at each end (Leech et al. 2001: 134). In which case, ah is clearly a word.
Does ah have meaning? That depends on your defnition of meaning. If
meaning is associated with ideational meaning, to use Hallidays (e.g. 1978)
terminology, and not textual or interpersonal meanings, then words like
ah do not have meaning. One of the reasons such words are not gener-
ally included in Shakespearean dictionaries is that words that refect some
aspect of the world are privileged above words that help organize other
words or words that help organize people. In my view, this approach is
entirely inappropriate for a dictionary of Shakespeares plays because those
plays are made up of dialogue. What lies at the heart of dialogue are those
pragmatic and discoursal words that structure and mediate the interaction
between characters.
Let us consider the pragmatic and discoursal meanings of ah, and also
its social and stylistic meanings. If a concordance of ah is scrutinized, one
can discern the three key pragmatic meanings following (an illustrative
example is provided of each):
Speaker attitude/state communicated = sorrow, emotional distress
desdemona To whom, my lord? with whom? how am I false?
othello Ah Desdemon! away! away! away!
desdemona Alas the heavy day! Why do you weep?
Am I the motive of these tears, my lord?
(Othello 4.2.403)
Speaker attitude/state communicated = pity
gloucester Canst thou blame him? . . .
His daughters seek his death; ah, that good Kent!
He said it would be thus, poor banished man!
Thou sayst the king grows mad; Ill tell thee, friend.
I am almost mad my self: . . .
(King Lear 3.4.1615)
Speaker attitude/state communicated = surprise, realization
[Enter Adriana and Luciana.]
adriana Ah, Luciana, did he tempt thee so?
(The Comedy of Errors 4.2.1)
And one can discern the two key discoursal meanings following:
Discourse marker: preface to the correction/rejection of the previous speakers
proposition(s), emotions or actions
menas These three world-sharers, these competitors,
Are in thy vessel: let me cut the cable;
And, when we are put off, fall to their throats:
All there is thine.
pompeius Ah, this thou shouldst have done,
And not have spoke ont! In me tis villainy;
In theet had been good service. . . .
(Antony and Cleopatra 2.7.705)
Discourse marker: reinforces elicitation
leonato All thy tediousness on me, ah?
dogberry Yea, an twere a thousand pound more than tis; for I hear
as good exclamation on your worship as of any man in the
city; and though I be but a poor man, I am glad to hear it.
Turning to stylistic and social meanings, consider the distribution of ah:
Distribution: All = 179 (19.9); T = 54 (21.3), C = 32 (8.9),
H = 93 (35.4); M = 121 (16.1), F = 59 (41.9).
Comparisons: Pla = 1,573 (14.4), Fic = 9 (10.9), Tr = 1 (2.9),
Ha = 11 (11.2), Sc = 0.
Within Shakespeare, ah is characteristic of the histories, to some extent
the tragedies, but to a much lesser extent the comedies. This distribution
may refect the frequent functions of ah in signalling emotional distress and
pity. Interestingly, the distribution across genders is far from even: it is more
than twice as dense in female dialogue. Compared with other playwrights
of the period, Shakespeare can be said to be fairly fond of this item. Also,
we can see that it is more characteristic of plays than other contemporary
genres. Moreover, there is evidence that ah is a strong colloquial marker. It
does not appear at all in scholarly works, the genre that is far removed from
colloquial genres; it hardly appears in trial proceedings, a genre that infu-
enced by the formal setting, legal routines and need to create an offcial doc-
ument tends to be remote from colloquial language; while on the other
hand, it appears in fctional prose (the choice of prose for this data set being
specifcally geared towards more colloquial prose) and handbooks in dia-
logue form. Interestingly, and remarkably, the density of ah in a sample of
fve present-day plays is 94.27 (contrasting with Shakespeares 19.9), some-
thing which presumably refects the drift of genres, including plays, towards
more colloquial language (see, for example, Biber and Finegan 1992).
8. The inclusion of grammatical words: the case of and
The most frequent words in any body of texts are closed-class. Yet
Shakespearean dictionaries do not, or do not adequately, treat such grammati-
cal items, despite or may be because of their high frequency of occur rence.
For example, the entries for the second most frequent word in Shakespeare,
the word and, in general Shakespearean dictionaries are as follows:
Foster (1908): Cross-references Abbots Shakespearean grammar.
Onions (1986 [1911]): (1) Coordinating conjunction (nouns, adjec-
tives and phrases); (2) Subordinating conjunction: if, even if, though,
as if, whether.
Crystal and Crystal (2002): [also spelling variant an] (1) if, even if;
(2) as if; (3) if, whether.
As can be seen, it is not treated at all in Foster (1908), while Crystal and
Crystal (2002) only mention conditional and (used as a subordinate con-
junction introducing a conditional clause with the sense if). Conditional
and is likely to be the focus of attention in Shakespearean dictionaries,
because of editorial policies to select items with which the modern reader is
assumed to be unfamiliar and thus likely to experience diffculty. Examples
of conditional and include the following:
bawd What would you have me be, an I be not a woman?
(Pericles 4.2.81)
romeo Noting this penury, to myself I said
An if a man did need a poison now,
Whose sale is present death in Mantua,
Here lives a caitiff wretch would sell it him.
(Romeo and Juliet 5.1.4952)
Only Onions (1986 [1911]) acknowledges the fact that words such as and
play an important grammatical role. It is the coordinating function of and
that accounts for the overwhelming majority of instances in Shakespeare.
And makes a signifcant contribution to textual meaning in Shakespeare
in the way it conjoins nouns, adjectives, (nominal or adjectival) phrases and
clauses, and it is also used as a pragmatic connective.
Compare the following two extracts in which instances coordinating
clauses are underlined and instances coordinating words/phrases are
emboldened:
vincentio She should this Angelo have married; was affanced to her
by oath, and the nuptial appointed: between which time
of the contract and limit of the solemnity, her brother
Frederick was wracked at sea, having in that perished
vessel the dowry of his sister. But mark how heavily this
befell to the poor gentlewoman: there she lost a noble
and renowned brother, in his love toward her ever most
kind and natural; with him, the portion and sinew of her
fortune, her marriage-dowry; with both, her combinate-
husband, this well-seeming Angelo.
(Measure for Measure 3.1.21323)
second gentleman Whos that that bears the sceptre?
first gentleman Marquess Dorset:
And that the Earl of Surrey, with the rod.
second gentleman A bold brave gentleman. That should be
The Duke of Suffolk?
first gentleman Tis the same: high steward.
second gentleman And that my Lord of Norfolk?
first gentleman Yes.
(King Henry VIII 4.1.3744)
The density of lexical/phrasal coordination in the frst extract contrasts
with the density of clausal coordination in the second. This grammatical
difference refects differences in style and communicative purpose. Lexical/
phrasal coordination in Vincentios speech helps create a high rhetorical
style, underscoring the seriousness of what he is saying. More specifcally,
the conjoins of coordinated pairs tend to be closely related in meaning.
Thus, noble and renowned overlap in meaning (refecting the rhetori-
cal fgure of pleonasm), and kind and natural could be viewed as being
in a hierarchical relationship such that one is subordinate to the other
(refecting the rhetorical fgure of hendiadys, that is amounting to nat-
urally kind). In contrast, the clausal coordination of the second extract
creates a low rhetorical style, underscoring the casual conversation, a style
which is, of course, reinforced by the ellipsis. In fact, in this particular case,
and is not merely coordinating clauses but also acting as a pragmatic con-
nective. Specifcally, the second instance is used to create a series of ques-
tions, or, as Schiffrin puts it, to link questions in a question agenda (1994
[1987]: 146). As a consequence of their rather different functions, lexical/
phrasal coordination tends to correlate with rather different genres com-
pared with clausal coordination. I cannot prove this claim with regard to
Shakespeare, as the computational analysis of Shakespeares grammar is
not yet suffciently accurate or sophisticated; indeed, one of the aims of my
dictionary project is to solve this (however, see Culpeper and Kyt 2002,
which provides supporting evidence for this correlation with regard to
four Early Modern genres). In sum, my argument is that such grammatical
items should be included in a dictionary of Shakespeare, and that dictio-
nary should focus widely on the contribution of those items to meaning.
9. Multiword units
John Sinclair (e.g. 1991), among other linguists, has argued that words
may belong to semi-fxed phrases that constitute single lexical choices (e.g.
of course, where the individual words cannot be assumed to produce
the sense of the phrase). Current Shakespearean dictionaries pay scant
attention to these. An empirical way of retrieving lexical items that tend
to bunch together is to run an n-gram analysis. Essentially, the computer
works through the text, recording the co-occurrence of every word with its
neighbours, and then calculates which groups of words most frequently co-
occur. Multiword units, thus defned, may be considered a kind of extended
collocational unit and are frequently referred to as lexical bundles or clus-
ters. The results for Shakespeare, retrieved by WordSmith Tools (Scott 1999),
are included in Table 3.3 along with the results for three other data sets for
comparison (the underlining, italics and emboldening show that a partic-
ular lexical bundle is used in another data set; no lexical bundle is used in
more than two data sets).
It has been noted in the literature that lexical bundles are good discrimi-
nators of different styles (e.g. Stubbs and Barth 2003). The bulk of the items
in Table 3.3 are unique to the specifc data sets. Lexical bundles in Early
Modern trials refect the fact that discourse is made up of question-answer
routines (e.g. do you know, did you see vs. I did not, I do not) and crime-
narrative report (e.g. he told me, at that time, out of the, I told him).
Lexical bundles in present-day plays seem to gravitate towards questions and
assertions to do with knowing, wanting and thinking perhaps refecting
the essence of present-day drama in which plot and character development
is conveyed through highly interactive character-to-character dialogue (in
other words, what is said between characters is partly designed to inform the
audience of character and plot). A characteristic of both Shakespeare and
other Early Modern plays is that many of the bundles begin with the frst
person pronoun I, perhaps refecting the essence of Early Modern drama
with its more direct presentation of characters and plot to the audience
(the epitome of this being the use of soliloquies and asides). Shakespeares
lexical bundles are distinguished by the fact that his top fve most frequent
bundles begin with the frst person pronoun. Also, it is interesting to note
that the most frequent three-word unit in Shakespeares plays, I pray you,
is something that is not characteristic of other Early Modern plays, other
genres or, of course, of present-day plays.
The kind of distributional stylistic information I have been discussing
here could, of course, be recorded along with the entries for the most fre-
quent lexical bundles in Shakespeare in the dictionary. Perhaps even more
Table 3.3 The top ranked-ordered three-word lexical bundles in Shakespeare
and other texts
Shakespeare EModE plays EModE trials Present-day plays
I pray you
I will not
I know not
I am a
I am not
my good lord
there is no
I would not
it is a
and I will
it is a
what do you
and I will
it is not
I have a
I will not
in the world
I tell you
I know not
I warrant you
do you know
I did not
did you see
I do not
he told me
at that time
out of the
I told him
he did not
there was a
I dont know
what do you
I dont want
do you think
do you want
I dont think
to do with
do you know
going to be
dont want to
importantly, such n-gram analysis can feed into the grammatical descrip-
tion contained in the dictionary. I will attend to this issue in the following
section.
10. A note on grammatical description
Linguists like Sinclair (e.g. 1991, 2004) emphasize that grammar is in the
lexicon and not in some a priori set of abstract categories (e.g. parts of
speech) imposed on the language. A way into describing the lexico-gram-
mar of Shakespeare would be to describe the grammatical frames or pat-
terns, revealed through collocational analyses, as discussed in section 6,
and multiword analyses, as discussed in section 9 (see Hunston and Francis
2000, for this approach). I have already hinted that collocational analyses
could be deployed in the exploration of grammatical relations, noting the
case of horrid (and good is similar).
13
In fact, my discussion of and was
very much geared towards the grammatical relations of co-occurring units.
Regarding multiword units, I pray you, for example, is a grammatical
pattern partly consisting of a frst person pronoun (i.e. either we or I),
a verb in the present tense and a second person pronoun (i.e. either
you or thee). While the items that can occur as pronouns are relatively
restricted, a much wider range of verbs can occur in the middle of this
particular pattern. However, not any verb is used: the set is also restricted.
One subset of those verbs is comprised of speech act verbs such as advise,
arrest, assure, beseech, charge, tell, thank and warrant. Such verbs
occur when the grammatical pattern is used in isolation or parenthetically
to a matrix clause. Making the step from an n-gram analysis to the descrip-
tion of grammatical patterns or frames is not necessarily straightforward.
N-gram analysis results in units which are not necessarily complete idioms
or grammatical structures. Nevertheless, such analysis offers a way into it
identifying grammatical frames, and the results can be complemented by
collocational analyses.
I would not argue for quite as radical an approach to grammar (i.e. ditch
all abstract grammatical categories) as Sinclair, for four reasons. First, my
analysis of and already demonstrated that grammatical categories can be
useful. Knowing the grammatical status of the conjoins (i.e. lexical/phrasal
versus clausal) helps us account for textual meanings. Second, grammatical
categories can provide a useful way of tracking variation and change in the
language; specifcally, in the case of the dictionary, it can help provide a
way of understanding how language varied in Shakespeares time (e.g. from
register to register, from person-to-person) and how language has changed
since Shakespeare. For example, the proposed dictionary could quantify
parts of speech, particularly in cases where an item can function as more
than one part of the speech, and thereby reveal differences in distribution
(e.g. the distribution of verbal versus nominal usages of the lexeme love
used to be weighted in favour of nominal but is now approximately even).
Third, supplying such information about words would enable researchers
to compare and contrast with extant research. Fourth, supplying such infor-
mation can simply be one additional means by which a dictionary can help
users understand words.
11. Character and play profles
Some Shakespearean dictionaries contain non-linguistic descriptions of
characters and plot summaries. I propose providing a description of the
idiolect of each major character. This can be done by conducting a sta-
tistical comparison between the vocabulary of one character and that
of the other characters in the same play, in order to reveal words that
are statistically characteristic of particular characters. Those words are
keywords. As an illustration, consider some of the results relating to
characters in Romeo and Juliet (see Culpeper 2002, for a more detailed
discussion). Table 3.4 contains the keywords of Romeo and Juliet (rank-
ordered in terms of the statistical keyness) produced by the program
WordSmith Tools.
This reveals, for example, the predictable result that Romeos two most
unusually frequent words (or keywords) are beauty and love, as well
as the less predictable and thus possibly more interesting result that
Juliets two most unusually frequent words are if and be. Although the
results for Juliet are less predictable, they can readily be explained by a
qualitative analysis of the text (i.e. they are motivated). Furthermore, and
Table 3.4 Rank-ordered keywords for Romeo and Juliet (raw frequencies in brackets)
romeo juliet
Beauty (10), Love (46), Blessed (5),
Eyes (14), More (26), Mine (14),
Dear (13), Rich (7), Me (73), Yonder (5),
Farewell (11), Sick (6), Lips (9), Stars (5),
Fair (15), Hand (11), Thine (7),
Banished (9), Goose (5), That (84)
If (31), Be (59), Or (25), I (138), Sweet (16),
My (92), News (9), Thou (71), Night (27),
Would (20), Yet (18), That (82),
Nurse (20), Name (11), Words (5),
Tybalts (6), Send (7), Husband (7),
Swear (5), Where (16), Again (10)
following the line of argument articulated above, although many of Juliets
keywords are grammatical in nature, they are no less meaningful. Upon
closer inspection of Juliets keywords, one can see that keywords such as if,
be (often subjunctive), or, would and yet refect Juliets anxieties and
worries about Romeos intentions and welfare, as the following examples
illustrate:
. . . if he be married, / My grave is like to be my wedding-bed
(1.5.1323)
If they do see thee, they will murder thee. (2.2.70)
. . . But if thou meanst not well, (2.2.150)
Is thy news good, or bad? answer to that;
Say either, and Ill stay the circumstance:
Let me be satisfed, ist good or bad? (2.5.357)
Tis almost morning; I would have thee gone:
And yet no further than a wantons bird, (2.2.1767)
The key point about such analysis is that, although a reading of the play
would obviously have resulted in an understanding of Juliets anxieties and
worries, such a reading would not necessarily have led to the identifcation
of the linguistic source of that very understanding. Indeed, no manual
critical analysis to date, literary or linguistic, has fully accounted for the
source.
Regarding plays, plot summaries tend to include information about the
plays themes. Such information relies on the intuitions of the editor. I
propose something more empirical: providing a description of the seman-
tic categories (or lexical felds) characterizing each play. This can be done
by getting the computer automatically to assign each word in the plays
to a semantic category (this assignment can, of course, be recorded in
the dictionary entry for each word). The dominance of categories within
plays can be statistically compared. For example, in an earlier study I con-
ducted with Dawn Archer and Paul Rayson (Archer et al. 2009), we com-
pared three love tragedies (Othello, Antony and Cleopatra and Romeo and
Juliet) with three love comedies (A Midsummer Nights Dream, The Two
Gentlemen of Verona and As You Like It). Each word was assigned to the cate-
gories in Table 3.5 using the USAS suite of programs (for further details,
see section 12).
Then a statistical comparison was conducted in order to establish which
semantic categories were characteristic of each data set (each semantic cat-
egory has several subcategories). Our fndings are displayed in Table 3.6.
It is love comedies that are characterized by the most obviously love-re-
lated category, intimate/sexual relationship. The love tragedies, by con-
trast, are characterized by categories far removed from love: warfare, etc.,
lack of life, etc., and so on. Closer inspection of the results in the context
of the plays reveals many points of interest. For reasons of space, I will just
comment on a few. The appearance of plants as highly characteristic of
comedies may seem puzzling. In fact, there is a connection with love, as the
following extract illustrates (Silvius explains why he loves Phoebe despite
Table 3.5 The semantic categories used (derived from McArthur 1981)
A
general and
abstract terms
B
the body and
the individual
C
arts and crafts
E
emotion
F
food and farming
G
government
and public
H
architecture, housing
and the home
I
money and commerce
in industry
K
entertainment,
sports and games
L
life and
living things
M
movement, location,
travel and transport
N
numbers and
measurement
O
substances,
materials, objects
and equipment
P
education
Q
language and
communication
S
social actions,
states and processes
T
time
Q
world and
environment
X
psychological actions,
states and processes
Y
science and technology
Z
names and grammars

Table 3.6 Love comedies and tragedies: characteristic semantic categories
(rank-ordered)
Most overused categories in
comedies relative to tragedies
Most overused categories in
tragedies relative to comedies
S3.2 = intimate/sexual relationship
L2 = living creatures
L3 = plants
S1.2.6- = (not) sensible
X3.1 = sensory: taste
E2+ = liking
T3- = old, new, young: age
G3 = warfare, defence, and the army
L1- = (lack of) life/living things
Z2 = geographical names
E3- = (not) calm/violent/angry
M4 = movement (by sea/through water)
S9 = religion and the supernatural
S7.1- = (lack of) power/organizing
the fact that she is a prostitute) (words assigned to the plants semantic
category are emboldened):
silvius So holy and so perfect is my love,
And I in such a poverty of grace,
That I shall think it a most plenteous crop
To glean the broken ears after the man
That the main harvest reaps: loose now and then
A scattered smile, and that Ill live upon.
(As You Like It 3.5.98103)
More precisely, the connection is a metaphorical one. As Oncins-Martnez
(2006) has pointed out, the underlying cognitive metaphor here is sex is
agriculture and its sub-mappings include a womans body is agricul-
tural land. Similarly, metaphor accounts for the presence of the semantic
category sensory: taste, as illustrated in the following example:
julia Nay, would I were so angered with the same!
O hateful hands, to tear such loving words!
Injurious wasps, to feed on such sweet honey
And kill the bees that yield it with your stings!
(The Two Gentlemen of Verona 1.2.1047)
The underlying cognitive metaphor here is love is food (see Barcelona
1995: 6723; see also Oncins-Martnez 2006).
12. Conclusions
The main features of my proposed dictionary can be summarized as
follows:
All words will be treated equally (e.g. not just hard words or content
words).
Meanings will not be restricted to semantic or ideational meaning.
Meanings will be based on usage in context (e.g. not etymology).
Context will include linguistic co-text (e.g. collocations) and non-
linguistic context (e.g. social properties of the speaker).
Linguistic description will be relative, that is, it will compare Shake-
speares usage with that of contemporary texts.
The dictionary will include linguistic profles of characters and plays.
Perhaps the most important question to raise at this stage is: to what extent
is this agenda feasible? In fact, the reason why I am proposing this kind of
dictionary now is that until recently it would have been impossible. With
developments in both corpora and computational techniques, we are now
at a point when it can be realized. Here, I will briefy list some methodologi-
cal problems and indicate the extent to which they have been solved:
There used to be a lack of comparative textual data in electronic form.
However, this has been partially solved by, for example, the Helsinki Cor-
pus of English Texts, the Corpus of English Dialogues 15601760, and so on.
Early Modern spelling variation has been perhaps the major stumbling
block for historical corpus linguistics, and hitherto the major stumbling
block for the proposed dictionary, for the reason that one cannot search
for a particular word-spelling and assume that all the relevant words will
be retrieved. However, this problem has largely been solved by the Vari-
ant Detector (VARD), primarily devised by Dawn Archer, University of
Central Lancashire and Paul Rayson, Lancaster University (see, for
example, Archer and Rayson 2004, Archer et al. 2003).
Studying abstract grammatical patterns in a corpus requires grammatical
annotation. The Lancaster-developed CLAWS part-of-Speech annotation
system works fairly well for present-day English (for descriptions of how
CLAWS works, see Leech et al. 1994 or Garside 1987). It has been recently
adapted at Lancaster for Early Modern English. However, it is not suff-
ciently accurate for the dictionary and manual correction is required
(once this is done, of course, a powerful resource will be created).
Semantic annotation has received attention from generations of research-
ers at Lancaster University, including Geoffrey Leech, Jenny Thomas,
Roger Garside, Andrew Wilson, Paul Rayson and Dawn Archer. The
USAS semantic annotation system has been adapted for Early Modern
English, and demonstrated to have value (see, for example, Archer et al.
2003). However, it is not suffciently accurate for the dictionary and
would require a further round of development. There is also the thorny
problem of what world view the system should adopt.
Social annotation, information about, for example, gender, status, age,
has not yet been comprehensively and systematically applied to Shake-
speare, but the methodology has been developed and applied to Early
Modern English texts (see Archer and Culpeper 2003), and so it would
be fairly straightforward to extend this to Shakespeare.
A fnal problematic area to note, and one that is philological and not
computational, is that the paper-based dictionary will need to be pri-
marily based on one particular edition of Shakespeare, and this will
involve an evaluation of available editions to arrive at a fnal choice.
Having said that, the dictionary could incorporate important informa-
tion arising from a comparison of different editions. Moreover, the
project would deliver not just a paper-based dictionary, but also a web-
site, which would allow researchers to pursue their own searches in
whatever editions that could be made available. In fact, such searches
are already possible via the Shakespeare Database Project (see: www.
shkspr.uni-muenster.de/index.php). However, this database only
includes Shakespearean texts (it does not have the wider comparative
capability of the dictionary project described here) and is not publicly
available.
Notes
1
This chapter is a revised version of: Culpeper, J. (2007) A New Kind of Dictionary
for Shakespeares Plays: an Immodest Proposal, SEDERI (Yearbook of the Spanish
and Portuguese Society for English Renaissance Studies), 17: 4773.
2
Onions was in fact one of the editorial team of the OED.
3
This kind of approach is akin to the corpus-based grammar produced by Biber et
al. (1999).
4
I am grateful to Antony Warner for this suggestion.
5
Although the evidence is thin, explorations in the present-day British National Cor-
pus suggest that women do tend to use the term horrid more than men.
6
In the period Foster was writing, this could have the earlier sense of causing
terror.
7
Of course, a concordance of a word will vary in terms of how many instances it
contains according to the edition of Shakespeare used (and occasionally accord-
ing to how good the search software is). The particular Shakespeare edition used
in this chapter is outlined in note 9.
8
Hent means clasp.
9
The Shakespeare frequencies given in this chapter are based on The Nameless
Shakespeare (2003), a joint project of the Perseus Project at Tufts University, The
Northwestern University Library, and Northwestern University Academic Tech-
nologies. It is derived from The Globe Shakespeare, the one-volume version of the
Cambridge Shakespeare, edited by William G. Clark, John Glover, and William A.
Wright (18911893). There is no claim here that this constitutes the ideal edition
of Shakespeare. It is searchable via WordHoard (the concordance in section 4
was derived by this). The comparative Pla corpus is the Korpus of Early Modern
Playtexts in English (KEMPE), initially compiled by Lene B. Petersen and Marcus
X. Dahl, University of Bristol, 20012003. It is searchable via Corpuseye. Note: a
particular problem with the Corpuseye search engine is that it only searches the
whole corpus and that corpus includes Shakespeare. Nevertheless, given the great

size of the corpus 10.7 million words the results will still mean something. The
samples for Early Modern prose fction, trial proceedings and handbooks are
sourced from the Corpus of English Dialogues 15601760, and the scholarly works
comprise half history writing and half science writing, sourced from the Helsinki
Corpus of English Texts.
10
Schmidts (1971 [1902]) complete treatment of Shakespeares lexicon stretches
over two volumes of small print and thin paper, yet only contains the briefest of
defnitions.
11
There are also diffculties in applying statistical signifcance tests to differences in
distribution that involve low frequencies.
12
It is a matter of debate as to which statistical measure to use. Mutual information
scores are frequently used, some use t-scores and some argue for the Fisher exact
test. These results were produced using the software Xiara.
13
As I have already indicated in this chapter, a highly accurate part-of-speech
tagged corpus of Shakespeare does not exist. Also, there are issues to do with
the compatibility of tags and software, as well as devising software to assess
adequately grammatical relations. One possibility to be explored is SketchEngine
(see Kilgarriff et al. 2004), used for lexicography by Oxford University Press,
for example.
References
Shakespeare frequencies and examples given in this paper are based on The
Nameless Shakespeare (2003); see note 9.
Archer, Dawn and Jonathan Culpeper 2003. Sociopragmatic annotation: new direc-
tions and possibilities in historical corpus linguistics, in: Andrew Wilson, Paul
Rayson and Anthony M. McEnery (eds), Corpus Linguistics by the Lune: A Festschrift
for Geoffrey Leech. Frankfurt/Main: Peter Lang, 3758.
Archer, Dawn, Tony McEnery, Paul Rayson and Andrew Hardie 2003. Developing an
automated semantic analysis system for Early Modern English, in: Dawn Archer,
Paul Rayson, Andrew Wilson and Tony McEnery (eds), Proceedings of the Corpus
Linguistics 2003 conference. UCREL technical paper number 16. UCREL, Lancaster
University, 2231. www.comp.lancs.ac.uk/computing/users/paul/public.html.
Archer, Dawn and Paul Rayson 2004. Using an historical semantic tagger as a diag-
nostic tool for variation in spelling. Presented at Thirteenth International Conference
on English Historical Linguistics (ICEHL 13) University of Vienna, Austria 2329
August, 2004. (Also available at: www.comp.lancs.ac.uk/computing/users/paul/
public.html).
Archer, Dawn and Jonathan Culpeper 2009. Love a familiar or a devil? An explo-
ration of key domains in Shakespeares comedies and tragedies, in: Dawn Archer
(ed.), Whats in a Word-List? Investigating Word Frequency and Keyword Extraction.
London: Ashgate, 13657.
Barcelona Snchez, Antonio 1995. Metaphorical models of romantic love in Romeo
and Juliet, Journal of Pragmatics 24: 66788.
Biber, Douglas and Edward Finegan 1992. The linguistic evolution of fve written
and speech-based English genres from the 17th to the 20th centuries, in: Matti
Rissanen, Ossi Ihalainen, Terttu Nevalainen and Irma Taavitsainen (eds), History
of Englishes: New Methods and Interpretations in Historical Linguistics. Berlin: Mouton
de Gruyter, 688704.
Biber, Douglas, Susan Conrad and Randi Reppen 1998. Corpus Linguistics: Investigating
Language Structure and Use. Cambridge: Cambridge University Press.
Biber, Douglas, Stig Johansson, Geoffrey N. Leech, Susan Conrad and Edward
Finegan 1999. Longman Grammar of Spoken and Written English. Harlow: Pearson
Education Limited.
Blake, Norman Francis 2004. Shakespeares Non-standard English: A Dictionary of his
Informal Language. Athlone Shakespeare Dictionary series. London and New
York: Continuum.
Boyce, Charles 1996. Dictionary of Shakespeare. Ware, Hertfordshire: Wordsworth
Editions Ltd.
Burrows, John F. 1987. Computation into Criticism: A Study of Jane Austens Novels and
Experiment in Method. Oxford: Clarendon Press.
Crystal, David and Ben Crystal 2002. Shakespeares Words: A Glossary and Language
Companion. London: Penguin.
Culpeper, Jonathan 2002. Computers, language and characterisation: an analysis
of six characters in Romeo and Juliet, in: Ulla Melander-Marttala, Carin Ostman
and Merja Kyt (eds), Conversation in Life and in Literature: Papers from the ASLA
Symposium, Association Sudoise de Linguistique Applique (ASLA), 15. Uppsala:
Universitetstryckeriet, 1130.
Culpeper, Jonathan and Merja Kyt 2002. Lexical bundles in Early Modern English:
a window into the speech-related language of the past, in: Teresa Fanego, Beln
Mndez-Naya and Elena Seoane (eds), Sounds, Words, Texts, Change. Selected Papers
from the Eleventh International Conference on English Historical Linguistics (11 ICEHL),
Current Issues in Linguistic Theory. Amsterdam: John Benjamins, 4563.
Enkvist, Nils Erik 1964. On defning style, in: Nils Erik Enkvist, John Spencer and
Michael Gregory (eds), Linguistics and Style. Oxford: Oxford University Press,
156.
Firth, John R. 1957. A synopsis of linguistic theory 19301955, in: Studies in
Linguistic Analysis, Philological Society, Oxford. Reprinted in F. Palmer (ed.) 1968.
Selected Papers of J. R. Firth. Harlow: Longman, 168205.
Foster, John 1908. A Shakespeare Word-Book: Being a Glossary of Archaic Forms and Varied
Usages of Words Employed by Shakespeare. London: Routledge.
Garside, Roger 1987. The CLAWS Word-tagging System, in: Roger Garside,
Geoffrey N. Leech and Geoffrey Sampson (eds), The Computational Analysis of
English: A Corpus-based Approach. London: Longman, 3041.
Halliday, Michael A. K. 1978. Language as Social Semiotic: The Social Interpretation of
Language and Meaning. London: Edward Arnold.
Howard-Hill, Trevor H. 19691972. Oxford Shakespeare Concordances. Oxford: Oxford
University Press.
Hunston, Susan and Gill Francis 2000. Pattern Grammar: A Corpus-Driven Approach to
the Lexical Grammar of English. Amsterdam: John Benjamins.
Kilgarriff, Adam, Pavel Rychly, Pavel Smrz and David Tugwell 2004. The Sketch Engine.
Proc EURALEX, Lorient: France: www.sketchengine.co.uk.
Leech, Geoffrey N. 1981. Semantics. Middlesex: Penguin Books.
Leech, Geoffrey N., Roger Garside and Michael Bryant 1994. CLAWS4: The tag-
ging of the British National Corpus. Proceedings of the 15th International Conference
on Computational Linguistics (COLING 94) Kyoto, Japan, 6228. www.comp.lancs.
ac.uk/computing/research/ucrel/papers/coling.html.
Leech, Geoffrey N., Paul Rayson and Andrew Wilson 2001. Word Frequencies in
Written and Spoken English Based on the British National Corpus. Harlow: Pearson
Education.
McArthur, Tom 1981. Longman Lexicon of Contemporary English. Longman: London.
Meyer, Charles F. 2002. English Corpus Linguistics: An Introduction. Cambridge:
Cambridge University Press.
Oncins-Martnez, Jos Luis 2006. Notes on the metaphorical basis of sexual lan-
guage in Early Modern English, in: Juan G. Vzquez Gonzlez, Montserrat
Martnez Vzquez and Pilar Ron Vaz (eds), The Historical Linguistics-Cognitive
Linguistics Interface. Huelva: Universidad de Huelva, 20524.
Onions, Charles T. 1986 [1911]. A Shakespeare Glossary (3rd edition enlarged and
revised by Robert D. Eagleson). Oxford: Clarendon Press.
Partington, Alan 1998. Patterns and Meanings. Amsterdam: John Benjamins.
Schiffrin, Deborah 1994 [1987]. Discourse Markers. Cambridge: Cambridge University
Press.
Schmidt, Alexander 1971 [1902]. Shakespeare Lexicon and Quotation Dictionary (Vols 1
and 2). Mineola, N.Y.: Dover Publications.
Scott, Mike R. 1999. WordSmith Tools. Oxford University Press. www.liv.
ac.uk/~ms2928.
Sinclair, John 1987. Collins Cobuild English Language Dictionary. London: Harper
Collins.
Sinclair, John 1991. Corpus, Concordance, Collocation. Oxford: Oxford University
Press.
Sinclair, John 2004. Trust the Text: Language, Corpus and Discourse (edited with
R. Carter). London and New York: Routledge.
Spevack, Marvin 19681980. A Complete and Systematic Concordance to the Works of
Shakespeare. Hildesheim: Georg Olms.
Stubbs, Michael and Isabel Barth 2003. Using recurrent phrases as text-type dis-
criminators: a quantitative method and some fndings, Functions of Language
10(1): 65108.
Wells, Stanley 1998. A Dictionary of Shakespeare. Oxford: Oxford Paperbacks.
Chapter 4
If I break time: Shakespearean Line
Endings on the Page and the Stage
Peter Kanelos
1. The ends of Shakespeares lines
For scholars surveying the development of Shakespeares poetic style,
the question of his progressively more enjambed and knotty verse is
a vital one. As Frank Kermode (2000) and Russ McDonald (2001)
have argued, Shakespeares increasingly complex and irregular poet-
ics developed concurrently with his increasingly complex and irregu-
lar representations of the world. Yet this evolution must have also had
dramaturgical purpose, serving the need for Shakespeares characters
to speak in a particular manner. By attending to Shakespeares dra-
matic texts as the working documents of players, scholars of perfor-
mance studies have helped us to gain a more nuanced understanding
of Shakespeare as a dramatic poet (see Palfrey 2005, Palfrey and Stern
2005, Stern 2004, Tucker 2002, Weimann 2000). In this spirit, I will
apply to Shakespeares dramatic verse questions of critical import to
our understanding of Shakespearean performance. Did Shakespeare
expect players speaking his verse to follow cues laid out by punctua-
tion? Did he intend players to be guided by lineation, indicating the
independent status of the line by speaking that line through, perhaps in
a single breath, and marking its end prosodically (via intonation, stress,
a pause, a breath, etc.)? Or did he anticipate that actors, when encoun-
tering an enjambed line, would not mark it in the phrasing? In other
words, when lines are enjambed by Shakespeare, did Elizabethan and
Jacobean players signify through some gesture of the voice that break
as a break, or did they follow organically the shape of the sentence? We
are aware how finely Shakespeare tuned the rhythms of his characters
speech through textually-signalled distinctions short lines, feminine
endings, overlapping cues and so forth. Were line endings also under-
stood as a sort of punctuation?
Line Endings on the Page and the Stage 85
To address these questions, it seems prudent to draw on experience from
both literary study and theatrical practice. As a professor of English, my
work is primarily of a textual order. In my Introduction to Shakespeare
class, howsoever often we might read lines aloud, our attention is invariably
drawn to the page. I teach as well in the Old Globe Master of Fine Arts in
Acting program and although my focus when working with actors is also
textual, given their profession, a slate of concerns emerges of an entirely
different order. Reading The Comedy of Errors, for example, when students in
a literature course discover a clever bit of enjambment, such as the Sprung
that springs up at the beginning of line 10 in Act 1 (see below), they tend
frst to notice this phenomenon spatially, their eyes registering diagonally
the shift from right to left and the scrolling movement down the page (what
I call the hypotenuse factor), as well as the capital S, all ensuring the
visual impact of springing sprung to their attention. Students of acting,
however, thinking ahead to performance, immediately begin to consider
how, when speaking these lines, one might replicate this effect. Of course,
Sprung from is a trochee, and Sprung will be lifted in pitch and/or
lengthened; yet is that enough? Or should line 9 end with a pause, breath
or some other mark of emphasis (even though we are mid-phrase), after
which Sprung presents itself with greater force and, perhaps, a mild sense
of surprise?
Merchant of Siracusa, plead no more.
I am not partiall to infringe our Lawes;
The enmity and discord which of late
Sprung from the rancorous outrage of your Duke,
To Merchants our well-dealing Countrimen,
Who wanting gilders to redeeme their liues,
Haue seald his rigorous statutes with their blouds,
Excludes all pitty from our threatning lookes:
(TLN 714)
The lines above are representative of the early plays, the great majority
end-stopped, so that enjambment is highly conspicuous. They compel us to
ask whether Shakespeare expected the end of lines to be marked prosodi-
cally that were not marked grammatically (lines where the syntax contin-
ues, without intervening punctuation), achieving one effect; or whether he
expected an actor to continue through the enjambment, lengthening the
sentence over the line, achieving another effect.
How to answer such a question responsibly? Peter Hall demands that the
line always be honoured:
In order to phrase naturally, there is an absolute need to feel where the
end of the verse line occurs. To run lines together risks incoherence and
often necessitates a pause in the middle of the next line for breath or
sense. This wrecks the structure of the verse. (Hall 2003: 29)
Neither John Barton nor Cicely Berry is as insistent about capping off
the ends of verse line, but both ask actors to attend to the end of the line
in most cases (see Barton 1984: 401; Berry 1997: 5281). Many directors
I have worked with (and several of my academic colleagues) maintain that
actors should simply follow the given punctuation when it makes sense to do
so and abandon it when it does not. Rather than trying to be original and/
or defnitive, I will attempt to lay out coherently the question of line breaks
in Shakespeare and to suggest possible avenues of response in the hope of
generating further discussion. I do want to take seriously, however, the possi-
bility of recapturing some sense of practice on the Elizabethan and Jacobean
stage, and thus I will be turning primarily to the First Folio as a control text,
with reference to several quartos, to achieve some sort of grounding.
2. The question of enjambment
The issue of enjambment, of course, is in the nature of the beast when deal-
ing with dramatic verse. Formal poetry, as Northrop Frye notes, has always
utilized the tension between form and syntax:
In every poem we can hear at least two distinct rhythms. One is the recur-
ring rhythm . . . a complex of accent, metre, and sound-pattern. The other
is the semantic rhythm of sense, or what is usually felt to be the prose
rhythm. (Frye 1957: 263)
The friction between these separate rhythms is felt acutely at the end
of enjambed verse lines: Viewed as competitors for the rhythmic control
of the poem, writes Mary Kinzie, the line is the conservative force and
the sentence the anarchist as it pulls attention away from meaning and
line, which seek to contain it (Kinzie 1999: 76). The interplay between
line and syntax is readily available on the page. Yet, verse on the stage,
ephemeral, subordinate to stylistic interpretation, maintains its structure
less evidently. The place to start to examine whether or not an Elizabethan
or Jacobean actor would have tried to preserve, when speaking lines of
verse, the sense of each line as a fxed unit seems to me to be the evolving
nature of Shakespeares use of enjambment. In the early plays, thought unit
and line almost always correspond, and each of these units fts rather neatly
into the larger grammatical structure dictated by the sentence. Here are
two examples of the openings of plays from the First Folio:
Cease to perswade, my louing Protheus;
Home-keeping youth, haue euer homely wits,
Wert not affection chaines thy tender dayes
To the sweet glaunces of thy honourd Loue,
I rather would entreat thy company,
To see the wonders of the world abroad,
Then (liuing dully sluggardizd at home)
Weare out thy youth with shapelesse idlenesse.
But since thou loust; loue still, and thriue therein,
Euen as I would, when I to loue begin.
(The Two Gentlemen of Verona, TLN 113)
Hvng be y heauens with black, yield day to night;
Comets importing change of Times and States,
Brandish your crystall Tresses in the Skie,
And with them scourge the bad reuolting Stars,
That haue consented vnto Henries death:
King Henry the Fist, too famous to liue long,
England nere lost a King of so much worth.
(The First Part of King Henry the Sixth, TLN 915)
End-stopping seems to ensue as a matter of expressing the sense of
the phrase, buttressed by the need to preserve the shape of rhetorical
fgures. Caesuras in these early works are rare, indicating that each line
was generally conceived of as a whole, even if part of a greater syntacti-
cal structure. The composition of verse in this manner seems to pattern
itself on the way in which the lines would be delivered; of course, con-
ventions for speaking dramatic verse were also established by the text
that made that verse available to the players. It may have been a closed,
self-reinforcing system of composition and delivery that Shakespeare
worked within in his early career; and it may have been those neat
boundaries that he worked against as he gained ever-greater mastery of
his own technique.
Tudor poetry, which utilized an iambic pentameter line, was generally
composed of two phrases, the frst of four syllables, the latter of six: as
George Gascoigne advises, There are also certayne pauses or restes in a
verse whiche may be called Ceasures . . . in a verse of tenne [syllables] it will
best be placed at the ende of the frst foure sillables (in Sylvester 1984:
325). In such an arrangement, as Thomas Wyatts Sonnet 33 illustrates, the
sentence is always subordinate to the line: What rage is this? what furour
of what kynd? / What powre, what plage, doth wery thus my mynd? (in
Sylvester 1984: 155). Elizabethan poets played a bit looser with this structure,
often alternating the syllabic groupings, but they maintained the division
of the line into two phrases and the priority of the line over the sentence,
as in Philip Sidneys Sonnet 36: Stella, whence doth this new assault arise, /
A conquerd yelden, ransackt heart to winne? (in Sylvester 1984: 437). The
voices of Shakespeares characters in the early plays, when speaking in
verse, have the timbre of such poetry. Compare the linear structure of the
passages above to the opening of Venus and Adonis:
Even as the sunne with purple-colourd face,
Had tane his last leaue of the weeping morne,
Rose-cheekt Adonis hied him to the chace,
Hunting he loud, but loue he laught to scorne:
Sick-thoughted Venus makes amaine vnto him,
And like a bold facd suter ginnes to woo him.
(Venus and Adonis, Quarto 1593)
Shakespeares early dramatic characters, like the speakers in his narrative
poems, express themselves in formal, structured language, a language in
which the explicit expression of the form itself is required and in which the
line is the primary vehicle for organizing complex thought.
There is self-evident tension at the opening of Richard III between the
frst and second lines of verse; that is to say that, for Richard, Now is the
winter of our discontent, is a distinct articulation of his interior world, set
in contradistinction to the effacement of this winter (and his identity) by
the son/sun of York. Syntactically speaking, there is no break between the
frst two lines; therefore modern editors nearly always leave the opening
lines end without punctuation:
Now is the winter of our discontent
Made glorious summer by this son of York;
(Wells and Taylor Oxford, 2nd edition)
Yet this is actually a rather bold instance of editorial prerogative. From
the frst extant quarto (1597) through fve other quarto editions (1598,
1602, 1605, 1612, 1622) to the First Folio (1623), and in two later quartos
(1629, 1634) a comma follows discontent.
Now is the winter of our discontent,
Made glorious summer by this sonne of Yorke:
(Q1 1597)
Now is the Winter of our Discontent,
Made glorious Summer by this Son of Yorke:
(Folio 1623)
Whereas modern editors prioritize the integrity of the sentence, Early
Modern editors were quite willing to retain a syntactically unnecessary
comma in order to retain, one would imagine, the implied break at the
end of the line. Were the original compositors/printers/publishers work-
ing from a scribal copy (foul/fair papers, prompt book) that included the
comma? If Shakespeare wanted a break, would he indicate it through punc-
tuation, sense of the sentence be damned? Were they indicating a pattern
of delivery that had become customary on the stage? Or was some confa-
tion of these two factors at play?
The opening lines of Richard III, then, as now, likely lodged themselves
in the collective cultural memory, so that the comma may have denoted
simply what audiences heard again and again during the plays perfor-
mance: Richard commenting on his despondency pause for effect fol-
lowed by the ironic turn towards the general state of delight pervading
the kingdom. Yet a question arises from this train of argument: if it was
conventional for actors to mark every line as a whole before continuing
on, and if printers/publishers were willing to use commas to indicate such
markings, why would any lines at all be left without some sort of punctua-
tion mark? Or, conversely, if it was the custom to break at every line, why
would the frst lines of Richard III need a seemingly anomalous comma
to signify something that would have been marked anyway? If we accept
Peter Halls contention that, The end of each line is in fact a punctua-
tion often more crucial than the regular punctuation itself (Hall 2003:
28), we may begin to fnd an answer. Assuming that to preserve the sense
of the whole line one must call attention to the close of the line in some
fashion, then there must have been a qualitative difference in the sorts of
ways that lines were brought to a close, perhaps indicated by punctuation
marks. According to Renaissance manuals of rhetoric, a comma, semi-co-
lon, colon, period, question mark, exclamation point, each indicated a
particular quality of marking of length, pitch, tone, etc. In Elementarie
(1582), Richard Mulcaster writes that the comma warneth us to rest there,
and to help our breath a little, while the period demands a longer pause
to help our breath at full, and a parenthesis signifes a lower & quicker
voice (Hunter and Lichtenfels 2005: 201). Thus it was not simply a break
at the end of Richards frst line that was indicated, but a certain sort of
break. I would further suggest that although every verse line ended with a
break, those that ended with a punctuation mark indicated a sort of inten-
tionality in the break. A punctuation mark assumed a self-conscious break
on the part of the speaker as he or she accommodated grammatical neces-
sity. Speech and sense coincided. In contrast, a line without punctuation
was still marked as a complete line even as it broke, in order to help the
auditor gain or retain a sense of meaning; yet that meaning may or may
not be entirely under the control of the character speaking. If we remind
ourselves that the line is the conservative force and the sentence the cha-
otic, in that open space at the end of unpunctuated verse lines, we can
see how Shakespeare began to explore something like what we in modern
terms call subtext.
3. Shakespeares dramaturgical development
As Shakespeares career progressed, he used enjambment with greater fre-
quency and fuidity, breaking lines in an ever-more radical fashion. Given
that Shakespeare was deviating progressively more from contemporary
practice and given that Shakespeare is always writing for the stage, we must
question the dramaturgical purpose here. At least two possibilities seem to
present themselves: (A) that there was on the stage persistent, habitual
end-stopping, inured by practice and custom, and that Shakespeare used
this convention to fragment units of thought and meaning, or (B) that
Shakespeare was encouraging, even demanding, his actors to allow the sen-
tence to trump the line, replicating more natural speech patterns. Looking
at the frst scene in Twelfth Night, we can return to the question of intention-
ality in pausing:
O she that hath a heart of that fne frame
To pay this debt of loue but to a brother,
How will she loue, when the rich golden shaft
Hath killd the focke of all affections else
That liue in her. When Liuer, Braine, and Heart,
These soueraigne thrones, are all supplyd and flld
Her sweete perfections with one selfe king:
Away before me, to sweet beds of Flowres,
Loue-thoughts lye rich, when canopyd with bowres.
(Twelfth Night, TLN 3947)
Surely part of the emotional and psychological agitation of this moment
is communicated by Orsino, lost in self-indulgent erotic reverie, lingering
on shaft and flld. To achieve this frisson, Shakespeare would have had
to have some confdence in his actors proclivity to mark the end of the
verse line. Orsinos narcissistic fantasy is most clear to us if we hear him
mark suggestively shaft and flld; yet if he does so consciously, even
intentionally, his sentiments are reduced to the vulgar. As John Barton
argues, You breathe at the end of verse-lines . . . [thus Shakespeare]
catches our trick of often pausing in ungrammatical places and running
on at full stops (Barton 1984: 42). The purpose of this practice is always
to reveal something critical about the speakers state of mind and/or
heart. Jan Gist, vocal coach for the Old Globe Theatre, widens the feld
of possibilities for marking line endings beyond just taking a breath. In
addition to or instead of inhaling air into the body after the last word of
each line, or taking a catch breath, or quick inhale, before the thought
continues, one might lengthen or strengthen the vowels and consonants
of the last word, fnalize the last sound of the last word, infect the pitch
on the last word to relate it forward, or pause, that is, take a moment of
silence after the last word. Alternatively, an actor can commit to and
fll out the idea or image of the fnal word, or emphasize the last word
as operative by separating it from the fow of the whole sentence (Gist
2008). The range of possibilities available to modern actors, even if not
organized according to the terms of formal rhetoric, was available to Early
Modern players as well, and Shakespeares increasing experimentation
with verse-endings indicates that he was experimenting foremost with the
sound of language on the stage. By working sometimes with, sometimes
against syntactical structures and the conventions of playing, Shakespeare
suggests the thought, emotion or impulse that lies beneath, behind or
between what is said. Sometimes a character is aware of and in control of
what is revealed, very often he or she is not. Thus naturalness of speech,
paradoxically, emerges from honouring the artifcial constraints imposed
by the line.
4. Line endings in late Shakespeare
By the late phase of Shakespeares career enjambment had become radical,
often splitting subject from verb, even articles from their referents. In the
example below, we see the frst lines literalizing the action of hanging,
as the eye drops from one line to the next, similar to the earlier example
of Sprung from The Comedy of Errors. Yet a quick glance over the passage
indicates that the line has loosened in extreme ways: I is split from Had,
that from Which, thou from His. To maintain the integrity of the line
by breathing or pausing at the lines end seems to risk losing entirely the
sense of the sentence.
Why he that weares her like her Medull, hanging
About his neck (Bohemia) who, if I
Had Seruants true about me, that bare eyes
To see alike mine Honor, as their Profts,
(Their owne particular Thrifts) they would doe that
Which should vndoe more doing: I, and thou
His Cup-bearer, whom I from meaner forme
Haue Benchd, and reard to Worship, who mayst see
Plainely, as Heauen sees Earth and Earth sees Heauen,
How I am galld, mightst be-spice a Cup,
To giue mine Enemy a lasting Winke:
Which Draught to me, were cordiall.
(The Winters Tale, TLN 40415)
This passage is taken admittedly from someone in a rather acute state
of emotional distress Leontes in The Winters Tale yet from this extreme
example it seems we can discern what it is that Shakespeare has come to
learn about verse and what he is trying to achieve. In breaking the line,
he breaks thought, not into distinct units but into fragments. Trying to
piece those fragments together is what speech itself is about attempting
to articulate both for others and for oneself the truth of ones experi-
ence. The need to maintain some sort of whole sense of the self is often
at odds with our fragmentary, mercurial experience of the world. On
the one hand, we have often imperfect traces of information, on the
other, we are compelled to shape this provisional sense of ourselves and
our environment into a whole, as if trying to piece together a mosaic in
the dark. This is certainly how Leontes must feel looking at Hermiones
bastard son:
Ynch-thick, knee-deepe; ore head and eares a forkd one.
Goe play (Boy) play: thy Mother playes, and I
Play too; but so disgracd a part, whose issue
Will hisse me to my Graue: Contempt and Clamor
Will be my Knell. Goe play (Boy) play, . . .
(The Winters Tale, TLN 26872)
In three consecutive lines, Leontes splits subjects from their actions:
I/play, issue/will hisse, Clamor/will be. George T. Wright suggests that
Shakespeare at this stage in his writing has moved to prioritize the sen-
tence over the verse line: Such endings lead us without pause over the
line-ending (Wright [1988], in McDonald 2004: 873). Yet, it is precisely a
split the unexpected, yet absolute division he now feels between father
and son that Leontes is experiencing. Patrick Stewart, breathing at the
end of each Leontes line above, described the effect: It gave me the sensa-
tion of somebody who was beginning to spiral out of control (Barton 1984:
43). And in the same way that, breathing at, pausing for, or attending to
the end of each line, he takes a moment to fnd the verb to connect to its
preceding subject, so, too, does Leontes, given the unanticipated turn life
has presented to him, fnd what he feels to be an appropriate action. Each
turn of the line allows for a discovery. In the brief moment it takes to make
that discovery, we witness Leontes attempt at self-cohesiveness and judge
for ourselves his success.
A glance at Posthumus in Cymbeline will illustrate the same principle in
action. Believing that he is responsible for the death of Imogen, his words
are whirling as he seeks his purpose:
Yea bloody cloth, Ile keep thee: for I am wisht
Thou shouldst be colourd thus. You married ones,
If each of you should take this course, how many
Must murther Wiues much better then themselues
For wrying but a little? Oh Pisanio,
Euery good Seruant dos not all Commands:
No Bond, but to do iust ones. Gods, if you
Should haue tane vengeance on my faults, I neuer
Had liud to put on this: so had you saued
The noble Imogen, to repent, and strooke
Me (wretch) more worth your Vengeance . . . .
. . . so Ile fght
Against the part I come with: so Ile dye
For thee (O Imogen) euen for whom my life
Is euery breath, a death: . . .
(Cymbeline, TLN 285868/28814)
We fnd through a series of line breaks Posthumus discovering what course
of action he must take. From wishing he neuer / Had liud and that ven-
geance had strooke / Me, to resolving to fght / Against the Romans
and dye / For thee, we witness the quicksilver nature of Posthumus emo-
tions and the progressively cogent plan he is shaping. Yet, if these lines are
spoken without some sort of break at each lines end, then Posthumus
speech is smoothed out so completely as to have the quality of resolve
before that quality is discovered. In other words, at each turn of the line, an
actor, whether he or she chooses to indicate this through a breath, a pause,
emphasis or some other vocal gesture, has the opportunity to represent
the active thought processes of a character, formed in the crucible of cir-
cumstance. The more radical the disjuncture, that is, the more that a line
splinters a unit of thought where we would not conventionally have that
thought broken, the more evident the fracture and the more aware we are
that, from a myriad of possibilities, the words that continue the thought
as the next line continues are chosen in the moment, refecting a living,
breathing intelligence.
It is curious that as Shakespearean lineation becomes increasingly bro-
ken and irregular, his plays are increasingly concerned with the overarch-
ing power of art and artifce and its ability to give structure to a seemingly
disordered world. Consider the nine scenes of The Tempest frst balanced
against last, second against penultimate, so that the play folds out from its
centre the betrothal of Miranda and Ferdinand, engineered by Prospero
like the halves of a paper snowfake. This play begins with a raging storm,
yet ultimately bears witness to our ability to control what is outwardly unruly
and untamed. Prospero is consistently confdent that he is in control of his
Art. Yet in the moment when Prospero chooses mercy over vengeance, we
fnd in his speech some rather interesting slippage:
Hast thou (which art but aire) a touch, a feeling
Of their affictions, and shall not my selfe,
One of their kinde, that rellish all as sharpely,
Passion as they, be kindlier moud then thou art?
Thogh with their high wrongs I am strook to th quick,
Yet, with my nobler reason, gainst my furie
Doe I take part: the rarer Action is
In vertue, then in vengeance: . . .
(The Tempest, TLN 19718)
Prosperos decision to forgive the trespasses of those who have wronged
him is much more engaging if we witness it in process. Ariels feeling, which
is what prompts Prosperos forgiveness, is highlighted by its placement at
the end of the line and, more critically, by the unexpected (syntactically
speaking) break that follows. Prosperos furie must be relinquished; but
its hold lingers with the line. Finally, if his conclusion is read without inter-
ruption the rarer Action is in vertue, then in vengeance Prosperos
sentiment is rather sententious. If, however, the rarer Action is . . ., and
we observe a moment flled with something before Prospero proceeds res-
ignation, enlightenment, relief then his commitment to vertue over
vengeance is active, in the moment, and most importantly, purposeful.
We see order, imposed by humane sentiment, overcoming the potential
for chaotic vengeance (Bacons wild justice); and as this is a dramatic
performance, we fnd it much more dramatically compelling to watch this
happen than to feel that it had been concluded before the action even
commenced.
The radical turns of the line in Shakespeares late plays mirror the radical
turns in the narrative of these stories. Each line is a horizon on a voyage.
The unexpected is discovered around corners. Yet we cannot see where
the turns are taken if the horizon has not been marked. Poetry ultimately
engages in the play between expectation and surprise, between pattern and
variation. As I am constrained to advise the actors that I work with, I fnd
it my obligation to make as clear as possible the patterns of Shakespeares
verse, not to impose those patterns rigidly, but to do so frmly so that when
an actor swerves, he or she swerves with purpose and intent, retaining con-
trol even where chaos seems to be imminent. In fact, The Tempest is the end-
point, begun in Pericles, working through The Winters Tale and Cymbeline,
of arts conquest over chaos. In Pericles, the vicissitudes of chance are over-
come through the direct intervention of the goddess Diana; in Cymbeline,
Jupiter intercedes. Yet in The Winters Tale, the oracle of the gods fails to avert
disaster and must be supplemented by artifce in the end. By The Tempest,
Shakespeare has dispensed with any sort of divine interference and has
located in the hands of a human character the power to use art to super-
sede chaos. Insofar as Prospero is an avatar of Shakespeare, one who self-
consciously uses his art to order the world, there seems to be a concession
to the playwrights own role in shaping out of even the most wildly irregular
verse something close to a cohesive vision.
References
Quotations from Shakespeares plays are taken from Charlton Hinman
(ed.), The Norton Facsimile: The First Folio of Shakespeare (New York: W.W.
Norton & Co., 1968), unless otherwise noted, with line numberings cor-
responding to the Through-Line Number system (TLN).
Other Shakespeare references are to:
Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery
(eds) 2005 [1986]. William Shakespeare: The Complete Works. Oxford:
Clarendon Press.
Facsimile of the 1593 Quarto edition of Venus and Adonis accessed from
Internet Shakespeare Editions at http://internetshakespeare.uvic.ca/
Library/plays/Ven.html.
Facsimiles of quarto editions of Shakespeares plays accessed from the
British Library online at www.bl.uk/treasures/shakespeare/homepage.
html
Barton, John 1984. Playing Shakespeare. New York: Anchor Books.
Berry, Cicely 1997. The Actor and the Text. New York: Applause Theatre Books.
Frye, Northrop 1957. The Anatomy of Criticism. Princeton: Princeton University
Press.
Gist, Jan 2008. Inhaling Shakespeare. Presented at the Voice and Speech Trainers
Association Conference, New Orleans, Louisiana, July 2008.
Hall, Peter 2003. Shakespeares Advice to the Players. New York: Theater Communications
Group.
Hunter, Lynette and Peter Lichtenfels (eds) 2005. Shakespeare, Language and the
Stage. London: Thomson Learning.
Kermode, Frank 2000. Shakespeares Language. New York: Farrar, Straus and Giroux.
Kinzie, Mary 1999. A Poets Guide to Poetry. Chicago: The University of Chicago
Press.
McDonald, Russ 2001. Shakespeare and the Arts of Language. Oxford: Oxford University
Press.
Palfrey, Simon 2005. Doing Shakespeare. London: Arden.
Palfrey, Simon and Tiffany Stern 2005. Shakespeare in Parts. Oxford: Oxford University
Press.
Stern, Tiffany 2004. Making Shakespeare. London: Routledge.
Sylvester, Richard (ed.) 1984. English Sixteenth-Century Verse: An Anthology. New York:
W.W. Norton & Co.
Tucker, Patrick 2002. Secrets of Acting Shakespeare: The Original Approach. New York:
Routledge.
Weimann, Robert 2000. Authors Pen and Actors Voice. Cambridge: Cambridge
University Press.
Wright, George T. 1988. The play of phrase and line, from Shakespeares Metrical Art.
Berkeley: University of California Press, 20728, in: Russ McDonald (ed.) 2004.
Shakespeare: An Anthology of Criticism and Theory 19452000. Oxford: Blackwell
Publishing: 86179.
Further reading
Adamson, Sylvia, Lynette Hunter, Lynne Magnusson, Ann Thompson and Katie
Wales (eds) 2001. Reading Shakespeares Dramatic Language. London: Arden.
Hope, Jonathan 2005. Shakespeares Grammar. London: Arden.
Hyland, Peter 2003. An Introduction to Shakespeares Poems. New York: Palgrave
Macmillan.
Prince, F. T. (ed.) 1960. The Poems. London and New York: The Arden Shakespeare,
Methuen.
Chapter 5
Subject-Verb Inversion and Iambic Rhythm
in Shakespeares Dramatic Verse
Richard Ingham and Michael Ingham
1
1. Introduction
Studies of the interaction of metre and syntax in Shakespearean verse,
though uncommon, are of great potential interest for the prosodic aspects
of his work in performance. Culpeper (2001: 202) has drawn attention to
the often underrated signifcance of syntactic features in Shakespearean
texts. His discussion is restricted to instances where syntactic features relate
to cognitive organization of speech, but can be usefully complemented by
considering what choices syntax makes available to the writer, and what
their prosodic consequences are for the performer. One signifcant type of
syntactic choice is the inversion of main constituents. Examples of inverted
word order occur in a range of Shakespearean contexts, both verse and
prose, with a range of textual impacts. The device is commonly assumed
to have provided English verse writers, especially in the pre-contemporary
period, with a ready means of complying with metrical exigencies. In this
chapter, using an empirical database of 998 target instances sampled across
Shakespeares career, the validity of such an assumption is examined with
respect to one particular syntactic option: the construction in which a sub-
ject could optionally be inverted round the verb in the context of an initial
non-subject constituent. We address the issue by means of a dual approach:
frst a linguistic analysis across Shakespeares dramatic verse career, and sec-
ondly a comparison of interpretative styles by contemporary Shakespearean
performers.
The subject-verb inversion option can be simply illustrated from one
of Shakespeares main sources, Holinsheds Chronicles. The examples in
(1) invert the subject, while those in (2) do not:
(1)a. Then longed the duke sore to heare what he would haue said, . . .
Holinshed, Chronicles (405)
Subject-Verb Inversion and Iambic Rhythm 99
(1)b. . . . [t]his did he to driue time, and to put his enimies out of all sus-
picion . . . Holinshed, Chronicles (266)
(2)a. Then all the companie sware to him fealtie, and did to him hom-
age . . . Holinshed, Chronicles (420)
(2)b. The headlesse trunk he commanded to be hoong vp upon an high
paire of gallowes. Holinshed, Chronicles (266)
The inversion of subject and verb in (1) may be seen as a relic of the
inversion of subject and verb in Old and Middle English
2
after clause-initial
adverbials (cf. 1a) and direct objects (cf. 1b), a rule which continues to char-
acterize present-day Germanic languages. It was a fairly common minority
pattern in fourteenth- to sixteenth-century English (Baekken 1998, 2000),
and seems to have typifed a more literary register. Later in the Modern
English period, most types of subject-verb inversion fell out of use except
in archaizing verse styles
3
. An English author writing in the late sixteenth
century, such as Holinshed, is likely to have perceived it as a viable stylistic
option, to be used with discretion, rather than an out-and-out archaism.
The issue considered here was how the availability of inversion interacted
in Shakespeares verse drama with the presence of iambic rhythm, in par-
ticular how metre and syntax comported when the subject was a personal
pronoun (henceforth Spro). Unless emphatic or contrastive, a subject pro-
noun never bears stress in natural English speech rhythm. Therefore, a
regular iambic pentameter, with its alternation of weak and strong syllables,
would often not map directly onto the speech rhythms of English, cf. the
following examples from Shakespeares prose:
(3)a. Yet you began rudely. (Twelfth Night 1.5.203)
(3)b. . . . for truly I think you are damned. (The Merchant of Venice 3.5.5)
In the sequences you began and truly I the subject pronoun either precedes
or follows a phonologically weak syllable, so were this to be recast as blank
verse two weak beats would occur in succession, a problem for regular iam-
bic rhythm. Shakespeares verse appears to show numerous cases where the
inversion of verb and subject pronoun respects iambic metre, for example,
(4) The other part reserved I by consent, (Richard II 1.1.128)
The uninverted order The other part I reserved would not be metrical, since
the sequence I reserved fails to ft an iambic foot. It is therefore plausible to
see subject pronoun inversion as a useful option available to an author of
iambic verse, since in principle the respective positions of verb and Spro
could be varied, depending on the syntactic context, to ft the metre. In
fact, however, we show in this study that this is very unlikely to have been
Shakespeares practice, and that in earlier works he tended to place an
inverted subject pronoun in an ictic position,
4
seemingly contrary to a strat-
egy driven by the needs of regular iambic rhythm.
Shakespeares verse is known to be characterized by a striking fexibil-
ity in the interaction between syntax and metre (see for example, Wright
1983), and it is of course not argued here that the iambic scheme formed
an absolute requirement on the wording of the line. Nevertheless most
studies of Shakespeares metre consider that iambic rhythm is present as a
factor providing a counterpoint to the speech rhythms of English. This will
present an interpreter of a Shakespearean line with clear alternatives as to
whether iambic rhythm is used or not, for example:
(5) And are you yet to your own souls so blind/That you will war with God
by murdring me? (Richard III 1.4.2478)
In line 248 the three function words that you will succeed each other with
no intervening content word. A regular iambic rhythm would place a beat
on the subject pronoun you, but speech rhythm normally would not, as it is
a function word. We shall see later in this article how different actors react
to this alternative, in interpretatively sensitive ways.
It is interesting to consider how Shakespeare may have changed his prac-
tice with respect to the ft between syntax and metre over his authorial
career, and then to ask what approaches to rhythmic variation are taken
by performers of his verse drama. Here we seek to provide a preliminary
assessment of the major aspects of Shakespeares use of syntax and metre
in cases where these two components inverted syntax for stylistic effect,
and respect for iambic rhythm may have created a mismatch. We shall
examine how subject pronoun inversion evolved over Shakespeares writing
career, and then how rhythmic delivery in contemporary performance has
engaged with it. Although our study comprises a linguistic analysis and an
interpretative analysis which clearly differ in methodological approach, it is
intended to function as a single line of enquiry in which the initial linguis-
tic analysis provides the empirical basis for discussion of earlier versus later
practice by the poet, and of alternative interpretations by performers. The
study is organized as follows: in section 2 we briefy review previous fndings
relevant to the issue discussed in this enquiry. In section 3, we discuss the
sources and methodology we have used to track the evolution of subject
pronoun placement in Shakespeares verse drama, and its relationship with
iambic rhythm. Section 4 presents the results of these analyses. In section 5
we turn to the issue of how recent recorded performances of the plays stud-
ied have handled the tension between syntax and metre, drawing attention
to the subtle interpretative effects derived by particular performers. Section
6 concludes the chapter with some observations on the earlier as compared
with the later authorial style of Shakespeare as to the link between speech
rhythm and metrical rhythm.
2. Research background
Shakespeares use of inversion constructions has been studied by Houston
(1988), who found that verb-subject (VS) order tended to decline in plays
thought to have been written from the end of the 1590s onwards. However,
whether this decline affected inversion with subject pronouns (henceforth
VSpro) to the same extent as inversion with full nominal subjects was not
stated. Since the same study also found increasing frequency of Subject-
Object-Verb patterns from roughly the same time-point, Shakespeare can-
not be said to have moved away from all types of inverted syntax in his
later plays. Hence VSpro could have remained constant as a stylistic option
throughout Shakespeares artistic career.
Hope (1994) studied syntactic patterns in Shakespeares work and found
that do-support fell within a percentage frequency band, which was distinct
from that of other contemporary authors such as Marlowe and Middleton.
His work showed the value of considering the use of a variable feature of
syntax in particular writers of this period. The use of a syntactic variable is
seen to be not merely a random feature but clearly under the control, con-
scious or not, of the author. Hope did not, however, investigate inversion
patterns, the focus of the present study. To our knowledge there has not
been any empirical investigation of how the interplay of metre with syn-
tactic variation was handled by Shakespeare in the domain of subject-verb
inversion. The present study is a modest attempt to remedy this defciency.
Shakespeares metre has received a great deal of critical attention which
can only briefy be summarized in the space available here. Abbott (1870),
as well as clarifying the notions of emphatic stress and regular, metrical
stress, addressed the problem of tension between iambic rhythm and speech
rhythm, which continues to cause diffculty for performers today, and
indeed for critics too, judging by the vigorously polemical online debate
in 2002 on the website Shaksper.net between David Wallace and David Evett
(Wallace 2002).
Wallace argued that a polysyllabic word was not rhythmically free in
Shakespeare: its stress pattern was respected by the rhythmic structure of
the line in which it appeared, insofar as its primary stressed syllable had to
be in an ictic position. A monosyllabic word, however, was not affected by
this requirement and so could occur either in or out of an ictic position.
The empirical prediction was thus that in the sample of plays used here
polysyllabic verbs would always map their primary stressed syllable on to an
ictus of iambic rhythm, while monosyllabic verbs would show no associa-
tion with an ictic beat. Evett, on the other hand, was sceptical as to whether
metre is determined by polysyllables alone, and of the claims of metrical
theorists in general. For the purposes of this study, Wallaces position offers
the advantage of being a hypothesis to be empirically investigated, by distin-
guishing between monosyllabic and polysyllabic contexts in which subject
pronouns occur. This point will be taken up again in section 4 below.
Their exchanges in some ways echoed the disputes in defence of alterna-
tive representations of the iambic pentameter in the late 1960s and early
1970s, informed by the then recent work in the feld of Chomskyan struc-
tural linguistics. Here too the leading theoretical model presented the
notion of a correspondence between the abstract metrical pattern and the
surface features of any given line. For Halle and Keyser (1971) the concept
of stress maximum or maxima allows the Abbott distinction between iam-
bic rhythm and emphatic accent to be handled in terms of the phrasal or
surface stress in relation to an underlying metrical pattern. They postulate
that such syllables bearing a stress maximum can only be preceded and
followed by unstressed syllables.
Wright (1983) traces the profound changes and radical experiments with
form undertaken by Shakespeare, from the more conventional metrical
structures of the early plays to the rupture with regular metre evident in
late plays like Cymbeline and The Tempest. He differentiates between what he
calls the underlying deep structure of the iambic rhythm and the surface
form of the increasingly irregular distribution of syllables and semantic
components in Shakespeares mature verse. By the end of Shakespeares
writing career, according to Wright, the formal restrictions of earlier tradi-
tion-bound iambic poets such as Gascoigne had been discarded. Instead of
the variables of beat (ictus), phrasal stress and major words coinciding, as
had been customary previously, in Shakespeares late plays there is a delib-
erate play of counterpoint (polyphony as Wright refers to it), by which the
conventional equilibrium is upset and the line more and more cast into
structural doubt (Wright 1983: 155). Scepticism about the power of smooth
rhetoric and the deceitful uses and users of language

(Wright 1983: 156)
is now much more evident in the frequently irregular or disruptive distri-
bution of phrasal stresses, according to Wrights insights. At this stage of
the development of his versifcation Shakespeare is prepared to take risks
and depart from tried and tested structures: The hallmark of the new verse
is the autonomy of the phrase, no longer obedient to the line or the line-
segment but with its own authority well established (Wright 1983: 156).
For Wright the polyphony allied to the metre is the key to Shakespeares
dramatic verse: semantic content and emotional design ride on the meter
and the syntax (Wright 1983: 157).
The notion of a counterpoint between underlying iambic rhythm and
surface speech rhythm was pursued empirically by Tarlinskaja (1983), who
conducted a detailed examination of how English phrasal stress, falling on
content words but not function words, interacted with the ictic beat. She
found a clear evolution over the course of Shakespeares writing career in
the likelihood with which phrasal stress would be placed on the sixth and
eighth syllables. This fnding is relevant to the concerns of the present study
in that the placement of phrasal stress on an ictic beat in Shakespeare was
not random, but appears to have varied in a coherent way across time. In
this study we use the convenient term ictic to refer to placement on an
even number syllable of a pentameter line, without prejudice to the issue of
where syllables bearing stress maxima in speech rhythm occur.
3. Source texts and method of analysis
For purposes of investigating language change, studying a particular idio-
lect over time is informative, as compared with analysing samples from a
large number of individuals, in which particular trends may get blurred.
As is well known, however, the textual transmission of the works ascribed to
Shakespeare is far from straightforward, which complicates the reliability
of its attribution to a single idiolect. Accordingly, the analysis we present
in this study cannot be claimed to represent more than that of a corpus of
works plausibly attributed to the same individual. In the case of a literary
author, stylistic variation across genres presents further complications. We
therefore opted to address the issues raised in the preceding sections by
analysing a subset of the verse dramas of Shakespeare, namely those based
on historiographical sources, in which it was hoped to observe a greater
number of relevant data-points
5
. For this purpose, we excluded comedies
and also tragedies lacking what might be called a public narrative, that is,
a succession of events in which a community beyond the protagonists own
immediate circle participates, for example, the wider public domain in
which King Lear is played out as compared with the more private sphere of
action in Romeo and Juliet. We thus identifed 16 dramas written principally
or exclusively in verse, which were divided into the following chronological
periods in terms of the dates conventionally attributed to them:
Period I (15921597): Richard III, Titus Andronicus, Richard II, King John,
Henry IV 1
Period II (15981603): Henry IV 2, Henry V, Julius Caesar, Troilus and
Cressida, Othello
Period III (16061610): The History of King Lear, Macbeth, Antony and
Cleopatra, Coriolanus, The Winters Tale, Cymbeline
Note that these temporal categories do not necessarily correspond to
those of previous studies of Shakespeare, but were chosen as a means of
roughly dividing the available data into three periods determined by their
known or presumed dates of composition. The reference edition used was
the Oxford Shakespeare (Wells and Taylor 1986).
The syntactic construction targeted in these texts was the inversion of the
fnite verb and subject pronoun in the context of a preceding non- subject
constituent. The latter was defned as an adverbial, a direct or indirect
object, or a subject/object complement. Examples of these constructions
from Shakespeares prose are provided below
6
:
(6)a. Now will not I deliver his letter, . . . (Twelfth Night 3.4.181)
(6)b. The maid will I frame and make ft for his attempt.
(Measure for Measure 3.1.2578)
(6)c. Bawd is he doubtless, . . . (Measure for Measure 3.2.65)
Because line-fnal metre poses particular issues in Shakespeares verse,
cases where the relevant verb or subject pronoun stood in the ffth foot
were not included. We also excluded cases where a subject pronoun had an
auxiliary contracted to it, as well as incomplete or hanging half lines and
other clear cases of irregularity as regards the iambic pentameter.
The variables analysed, then, were the syntactic order of subject pronoun
and verb in declarative main clauses with an initial non-subject constituent,
and the positioning of a subject pronoun in ictic position. All types of verbs
were counted for the purpose of the analysis, regardless of whether they
were main or auxiliary verbs. Monosyllabic verbs and verbs of two or more
syllables were analysed separately, in view of Wallaces argument that stress
placement in polysyllabic verbs in Shakespeare respects ictic beats. A more
refned analysis could subdivide our categories of analysis, but for the pur-
poses of an initial examination of the feld it was judged best to maximize
the data frequencies to see how common the phenomena were and what
was revealed by a quantitative investigation conducted on this basis.
4. Results of frequency analyses
The issue addressed in this part of the study is how Shakespeare negoti-
ated the relationship between word order and iambic rhythm, specifcally
in contexts where syntax allowed him fexibility in positioning a subject
pronoun. To recall the proposal of Wallace (2002) that iambic rhythm
was established by the presence of a polysyllabic word in the pentameter
line is adopted in this study as a starting hypothesis. In the texts analy-
sed, of 47 polysyllabic verb tokens co-occurring with a subject pronoun,
in every case the naturally stressed syllable of the word fell on an even-
numbered syllable. Wallaces claim is thus entirely upheld. Unless the
poet had been following a metrical scheme of some kind, this outcome
would have been an inexplicable coincidence. Since polysyllabic verbs
occurred in varying positions across the ten syllables of the line, that
scheme must therefore have been applicable to the ten-syllable line as a
whole. An iambic rhythm clearly constituted a default option for the ten-
syllable line, in accordance with the traditional stance adopted by adher-
ents of the iambic pentameter analysis of Shakespeares dramatic verse.
The position taken here thus endorses both the recently proposed claim
of a rhythmic constraint associated with polysyllabic verbs, and the tra-
ditional assumption that iambic rhythm is at least underlyingly present
in Shakespearean verse.
Monosyllabic verbs, however, are a different matter. On the approach fol-
lowed here, single-syllable words, including verbs, were free to appear in
either odd or even syllable positions in the line. There is no expectation
that, in pairs of monosyllabic verbs and subject pronouns, one or the other
will be favoured in, for example an odd syllable position. Shakespeare may
nevertheless have shown a preference for placing a particular form, pro-
noun or verb, in a particular position. In terms of natural speech rhythm,
non-contrastive personal pronouns are always unstressed, whereas verbs,
including monosyllabic ones, may bear stress. With polysyllabic verbs, an
underlying iambic rhythm was evidently at work. A preference for placing a
single-syllable verb in an even (ictic) syllable position and a subject pronoun
in an odd (non-ictic) position would tell us that Shakespeare was broadly
tending to align the rhythm of the line on iambic rhythm with monosyllabic
verbs as well. If this turned out to be the case, the question can then be
addressed whether Shakespeare manipulated the possibility of subject-verb
inversion as a device for keeping subject pronouns out of ictic positions.
An initial analysis was therefore carried out, investigating whether in the
16 target plays a monosyllabic verb fell on an odd or an even syllable. The
results are shown in Table 5.1.
This established that Shakespeare was four times more likely to position a
pronoun in a weak position than in a strong position, clearly indicating that
he wrote with an awareness of an underlying weak-strong rhythmic alterna-
tion infuencing the disposition of monosyllabic verbs and pronouns. This
feature of his verse did not change over the three periods studied.
Turning now to the question of how frequently Shakespeare deployed the
VSpro option, Table 5.2 frst presents the overall frequencies of inverted
versus non-inverted syntax.
In inverted contexts with a monosyllabic verb, therefore, SV and VS
orders were almost equally weighted options for Shakespeare in both
Table 5.2 Frequencies of VS and SV syntax with main clause subject pronouns in
inversion contexts featuring a monosyllabic fnite verb
VS SV Total
Period I (15921597) 214 (50.8%) 207 421
Period II (15981603) 109 (51.8%) 104 213
Period III (16061610) 95 (30.0%) 222 317
Average 418 533 951
Table 5.1 Syllable position of subject pronouns accompanying monosyllabic verbs

Odd Even T
N % N %
Period I 357 84.8 64 15.2 421
Period II 158 74.2 55 25.8 213
Period III 274 86.4 43 13.6 317
Average 789 83.0 162 17.0 951
periods I and II. It was only from 1606 onwards that a clear preference
emerges for the non-inverted order.
So far, then, it can be seen that VSpro was a freely available option for
Shakespeare at least up to and including Othello, but that he strongly
favoured putting the subject pronoun in an odd (non-ictic) position. Was
VSpro used as a way of complying with the latter preference? Table 5.3
shows whether subject pronouns following monosyllabic verbs fell on an
odd (weak) or an even (ictic) syllable.
Although the majority of the time Shakespeare tended to place an
inverted subject pronoun on an odd (weak) syllable, it can be seen that
in a substantial minority of cases, especially in periods I and II, inversion
caused the pronoun to stand in an even (ictic) syllable, contrary to our
initial hypothesis that VSpro syntax was used to obtain conformity with an
underlying iambic rhythm.
Indeed, when all subject pronouns in even syllables are considered, both
in VS and in SV order, a rather striking development emerges. In over three
quarters of cases in period I, Shakespeare appears to have used VS order to
position the subject pronoun in an even (ictic) syllable (see Table 5.4).
Table 5.3 Frequencies of inverted subject pronouns in even and odd syllables with
monosyllabic verbs in Shakespeares dramatic verse
Inverted subject
pronoun in
even syllable
Inverted subject
pronoun in
odd syllable
Total of inverted
subject pronouns
N % N % N
Period I
(15921597)
49 22.9 165 77.1 214
Period II
(15981603)
29 26.6 80 73.4 109
Period III
(16061610)
10 10.5 85 89.5 95
Table 5.4 Frequencies of main clause VSpro and SproV syntax in inversion
contexts with monosyllabic verbs, with subject pronouns in even syllables
VSpro SproV Total
Period I (15921597) 49 (76.6%) 15 64
Period II (15981603) 29 (52.7%) 26 55
Period III (16061610) 10 (23.3%) 33 43
Average 88 (54.3%) 74 162
That is, VSpro order was adopted proportionately far more often if the
pronoun was ictic than the overall average rate of VSpro order. The com-
parison is displayed graphically in Figure 5.1.
Whereas in periods II and III, pronouns in an even syllable (ictic) are
about as frequent as we would expect in terms of the overall rate of VSpro,
there was clearly a strong preference in the period I plays for VS order. Why
this should have been the case is an intriguing question which awaits fur-
ther exploration; it may be related to expressive dimensions such as char-
acterization in the particular plays, adopting the perspective of Culpeper
(2001).
Our analyses show, in any case, that inversion did not constitute a device
for managing compliance with iambic rhythm. Shakespeares use of the
two traits was quite clearly unrelated: on the one hand inversion shows a
monotonic decline in subject pronoun inversion from about half to about
a quarter of contexts (see Table 5.2), while on the other (see Table 5.1)
we found a fairly low incidence of ictic inversion showing non-monotonic
variation around the 20 per cent mark over the period studied.
The foregoing data analysis has thus established three initial points. First,
inversion of subject pronouns declines over the course of Shakespeares
work. Secondly, VSpro order is not associated with odd syllable placement
of a subject pronoun: Shakespeare placed subject pronouns on an even syl-
lable about as often in his later plays as in his earlier ones, but now mostly in
SproV order. Finally, Shakespeares earlier verse dramas strongly associate
0
10
20
30
40
50
60
70
80
90
Per I Per II Per III
VS pro
VS pro ictic
Figure 5.1 Syntax of subject pronoun after an initial non-subject constituent: per-
centage of VSpro order overall, and percentage of VSpro order when the pronoun
was in an even (ictic) syllable.
ictic position of the subject pronoun with VSpro order. These fndings,
though clearly of interest as regards authorial practice in the interaction
of syntax and metre, may further be enhanced by consideration of how the
verse is performed. It is, after all, in the interpretation of the lines that the
subtlety of the authors craft, as well as that of the actor, is brought most
keenly into contention. In the next section, therefore, we examine how
established actors deal with the tension between iambic rhythm and syntax
with particular reference to the cases of inverted subject pronouns in the
early plays.
5. Contemporary interpretations in performance
Performing Shakespeares verse poses considerable problems in terms of
understanding his sometimes complex metrical patterns, and especially in
respect of stress placement. The relationship between syntax and metre
and between natural speech rhythm and metrical rhythm is conceptu-
ally challenging. Native-speaker intuition alone is inadequate to explain
the complexities. In the teaching and directing of performance to non-
native speakers, invoking simplistic iambic rules for the delivery of lines
is essential, yet, at the same time, frustratingly inadequate. Likewise, for
native speakers we cannot assume competence in discerning appropriate
patterns of phrasing in the enunciation of Shakespearean verse or, for that
matter, of other verse dramatists. Our contemporary auditory sensibilities
are ill-attuned to the prosodic structure of metrical verse in performance,
accustomed as we are to free verse with fairly arbitrary patterns or to the fre-
quently monotonous beats of much popular music. The subtleties of synco-
pation in English verse and lyrics can still be perceived in modern rap and
other verse forms. However, sensitivity to metrical regulation and variation
in contemporary Shakespeare performance so challenges, or is assumed to
challenge, the contemporary ear that many flm directors of Shakespeare
adaptations virtually dispense with the verse altogether in favour of a prose-
like, naturalistic speech delivery.
For this reason BBC recordings of the Shakespeare canon between 1978
and 1983 have been used as a yardstick for assessing the sensitivity of con-
temporary actors to metrical features in general and iambic rhythm in par-
ticular. The rationale for selecting a sample of the BBC recordings is that the
actors represent universally acknowledged specialists in Shakespearean per-
formance, of the calibre of John Gielgud and Derek Jacobi, whose handling
of verse drama may be considered authoritative. Recorded performances
were essential for the purposes of the investigation, since impressionistic
recollections of live performances cannot be empirically studied. The con-
sistently high standard of enunciation in the BBC series facilitated close tex-
tual/aural analysis with the unaided ear and the assistance of pause and
playback commands.
7
Furthermore, the BBC versions enjoyed wide dis-
tribution and many, though by no means all, were critically acclaimed. By
comparison with the hit-and-miss approach of feature flm directors and
actors to Shakespearean versifcation, full-length stage-like performances,
as opposed to abbreviated and truncated flmic versions of the works, were
felt to be more appropriate. In the BBC versions flmic devices are used dis-
creetly in such a way that they underscore rather than dominate the verbal
delivery. In some respects the BBC series can be seen as taking up the baton
of the masterful flm performances of Laurence Olivier as Hamlet, Henry
V, Richard III and King Lear, which represented for many the apogee of
Shakespearean diction and the perfect fusion of sound pattern and seman-
tic signifcation in the phrasing of Shakespearean verse. Since these authori-
tative recorded performances, however, the decline in the use of RP English
in the electronic media and the tendency of flm directors and illustrious
contemporary actors, such as Jeremy Irons,
8
to blur the distinction between
Shakespearean verse and prose, is palpable. Perhaps there is a preconception
on the part of certain directors and performers concerning the reception by
contemporary audiences of Shakespearean pentameter. Perhaps also, fash-
ionable recent adaptations of Shakespeare and Jane Austen, inter alios, into
contemporary linguistic idiom inhibits any mass-audience Shakespeare per-
formance that would be perceived as adhering to standard, old-fashioned
verse delivery style.
All of the above critical debates concerning the complexities of the
iambic pentameter as formal unit inevitably have a strong bearing on
Shakespeare in performance. Where the iambic rhythm and the phrasal
stress pattern of normal speech coincide, the trained actor will have no
diffculty achieving a sense of rhythm in performance. What is more com-
plicated, however, is the increasingly frequent use of disparities between
metrical rhythm and phrasal or contrastive stress, quite apart from rhe-
torical emphasis in Shakespeares dramatic verse. Research into the deliv-
ery techniques of the actors in the BBC sample suggests an idiosyncratic
approach to metre and rhetorical emphasis, with certain actors adhering
largely to an underlying iambic pattern in the modern tradition of Olivier,
while others prefer a more prose-like enunciation of lines with very lim-
ited observance of line-pauses and verse cadences or, indeed, metrical
rhythm.
Our fndings above reveal that in the earlier use of iambic rhythm,
Shakespeares syntactic predilection for subject-verb inversion resulted
in many instances of a non-contrastive subject pronoun occurring in
even syllable position. In his later works, he produced verse that tended
to avoid subject-verb inversion with minimal additions to the tally of non-
contrastive subject pronouns. The present study set out to determine to
what extent sensitivity to these factors was in evidence in the performance
of leading Shakespearean actors of recent decades, whose interpretations
were sampled from the series of BBC productions available in digital video
format. Five plays were chosen from the BBC series, namely Richard III,
Richard II, Julius Caesar, Troilus and Cressida and Cymbeline, so as to repre-
sent a broad span of the plays analysed in section 4 above. The actors
included Ron Cook (Richard III), Zoe Wanamaker (Lady Anne), Jon Finch
(Bolingbroke), John Gielgud (John of Gaunt), Richard Pasco (Brutus), Sam
Dastor (Casca), Anton Lesser (Troilus), Suzanne Burden (Cressida), Helen
Mirren (Imogen), Robert Lindsay (Iachimo) and Michael Pennington
(Posthumus). Those who broadly adopted the phonological stress rules
of natural speech in order, presumably, to render their dialogue more
modern in feeling included Michael Byrne (Buckingham in Richard III),
Rowena Cooper (Queen Elizabeth in Richard III), Charles Gray (Duke of
York in Richard II, Julius Caesar in Julius Caesar and Pandarus in Troilus and
Cressida), Vernon Dobtcheff (Agamemnon in Troilus and Cressida), Keith
Michell (Antony in Julius Caesar) and Richard Johnson (Cymbeline).
An analysis of their stress placement in cases where Spro was positioned
on an even syllable showed that in the majority of cases in the BBC samples
the pronoun receives light but distinct stress. There is a high correlation
between actors who observe metrical rhythm more closely throughout their
performance as a whole and the discernible enunciation of stress on Spro
placed as even syllables in these specifc cases. Conversely, those actors who
are inclined to play more and more astride the verse instead of in it
9
,
to cite Granville-Barkers admonition to the young Gielgud, and treat the
verse as if it were prose, tend to ignore the question of subject pronoun
stress entirely unless it is strictly contrastive. Examples of both types are
given below (our own emphasis shown in bold type):
Pronoun in even position enunciated with stress (VS):
(7) A cockatrice hast thou hatched to the world, Richard III 4.1.54
(Annette Crosbie: Duchess of York)
(8) Again shall you be mother to a king, Richard III 4.4.317
(Ron Cook: Richard III)
(9) Come I appellant to this princely presence. Richard II 1.1.34
(Jon Finch: Bolingbroke)
(10) Hast thou tapped out and drunkenly caroused. Richard II 2.1.128
(John Gielgud: John of Gaunt)
(11) I wish we may. But yet have I a mind Julius Caesar 3.1.1456
That fears him much; . . . (David Collings: Caius Cassius)
Pronoun in even position enunciated without stress (VS and SV):
(12) More bitterly could I expostulate, Richard III 3.7.182
(Michael Byrne: Buckingham)
(13) To Bolingbroke are we sworn subjects now, Richard II 5.2.39
(Charles Gray: Duke of York)
(14) From your own mouth, my lord, did I this deed. Richard II 5.6.37
(Desmond Adams: Sir Piers of Exton)
(15) For certain she is dead, and by strange manner. Julius Caesar 4.2.241
(Brian Coburn: Messala)
A variable approach to delivery of the pentameter is evident in all fve of
the productions assessed. To some extent speech style and delivery probably
depended on directorial preference. Jonathan Millers Troilus and Cressida
deploys a prose-like delivery for much of the play, with little discernible
attempt at suggesting an underlying iambic rhythm, apart from in the
romantic exchanges of the lovers scenes. Nevertheless in all fve produc-
tions it was clear that some actors were palpably more rhythmically faithful
in their spoken verse technique. However, all of the actors studied (both
the iambic rhythm-oriented and the non-iambic rhythm-oriented) showed
awareness of verse scansion in telling instances. For example, the past tense
and adjectival -ed endings which are non-syllabic in modern English are fre-
quently sounded as a separate syllable in Shakespearean verse, the extra syl-
lable thereby contributing to the syllable count of a particular line. Unlike
a number of non-Shakespearean specialists in flm adaptations, the BBC
series actors fawlessly observe the metrically required articulation of -ed, or
non-articulation of -d.
Probably the most interesting case in the comparison between iambic-
oriented and non-iambic oriented actors was Derek Jacobi, who adopted a
deeply subtle approach to metre in his interpretation of the role of Richard II.
In the frst two acts of the play the capricious nature of the Kings character is
conveyed by his capricious, almost throw-away treatment of the verse with the
notable exception of a more ceremonial use of regular rhythm in his decrees
of banishment on Bolingbroke and Mowbray. As Richards dignity and power
are gradually stripped away in Acts 3 and 4, he exhibits a serious, contempla-
tive and sorrowful attitude with more measured tones, which is refected in
his greater observation of iambic regularity. In Richards Act 5 pre-death solil-
oquy Jacobi reverts at one point to a more syncopated approach to rhythm to
express his characters agitated state of mind, but what precedes and follows
is in many places more conventionally iambic. Below are four examples from
different stages of the play with stress/emphatic accent bolded:
(16) We will ourself in person to this war,
And for our coffers with too great a court
And liberal largess are grown somewhat light,
We are enforced to farm our royal realm,
The revenue whereof shall furnish us
For our affairs in hand . . . . Richard II 1.4.416 (Jacobi)
Prosodic features in Jacobis delivery observe the stress and tempo of natural
speech rhythm with additional optional rhetorical emphasis. Iambic rhythm
is not evident, although the rising and falling intonation, very distinct diction
and alliterative emphasis all confer a poetic quality on the delivery. The actors
choice in respect of pace and pause accentuates this non-iambic attribute
10
:
(17) My crown I am, but still my griefs are mine.
You may my glories and my state depose,
But not my griefs; still am I king of those. Richard II 4.1.1813 (Jacobi)
The trochaic interpretation of the third foot causes am to be unstressed,
but then the subject pronoun I receives stress. In this more ceremonial setting
Jacobi deploys iambic rhythm for the frst two lines, but, his character still torn
between dignifed resignation and bitter sorrow at his usurpation, he departs
from iambic rhythm again in the second half of the third line. After which, he
intones what amounts to a formal abdication speech in regular pentameter:
(18) Now mark me how I will undo myself.
I give this heavy weight from off my head,
And this unwieldy sceptre from my hand,
The pride of kingly sway from out my heart.
Richard II 4.1.1936 (Jacobi)
(19) Music do I hear.
O, Oh
11
keep time! How sour sweet music is
When time is broke and no proportion kept.
So is it in the music of mens lives.
And here have I the daintiness of ear
To check time broke in a disordered string;
But for the concord of my state and time,
Had not an ear to hear my true time broke.
I wasted time, and now doth time waste me,
Richard II 5.5.419 (Jacobi)
Here, and in particular in line 45 here have I there is recourse to iambic
rhythm, though not in all lines, as if to underline metrically the semantic
content, in particular the reference to broken time and the counterpoint
with the strict musical tempo of the lute which is discordant to Richards
thought processes. Clearly, Jacobis choices of when to employ iambic
rhythm and when to accentuate specifc syllables for rhetorical purposes
are far from arbitrary. His interpretation of the lines sets the character
apart as he intends for legitimate dramatic reasons, resulting in a defnitive
and critically acclaimed performance.
To add a further example, this time from the 1981 BBC version of
Richard III, the interchange between Richard III and Buckingham in
Richard III Act 3, scene 7 contrasts an iambically regular delivery by Ron
Cook (Richard) with a prose-like delivery from Michael Byrne (Buckingham).
The exchange between Cook and Rowena Cooper (Queen Elizabeth) in
Act 4, scene 4 sets up a powerful contrast between the smooth-tongued hyp-
ocrite Richard (a deceitful user of language to refer to Wright 1983: 156)
and the less intelligent and ultimately outmanoeuvred Duke and Queen
respectively. Cook rarely deviates from the predominantly iambic rhythm,
but again delivers a performance of calculated brilliance. He prefers to ride
on the verse to great effect, achieving a delivery of which Granville-Barker
would, presumably, have approved. It can be seen from these instances
that in Cooks personal interpretation of Shakespearean prosody, subject
pronouns can accordingly be stressed in even (ictic) position in the line,
whether or not they function in a contrastive sense.
6. Conclusions
The fndings of our linguistic analysis have already been set out in section 4.
They indicate that the option of inverting a subject pronoun around the
verb in Shakespeares dramatic verse had nothing to do with respecting
iambic pentameter. Rather it was a freestanding stylistic variant used by the
poet in order to achieve a high style effect, possibly even a monumental
quality. The later plays show a decline in the frequency with which this var-
iant is deployed. They display no tendency to favour or avoid placement of
a pronoun on an odd or even syllable, depending on whether inversion or
non-inversion of the subject pronoun was adopted. Unexpectedly, an early
period has been identifed of tension between versifcation and syntax, that
of the earlier plays, up to King John. In these, performers will often have
to disregard iambic rhythm if they wish to avoid stressing a subject pro-
noun contrary to sense, that is, where no emphatic or contrasted meaning
is intended. The performer seeking to avoid this unwanted effect will be
drawn towards a speech rhythm delivery in which the effect of dramatic
verse will be diminished, whereas the performance of later plays will very
rarely encounter mismatches between speech rhythm and iambic rhythm
occasioned by VSpro ictic
12
. This fnding goes against a strict interpretation
of the position taken by Wright (1988: notably 2203), to the effect that it
was only in Shakespeares late period that he achieved independence of
metre and speech rhythm. As regards the syntax of inversion, disruption of
iambic rhythm is not a distinctive hallmark of the late plays, but rather of
the earlier ones.
It would be reasonable to assume, based on the work of Wright and
others, that the structural, metrical and characterizational complexi-
ties of Shakespeares later plays would always pose more problems of
stress placement for performance practitioners. Paradoxically, the fnd-
ings of our study suggest that the high incidence in the earlier plays of
inverted subject pronouns in even syllable positions may induce some
actors to adopt greater natural speech rhythm strategies, contrasting
markedly with those actors whose preference is for more traditional iam-
bic rhythms.
More generally, it is to be noted that, although this research was under-
taken with the intention of addressing controversies on the status of iam-
bic rhythm in Shakespeares dramatic verse, our fndings have uncovered
further support for the position that he used it as an underlying default
scheme. There is clear evidence of his preference for placing pronouns
in odd syllables, as well as corroboration of earlier arguments for the pres-
ence of iambic rhythm on the basis of his alignment of iambic rhythm and
word stress in the case of polysyllabic verbs. Together, these fndings con-
stitute strong support for the view that Shakespeare skilfully managed the
relationship between syntactic position, rhythm and metre, and in distinct
ways, at successive stages of his play-writing career.
Notes
1
We gratefully acknowledge the generous support of the British Academys Sino-
British fund in the preparation of this research.
2
In fact, inversion of a subject pronoun, as in (1)b, was not generally the rule in
Old English and Early Middle English. It may plausibly be attributed to the infu-
ence of Anglo-Norman (Haeberli 2010).
3
Baekken (1998), using the Helsinki corpus, shows that the terminal decline of
inversion post-dates 1630.
4
That is, on an even-numbered syllable, cf. Tarlinskaja (1983).
5
We followed this approach on the basis of the frst authors earlier work on inver-
sion in Late Middle English chronicles (Ingham and Grohman 2008), in which
the presence of a narrative timeline favours the clause-initial placement of non-
subject constituents, notably temporal expressions.
6
We repeat the point that inversion was not yet an entirely archaic trait in later
sixteenth-century English, and not confned to verse.
7
A reviewer suggests that using an oscilloscope for more scientifc measurement
would give a set of uncategorized gradient-based readings. To categorize the
readings into different levels of stress is a task that the human ear and brain
can do, but the oscilloscope cannot, especially if the aim is to categorize the
readings according to metrical or rhythmic criteria. Objective acoustic measure-
ment methods, in addition to the present subjective ones, would be useful if
the present study targeted purely the performance pragmatics of Shakespearean
dramatic verse, but this is not our aim.
8
Irons interpretation of the role of Antonio in Michael Hoffmans 2004 flm of
The Merchant of Venice exemplifes the self-conscious avoidance of iambic stress
patterns in contemporary performance and the tendency to simplify the verse by
turning it into what is effectively prose.
9
H. Granville-Barker, letter to John Gielgud, cited in The New Penguin Shake-
speare, Richard II, p. 9.
10
The shift in rhythm and stress patterning on the second hemistich of the third line of
this extract can be said to constitute a choriambic cluster or segment, as an anony-
mous reviewer has pointed out; that is a mixture of trochaic and iambic rhythms with
an intervening spondee at least in Jacobis interpretation referred to here.
11
Most versions of the text have Ha, ha! but Jacobis performance employs the
variant given here.
12
In fact the 13 examples of VSpro ictic in the Period III plays studied all occur in
the frst foot of the line, where Shakespeare often uses a trochee anyway, so they
will not necessarily be perceived as disrupting the metre.
References
William Shakespeare: The Complete Works, Stanley Wells and Gary Taylor,
with John Jowett and William Montgomery (eds) 2005 [1986]. Oxford:
Clarendon Press.
Holinshed (Richard III): www.r3.org/bookcase/holinshed/h-408.htm
(Macbeth): www.clicknotes.com/macbeth/Holinshed/
The Shakespeare Collection BBC Shakespeare series (19781985) on
DVD.
www.en.wikipedia.org/wiki/BBC Television Shakespeare (for online
reference)
Wallace, David Accents, SHK 13.1285 posted 9th May 2002: Shaksper: The
Global Electronic Shakespeare Conference. Accessed 12 December 2007
www.shaksper.net/archives/2002/1252.html
Abbott, Edwin A. 1870. A Shakespearian Grammar An Attempt to Illustrate some of
the Differences between Elizabethan and Modern English. New York: Dover Editions.
Re-issued 2003. Originally published by Macmillan: London.
Baekken, Bjrg 1998. Word Order Patterns in Early Modern English. Oslo: Novus
Press.
Baekken, Bjrg 2000. Inversion in Early Modern English, English Studies 81(5):
393421.
Culpeper, Jonathan 2001. Language and Characterisation: People in Plays and Other
Texts. Harlow: Pearson.
Halle, Morris and Samuel Jay Keyser 1971. Illustration and defence of a theory of
the iambic pentameter, College English 33(2): 15476.
Hope, Jonathan 1994. The Authorship of Shakespeares Plays: A Sociolinguistic Study.
Cambridge: Cambridge University Press.
Houston, John Porter 1988. Shakespearean Sentences: A Study in Style and Syntax. Baton
Rouge: Louisiana State University Press.
Tarlinskaja, Marina 1983. Evolution of Shakespeares metrical style, Poetics: Journal
of Empirical Research on Literature, the Media and the Arts 12.6: 56787.
Wells, Stanley 1969. Introduction, Stanley Wells (ed.) Richard II. London:
Penguin.
Wright, George T. 1983. The play of phrase and line in Shakespeares iambic pentam-
eter, Shakespeare Quarterly 34(2), Folger Shakespeare Library, 14858.
Wright, George T. 1988. Shakespeares Metrical Art. Berkeley, Los Angeles and London:
University of California Press.
Further reading
Groves, Peter L. 1998. Strange Music: The Metre of the English Heroic Line. ELS
Monograph Series 74. Victoria, British Columbia: University of Victoria.
Haeberli, Eric 2010. Investigating Anglo-Norman infuence on Late middle English
syntax, in: Richard Ingham (ed.), The Anglo-Norman Language and its Contexts.
Woodbridge: Boydell & Brewer, 14363.
Ingham, Richard and Kleanthes K. Grohman 2008. On the post-fnite misagree-
ment phenomenon in Late Middle English, in: Marina Dossena, Richard Dury,
and Maurizio Gotti (eds), English Historical Linguistics 2006 (ICEHL) 14. Volume
I: Syntax and Morphology (Bergamo, 2125 August 2006). Amsterdam: John
Benjamins Publishing, 12540.
Chapter 6
Shakespeares Short Pentameters and the
Rhythms of Dramatic Verse
Peter Groves
1. The short pentameter
Just before the much anticipated joust in Richard II, the Marshal abruptly
calls a halt to the proceedings: Stay, the King hath throwne his Warder
downe (1.3.118). Because this line has only nine syllables it has worried
many of Shakespeares editors, who have generally assumed that metri-
cality in the pentameter requires a standard or at any rate a minimum
number of ten syllables, and who for three centuries have been conscien-
tiously tidying up his versifcation, anxious to defend him against the tra-
ditional charges of artlessness and lawlessness.
1
Eighteenth-century editors
simply added words ad libitum to fll in the apparent gaps: they repaired
this line, for example, by arbitrarily inserting a But before Stay. Later
editors have usually been more cautious, and have tried to remove appar-
ent gaps by relineation where possible. Where it is not, a note of desper-
ation sometimes creeps in: the Arden 2 editor of Richard II, for example,
suggests pronouncing the Q1 spelling throwen to regularize the metre of
the line (Ure 1961: 28n) thus creating a highly irregular sequence of three
successive reversed feet (a reversed foot traditionally a trochee is one
where the beat or ictus precedes rather than follows the off-beat or non-
ictus). More recent editors have tended to pay less attention to the metre,
but have (oddly) retained many previous relineations designed solely to
remove apparent gaps (see Groves 2007).
What I wish to suggest in this chapter is that many examples of syllabic def-
icit in the early texts of Shakespeare represent not compositorial botching,
metrical licence or authorial inadvertence, but rather choices within the
signifying system of the metre, a way of pointing performance for actors
(and readers) through the use of what I term short pentameters, or lines
that have the requisite fve beats but fewer than ten syllables because they
include silent off-beats or silent beats.
2
Take the Marshals headless line: the
surprise of the silent initial off-beat performatively a little like treading
in the dark on a step that isnt there arises from our perception that the
frst position should (by default) be an off-beat. We experience the same
surprise in the case of an initial reversal or trochee: with the headless line,
however, the surprise is prolonged as the metre fails to re-stabilize itself with
the familiar POM-titty-POM or O for a Muse opening. This silent off-beat
draws attention to Richards coup de thtre and mimetically it conveys the
abruptness of the Marshals injunction.
I am not, of course, the frst to suggest that Shakespeares metrical gaps
or lacunae might represent a rhythmical device rather than a lapse or a
licence see, for example, Sicherman (1982, 1984) and Wright (1988) but
there has been little attempt to theorize the phenomena in terms either of
English phonology or of a coherent theory of metricality. For Sicherman,
for example, they simply represent metrical pauses (1982: 175) which
may occur in succession, so that Hes tane becomes a pentameter when
followed by eight metrical pauses (1984: 185), as does No, my Lord when
preceded by seven (1984: 193). In the interests of theoretical rigour I
shall use a system of phonologically based prosodic and metrical analysis
which I call Base and Template scansion, which can distinguish metrical
and unmetrical lines; it is set forth fully in Groves (1998), but I will need
to sketch in some salient features here. It begins from the fact that metre
is an organization of speech, and that English speech is itself structured
(and, in part, timed) around periodic recurrences of articulatory inten-
sity called beats: they are not directly features of sound but rather of
the muscular organization of articulation, sympathetically perceived by the
listener, and distinct both from lexical stress, the inherent prominence
that obtains in at least one syllable of every lexical or dictionary word,
and accent, the pitch-slide that marks contrast in speech (ON the fridge,
not IN the fridge!). Those who accompany their speech with gestures, for
example, will tend to time those gestures on the beats. Beats occur most
frequently on lexically stressed syllables, but not on all: we say /three blind
/mice, not /three /blind /mice (unless we want to emphasize both the num-
ber and the blindness). They can, moreover, fall upon unstressed syllables
to break up a long run of them, and (since they are not a form of sound,
but of muscular activity, a means of organizing utterances in time), they
may (unlike stress and accent) fall (and be perceived) upon silence, like
a rest in music. The silent beat, represented as <!>, distinguishes (for
example) between the restrictive and non-restrictive relative clauses in a
pair like the following: I /spoke to the /waiter who /brought my /soup
Shakespeares Short Pentameters 121
(bringing precedes speaking); I /spoke to the /waiter, /<!> who /brought
my /soup (bringing follows speaking).
3
As Derek Attridge (1982) has sug-
gested, silent beats are a typical structural feature of chantable English
demotic verse nursery rhymes, protest chants and so on where they
represent a beat-long pause, a rest, in the chant. It is easy to clap on the
eight beats of an appropriate performance of the following, for example,
including on the fnal rest:
/What do we /want? /Ten per cent!
/When do we /want it? /Now! /<!>
A once well-known football chant goes:
/Georgie /Best! /Georgie /Best!
/Georgie! / <!> /Georgie /Best!
Metrical structure in English is produced through the placement of beats
in the spoken line, and this is enabled and limited by three prosodic phe-
nomena: by the disposition of lexical and syntactic stress, by the location
of syntactic junctures, and by the speakers (contextually motivated) plac-
ing of pragmatic accent (used for pointing contrast or highlighting infor-
mation) within the utterance. To simplify somewhat the argument of an
entire book, an iambic pentameter is an utterance or part-utterance that
can be accommodated (by rule-governed elisions where necessary) to fve
feet (pairs of syllable-positions), each of which contains at least one sylla-
ble capable of carrying a beat, normally the second syllable of the foot. A
syllable is prevented from carrying a beat when it is dominated by a fully-
stressed neighbour within the same intonational phrase; a syllable not so
dominated may be called independent; in my apartment, for example,
fully stressed (and thus independent) -part- dominates both a- and -ment
but leaves unstressed my independent. An independent syllable is thus one
capable of carrying a beat, and it must be either (a) a fully stressed syllable,
(b) an accented syllable, or (c) an unstressed syllable that is not dominated
by a neighbouring stressed syllable (i.e. it is protected by an intervening
syntactic break or liberated by accent). Where the independent occurs on
the frst syllable we say the foot is reversed, and enclose it in angle brackets:
<Wanton> as youth|ful goats|, <wild as> young bulls| (1 Henry IV 4.1.103);
each reversal must be followed by a normal foot. Syncopated syllables are
printed in superscript to suggest that they should be pronounced quickly
and lightly: <Myr
i
ads> of ri|v
u
lets hu|rr
y
ing through| the lawn|.
If (to illustrate) I enclose dominated syllables in brackets and bold all
independent ones, the following line can be shown to have fve normal feet
(or iambs), because by, though unstressed, is not dominated: (I n|der)
stnd| (your ma|ning) by| (your ye|). As George Gascoigne, the earliest
English metrist, points out, this verse may passe the musters metrically,
whereas the following re-arrangement unmetrical, or neyther true nor
pleasant (1575: 295); this is because the second pair of syllables has no
independent and cannot, therefore, function as a foot: (Your maning)
(I nder)(stnd by) (your ye).
The phonological conditions for silent off-beats and silent beats may be
simply stated: a silent off-beat can only occur either initially or in a position
that is fanked by fully independent syllables, and a silent beat can only
occur at the site of an obligatory intonational break (normally marked by
terminal punctuation). The adjacency of beats that in metre may generate
the perception of a silent off-beat is something that English generally tries
to avoid (three successive stresses in the same phrase will typically produce
only two beats (/three blind /mice). Adjacent beats will be produced in one
of two ways: by the occurrence of an intonational phrase-break between
two stressed syllables (/Run, /Spot!) or by the contiguity of pragmatically
accented syllables within a phrase (Not /white /coffee but /green /tea), and
each of these produces a distinct kind of silent off-beat, which I term respec-
tively the jolt (^) and the drag (~).
2. The jolt
A silent off-beat that is mapped onto a potential or obligatory intona-
tional break will accentuate that break (where in general regular
metre by its insistent continuity tends to bridge and smooth over such
gaps), and this is why I call it a jolt. It can have the expressive effect of
suggesting urgency and spontaneity, most commonly at the beginning of
the line, where it often underscores abrupt attention-seeking imperatives
and vocatives:
^ Goe|, take hence| that Tray|tor from| our sight|, (2H6 2.3.100)
^ Come| my Lord|, Ile leade| you to| your Tent|. (1H4 5.4.9)
^ Proue| it Hen|ry, and| thou shalt| be King. (3H6 1.1.131)
^ Gen|tlemen|, impor|tune me| no far|{ther}, (SHR 1.1.48)
^ Iay|lor, take| him to| thy cus|todie. (ERR 1.1.155)
^ Ti|tus, I| am come| to talke| with thee|, (TIT 5.2.16)
Such headless lines constitute the most common kind of short pentame-
ter, and do not represent an innovation on Shakespeares part. He probably
learned the trick of buttonholing vocative headlessness from Marlowe (105):
^ Bar|barous| and bloo|dy Tam|burlaine |, (1TAM 2.7.1)
^ Trea|cherous| and false| Theri |damas |, (1TAM 2.7.3)
^ Bloo|dy and| insa|tiate Tam | burlain|. (1TAM 2.7.11)
Shakespeares innovation consists in extending the possibility of lacuna to
every position in the line (though there are precedents for this in Wyatt see
Groves (2005) which Shakespeare may have been familiar with). A jolt, for
example, may draw attention to imperatives and vocatives within the line:
<Vnder> my Batt|lements|. ^ Come| you Spi|{rits}, (MAC 1.5.40)
Good Mar|garet|[,] ^ runne| thee to| the par|{lour}, (ADO 3.1.1)
<As may| compact| it more|. ^ Get| you gone|, (LR 1.4.339)
But roome|[,] ^ Fai|ry, heere| comes O|beron|. (MND 2.1.58)
That what| he feard|, is chancd|. Yet speake|[,] ^ (Mor|{ton}),
(2H4 1.1.87)
^ Nurse|, ^ wife|, what ho|? what Nurse| I say|? (ROM 4.4.24)
It can also mark a turning to a new addressee:
cornwall: [To Gloucester] . . . make your owne purpose,
<How in> my strength| you please|: for you| ^ Ed|{mund},
(LR 2.1.1112)
brutus: <Luc
i
us> my Gowne|: ^ fare|well good| Messa|{la},
(JC 4.3.231)
malcolm: [To Duncan] This is the Seriant,
Who like a good and hardie Souldier fought
<Gainst my> Capti|uitie|: ^ Haile| braue friend|;
(MAC 1.2.35)
The jolt can also work mimetically, to represent interrupted speech: it can,
for example, indicate a pause or hesitation. In the following examples it rep-
resents the thinking on the fy or er-um pause: Gonzalo, inventing a Utopia
on the spur of the moment, hesitates a couple of times in deciding on its
elements:
<Letters> should not| be knowne|: ^ Ri|ches, po|{uerty},
And vse| of ser|uice, none|: <Contract>, Succe|{ssion},
Borne, bound| of Land|, ^ Tilth|, ^ Vine|yard none|:
(TMP 2.1.1513)
Northumberland, rebuked by York for referring to the threatened king
as mere Richard, hesitates before arriving at an insolently lame excuse:
Your Grace| mistakes|: ^ one|ly to| be briefe|,
Left I his Title out. (R2 3.3.10)
Between two speakers the jolt may represent the hesitation of disagree-
ment, anything from Gloucesters polite demurral to Leontes outright
contradiction:
gloucester: Me thinkes the ground is eeuen.
edgar: Horrible steepe.
<Hearke, do> you heare| the Sea|?
gloucester: ^ No| ^ tru|{ly}.
(LR 4.6.34)
brutus: For he can do no more then Csars Arme,
When C|sars head| is off|.
cassius: ^ Yet| I feare |{him},
(JC 2.1.1823)
hermione: <You did|
4
mistake|.
leontes: ^ No|: <if I| mistake| (WT 2.1.100)
There are, of course, other kinds of hesitation: Messalas tactful reluc-
tance to confrm bad news, for example, or the Nurses cautious awareness
that her comfort is unlikely to be very welcome:
cassius: <Cice>ro one|?
messala: ^ Ci|cero| is dead|, (JC 4.3.179)
juliet: Some com|fort Nurse|.
nurse: ^ Faith| ^ here| it is|,
. . . I thinke it best you married with the Countie,
(ROM 3.5.2123/17)
Such interlocutory jolts may represent a kind of double take, as when
Hamlet (in F1) responds with incredulity to Gertrudes attempt to dismiss his
criticisms as madness, Bolingbroke is surprised to fnd King Richard in Flint
castle, or Caesars centurion is amazed to come across Enobarbus in abject
soliloquy (the arrow indicates that the printed line does not break in F1):
gertrude: This bodilesse Creation extasie
is ve|ry cu|nning in|.
hamlet: ^ Ex|tasie|? (HAM. 3.4.1389)
percy: The Castle royally is mannd, my Lord,
Against| thy en|t[e]rance|.
5
bolingbroke: ^ Roy|ally|?
Why, it containes no King?
percy: Yes (my good Lord)
(R2 3.3.23)
enobarbus: . . . (O thou blessed Moone) . . . poore Enobarbus
did Before| thy face| repent|.
centurion: ^ E|nobar|{bus}?
(ANT 4.9.7/910)
Sometimes the medial jolt represents a catch in the voice at a moment of
high emotion:
duncan: My plent
e
ous Ioyes,
Wanton in fullnesse, seeke to hide themselues
In drop|s of so|rrow. Sonnes|, ^ Kins|men, Thanes|,
(MAC 1.4.335)
helena: Is all the counsell that we two have shard, . . .
When wee haue chid the hasty footed time,
For par|ting vs|; ^ O|, is all| forgot|? (MND 3.2.201)
antony: . . . You did know
My Sword, made weake by my affection, would
Obey| it on| all cause|.
cleopatra: ^ Par|don, par|{don}.
(ANT 3.11.65/678)
Alternatively (it depends, of course, on the context of meaning), it can
evoke a sudden start: Achilles or Capulets anger, or the feigned exaspera-
tion of Petruchio:
ulysses: Tis knowne Achilles, that you are in loue
With one| of Pri|ams daugh|ters.
achilles: Ha|? ^ knowne|?
(TRO 3.3.1934)
nurse: <May not| one speake|?
capulet: ^ Peace| you mum|bling foole|,
(ROM 3.5.173)
petruchio: Whats this|, ^ Mu|tton?
first servant: I|.
petruchio: Who brought| it?
peter: I|.
(SHR 4.1.160)
Successive jolts are usually mimetic, for example of anger:
lear: Blowe windes|, & crack| your cheeks|; ^ Rage|, ^ blow
(LR 3.2.1)
solanio: <Why then| you are| in loue|.
antonio: ^ Fie|, ^ fe|. (MV 1.1.46)
timon: ^ Gold|? ^ Ye|llow, gli|tt
e
ring, pre|cious Gold|? (TIM 4.3.26)
Of jumpiness:
lady macbeth: ^ Hearke|, ^ peace|: <it was| the Owle| that shriekd|,
6
(MAC 2.2.3)
horatio: ^ Stay|: ^ speake|; ^ speake|: I Charge| thee, speake|.
(HAM 1.1.51)
Of surprise:
horatio: My Lord, I thinke I saw him yesternight.
hamlet: ^ Saw|? ^ Who|?
horatio: My Lord|, the King| your Fa|{ther}.
(HAM 1.2.18990)
Or of panic:
edmund: . . . my Writ / Is on the life of Lear, and on Cordelia:
Nay, send| in time|.
albany: ^ Run|, ^ run|, O, run|. (LR 5.3.246/49)
3. The drag
A drag, by contrast, is a silent off-beat between two syllables in the same
phrase where there is no potential intonational break, as between an adjec-
tive and its noun. Since within a phrase one of two adjacent stressed syl-
lables will normally be subordinated to the other, and since a silent off-beat
requires fully independent syllables on either side, the result is to force
the actor to accent the subordinated syllable, in order to promote it to
full independence. It is a method, in other words, of pointing the text for
performance. Take the following line: Not i th worst ranke of Manhood,
sayt, (Macbeth 3.1.102). One 19th-century editor says of it a syllable is
wanting and proposes most worst; others read worser (Furness 1873:
144). We fnd the pointing of the line, however, by negotiating between our
(tacit) knowledges of the metre and of the language: the only way to pro-
duce the requisite fve beats within the constraints of metre and phonology
is to place an emphatic accent on worst, which not only allows the neces-
sary silent off-beat but at the same time encourages (because of the rough
periodicity of beats) a suitably sarcastic dwelling on the word, as Macbeth
exhorts them Now, if you haue a station in the fle,
<Not i> th[e] worst| ~ ranke| of Man|hood, sayt|,
In prose speech, of course, accenting only worst would simply attract the
beat away from ranke; to speak the line as verse we must accent both to
ensure a beat on both. To take another occurrence: Aegeon, having spoken
of his own wifes giving birth, is instructed by the metre to put an appropri-
ate contrastive accent on mean by the drag in the second line:
That very howre, and in the selfe-same Inne,
A meane| ~ wo|man was| deli|uerd| (ERR 1.1.534)
Nevertheless the word is emended (without textual support) to mean-born by
Wells and Taylor (1986), and some modern editions print meaner. In the fol-
lowing post-mortem on the Battle of Actium, in which the two dragged pronouns
must be given contrastive accent to save the metre (what though you [a woman,
unused to battle] fed? Why should he [a hardened soldier] follow?), earlier
editors would routinely emend though to although in 4, but leave the structurally
similar line 6 alone, presumably because it already has ten syllables:
cleopatra: Is Anthony, or we in fault for this?
enobarbus: Anthony onely, that would make his will
<Lord of> his Rea|son. What| though you| ~ fed|,
From that great face of Warre, whose seu
e
rall ranges
<Frighted> each o|ther? Why| should he| ~ fo|{llow}?
(ANT 3.13.26)
4. The silent beat
A silent off-beat is only a presence and an absence in metrical terms: it has
no existence for someone not experiencing the performance as being gov-
erned by a metre (phonologically it is no more than a successive occurrence
of beats). By contrast a silent beat (represented in what follows as <!>) is a
distinct articulatory and perceptual phenomenon, a rest. In demotic verse,
the predictability of the silent beat both in structure (occurring, for example,
at every eighth beat in the ballad stanza) and in performance (due to the
rather regularly-timed sing-song performance) permits it to be adequately
signalled by a silence. Neither of these regularities obtains in sophisticated
forms like pentameter, however, and so the natural assumption is that
Shakespeare makes no use of the silent beat. Yet Fredson Bowers has drawn
attention to a couple of instances in the operatic love-duet at the beginning
of Act 5 of The Merchant of Venice : there are six mid-line changes of speaker
in the passage, and in each case the second speaker begins with the two-foot
refrain In such a night. At Lorenzos second serve he tries to put Jessica
off her stride by providing a feminine caesura, but she lobs it right back at
him; at his third he withholds the fnal beat of his part-line, but undismayed
she supplies it and returns a similarly truncated part-line to him. As Bowers
says, the symmetry . . . can only be designed (Bowers 1980: 99) two normal
transitions, two excessive ones, and then two defective ones showing that
Shakespeare clearly intended silent beats in this passage:
lorenzo 1: Where Cre|ssed lay| that night|. / In such| a night|
jessica 1: And ranne| dismayed| away|. / In such| a night|
lorenzo 2: To come| againe| to Car|{thage}. / In such| a night|
jessica 2: <That did| renew| old E|{son}. / In such| a night|
lorenzo 3: As farre| as Bel|mont. / <!>| In such| a night|
jessica 3: And nere| a true| one. / <!>| In such| a night|
(MV 6/9/12/14/17/20)
But how do Jessica and Lorenzo signal these silent beats? A mere pause
silence itself will not do the job in pentameter verse, as I have pointed out.
But since beats represent muscular activity, peaks of effort in articulation,
unexpected silent beats may be signalled by and thus in a sense cue some
form of gesture (perhaps in this case a playful admonitory tap from Jessica
and a patronizing kiss from Lorenzo).
Occasionally the beating of the verse is played out more violently on
the human body, as in the case of the messenger both unlucky enough to
bring the news of Antonys marriage and reckless enough to recommend
pa-ti-ence to Cleopatra (the stage direction is in F1):
messenger: Good Ma|dam pa|tence|.
cleopatra: What say| you? <!>| Strikes him
(ANT 2.5.63)
Less violently, a beat may be signalled by a touch: the two following are
marked by a touch on the shoulder, from the kings sword or the bailiffs
hand:
king john: <Kneele thou> downe Phi|lip, <!>| but rise| more great|,
Arise Sir Richard, and Plantagenet. (JN 1.1.1612)
antonio: <But be| of com|fort.
second officer: <!>| Come sir| away|. (TN 3.4.338)
A piece of stage-business may be cued by a rest: Bassanios opening of the
fateful casket or the fourishing of a prop:
portia: I feele too much thy blessing, make it lesse,
For feare| I sur|feit.
bassanio: <!>| What fnde| I here|? (MV 3.2.1134)
sicinius: Haue you a Catalogue
Of all the Voices that we haue procurd,
<set downe| byth Pole|?
aedile: I haue|:tis rea|dy <!>|.
(COR. 3.3.10)
jeweller: And rich|: <heere is> a Wa|ter looke| ye <!>|.
(TIM 1.1.18)
The rest may cue a gestural pointing to or indication of some object or
person:
england: And there|upon| <giue me> your Daugh|ter <!>|.
(H5 5.2.347)
lear: . . . Death on my state: wherefore
<Should he| sit he|ere <!>|? This act| perswades| {me},
(LR 2.3.1123)
hamlet: My fa|ther, <!>| me thinkes| I see| my fa|{ther}.
(HAM 1.2.184)
Horatios alarm (Oh where, my Lord?) makes more sense if the rest in
Hamlets line cues a gesturing, pointing or turning towards the imaginary
father.
A second function of the silent beat is to cue an expressive bodily ges-
ture that represents a characters response or emotion. Take Polonius
youre-a-man-of-the-world, work-it-out-for-yourself shrug as he ekes out
his list of suggested slips on Laertes part to Reynaldo, who has ten-
tatively suggested gaming:
polonius: ^ I|, or drink|ing, fenc|ing, swear|ing, <!>|
<Quarrel>ling, dra|bbing. You| may goe| so farre|.
(HAM. 2.1.25)
It can equally cue a mock-shrug: King Henrys sarcastically simulated baf-
fement (this silent beat was frst pointed out by George Wright 1988: 181)
or Flavius mimicry of the embarrassed prevarications of Timons ungrate-
ful parasites:
worcester: I haue not sought the day of this dislike.
king: <You haue| not sought| it: <!>| how comes| it then|?
(1H4 5.1.26)
flavius: They answer in a ioynt and corporate voice,
That now they are at fall, want Treasure[,] cannot
Do what they would, are sorrie: you are Honourable,
But yet| they could| haue wisht|, they know| not <!>|,
Something hath beene amisse; a Noble Nature
May catch a wrench; would all were well; tis pitty,
(TIM 2.2.2049)
Other possible gestures include the shudder or grimace:
ophelia: As if he had been loosed out of hell,
To speake| of ho|rrors: <!>| he comes| before| {me}.
(HAM 2.1.801)
claudius: Oh my offence is ranke, it smels to heauen,
It hath the primall eldest curse vpont,
A Bro|thers mur|ther. <!>| <Pray can> I not|,
(HAM 3.3.368)
The sigh:
portia: I must go in. Aye me! How weake a thing
The heart| of wo|man is|? O Bru|tus, <!>|
The Heauens speede thee in thine enterprize!
(JC 2.4.40)
aegeon: In Syracusa was I borne, and wedde
Vnto a woman, happy but for me,
And by| me; <!>| <had not| our hap| beene bad|:
(ERR 1.1.368)
The sob, genuine or simulated:
antony: Haply you shall not see me more, or if,
A mang|led sha|dow. <!>| Perchance| to mo|{rrow},
Youl serue another Master. . . . (ANT 4.2.268)
octavius: Looke you sad Friends,
The Gods| rebuke| me, <!>| <but it| is Ty|{dings}
To wash the eyes of Kings. (ANT 5.1.268)
A silent beat can cue a start of surprise:
volumnia: I kneele before thee, and vnproperly
Shew duty as mistaken, all this while,
Between| the Childe|, and Pa|rent.
coriolanus: <!>| Whats this|?
your knees to me? \ To your Corrected sonne?
(COR 5.3.547)
desdemona: And bid| me to| dismisse| you.
emilia: <!>| Dismisse| {me}?
(OTH 4.3.14)
Or a gesture of exasperation:
hotspur: Sicke now? droope now? this sicknes doth infect
The very Life-blood of our Enterprise,
He writes me here, that inward sicknesse, <!>|
And that his friends by deputation
(1H4 4.1.278/301)
othello: And he that is approud in this offence,
. . .
Shall loose| me. <!>| <What[,] in> a Towne| of warre|,
. . .
To Manage priuate, and domesticke Quarrell?
(OTH 2.3.211/213/215)
othello: The Handkerchiefe.
desdemona: A man that all his time
Hath founded his good Fortunes on your loue;
Shard dan|gers with| you.
othello: <!>| The Hand|kerchiefe|.
(OTH 3.4.936)
A silent beat may cue a theatrical registration of dismay or alarm at
being suddenly at a loss for an answer. One delightful example arises when
Cleopatra, wrinkled deep in time (1.5.29), confronts her messenger (still
smarting from their earlier interview) with a dangerous request:
cleopatra: The Fellow has good iudgment.
charmian: Excellent.
cleopatra: <Guesse at> her yeares|, I pry|thee.
messenger: Ma|dam, <!>|
she was a widdow.
cleopatra: Widdow? Charmian, hearke.
(ANT 3.3.257)
He is caught in a cleft stick: if he makes her too young, he risks another
beating, but if he makes her too old for the neer lust-wearyd Antony she
will think hes lying. His momentary alarm, registered in the facial gesture
that signals the silent beat, is effaced by an inspired answer, both satisfying
and non-committal. Some further examples of embarrassment:
olivia: But would you vndertake another suite,
I had rather heare you, to solicit that,
Then Mu|sicke from| the spheares|.
viola: Deere La|dy <!>|.
(TN 3.1.10810)
clarence: Who sent you hither? Wherefore do you come?
second murderer: To, to|, to <!>|
clarence: To mur|ther me|?
both murderers: I, I|.
clarence: You scarcely haue the hearts to tell me so,
(R3 1.4.1715)
George Wright (1988: 181) has drawn attention to a line (printed as two
fragments in most editions) in which Iago just hints a fault, and hesitates
dislike of Cassio:
othello: <Is he| not ho|nest?
iago: <!>| <Honest>, my Lord|?
(OTH 3.3.103)
Bob Hoskins, in the BBC Othello, takes advantage of the intimacy of tele-
vision to half-raise a sceptical eyebrow on the third beat.
In addition to these mimetic and expressive applications of the silent
beat, we also fnd a couple of interesting deictic uses, marking and drawing
attention to features of the interaction between characters. One common
one, for example, is to mark the end of a question, perhaps with a conven-
tional head-tilt or some similar item of body language:
king richard: What sayes| he <!>|?
northumberland: Nay no|thing; all| is said|: (R2 2.1.1489)
king richard: <Shall I| obtaine| it <!>|?
bolingbroke: <Name it,> faire Cou|{sin}. (R2 4.1.304)
gloucester: <Would he| deny| his Le|tter, said| he <!>|?
(LR 2.1.78)
Another kind of rest is the interruption, where the silent beat involves the
speaker breaking off in mid-sentence, as when the Jeweller breaks into the
Osric-like verbosity of the Merchants praise of Timon with an intrusive ges-
ture such as showing the jewel, or Gertrude interrupts her sons intolerable
tirade about Claudius by (perhaps) throwing up her hands:
merchant: A most incomparable man, breathd, as it were,
To an vntyreable and continuate goodnesse:
He pa|sses. []
jeweller: <!>| I haue| a Ie|well heere|.
7
(TIM 1.1.102)
hamlet: . . . A Cutpurse of the Empire and the Rule.
That from a shelfe, the precious Diadem stole,
And put| it in| his Po|cket.
queen: <!>| No more|! (HAM 3.4.99101)
And then theres the absolutely fnal interruption:
hotspur: . . . O, I could Prophesie,
But that the Earth, and the cold hand of death,
Lyes on my Tongue:
No Pe|rcy thou| art dust| / And food| for <!>|
8
(1H4 5.4.836)
Like the jolt, the rest can deictically underscore an imperative, presum-
ably with a supporting gesture. In the following, York perhaps points to or
lays a hand on the desired dagger; in King Richards case, the frst injunc-
tion leave me, marked by a jolt, doesnt seem to work, so the repetition is
accompanied by an intensifying silent beat (perhaps an impatient wave of
the hand or a thump on the table):
york: I pray| you, Vn|ckle, <!>| <giue me> ths Da|{gger}.
(R3 3.1.110)
king richard: ^ Set| it downe|. Is Inke| and Pa|per rea|{dy}?
ratcliff: It is| my Lord|.
king richard: <Bid my> Guard watch|. ^ Leaue| {me}.
Ratcliffe,
about the mid of night come to my Tent
And helpe| to arme| me. <!>| <Leaue me> I say|.
(R3 5.3.759)
ghost: <Thinke on> Lord Has|tings: <!>| dispaire|, and dye|.
(R3 5.3.156)
Another deictic use of the rest is in turning towards a new interlocutor
(perhaps cuing an actual turning of head or body):
bolingbroke: Thankes gen|tle Vn|ckle: <!>| <come Lords| away|,
(R2 3.1.42)
stanley: My Lord| good mo|rrow, <!>| good mo|rrow, Cates|{by}.
(R3 3.2.74)
brabantio: [To Othello] I here do giue thee that with all my heart,
Which but thou hast already, with all my heart
<I would| keepe from| thee. [Turns to Desdemona]
<!>| For your| sake (Ie|{well})
I am glad at soule, I haue no other Child,
(OTH 1.3.1936)
Sometimes exits are punctuated by a silent beat:
martius: Your va|lour puts| well forth|: Pray fo|llow. <!>|[Exeunt]
(COR 1.1.251)
imogen: I
a
m bound| to you|.
belarius: And shalt| be e|ver. <!>| [Exit Imogen]
(CYM 4.2.46)
brutus: Lets . . . carry with us Eares and Eyes for th time,
But hearts| for the| euent|.
sicinius: <Haue with> you. <!>| [Exeunt]
(COR 2.1.26870)
As are entrances:
lady: How my good name? or to report of you
<What I> shall thinke| is good|[?] The Prin|cesse <!>|.
[Enter Imogen] (CYM 2.3.85)
coriolanus: Shall I be tempted to infringe my vow
<In [the same] time|tis made|? I will| not. <!>|
[Enter Virgilia, etc.] (COR 5.3.201)
cassio: I humbly thanke you fort. I neuer knew
A Flo|rentine| more kinde|, and ho|nest. <!>|
[Enter Emilia] (OTH 3.1.3940)
An interesting variant is when the entrance-rest coincides with interrup-
tion of the speaker. Macbeth doesnt want his wife to fnd him contemplating
tergiversation, any more than Iago wants his general to fnd him gossiping
about him, and so the words side and Desdemona are not just cut off by the
entrances but replaced by gestural responses of surprise or alarm:
macbeth: Vaulting Ambition, which ore-leapes it selfe,
And falles| on th o|ther <!>|. [Enter Lady.]
How now|? What Newes|?
(MAC 1.7.279)
cassio: I do not vnderstand.
iago: Hes married.
cassio: To who?
iago: <Marry> to <!>| [Re-enter Othello]
Come Cap|taine, will| you go|?
(OTH 1.2.523)
To illustrate the short pentameter at work in an extended passage consider
the following, in which Emilia hears for the frst time of her husbands slan-
der of Desdemona. It contains four occurrences of the phrase my husband,
all of which are placed by their prosodic and pragmatic contexts in different
relations to the metre. The frst (l.138) is simply a puzzled query (Why men-
tion him? What has he got to do with all this?); the second (l.142) repeats
that query after a silent beat that signifes a kind of stunned double take (the
penny is beginning to drop). The third utterance (l.145) has a drag, cue-
ing an incredulous, horrifed accent on my; by the fourth (l.148) accentua-
tion has returned to normal as she desperately seeks a contradiction from
Othello (whose exasperation provides the jolts in this line):
othello: . . . thy Husband knew it all.
emilia: My Hus|{band}?
othello: Thy Hus|{band}.
emilia: <That she| was false| to Wed|{locke}?
othello: ^ I, with Cassio: had she bin true,
If Heau
e
n would make me such another world,
Of one entyre and perfect Chrysolite,
Ild not| haue sold| her for| it.
emilia: <!> | My Hus|{band}?
othello: ^ I,twas he that told me on her frst,
An honest man he is, and hates the slime
That stickes| on fl|thy deeds|.
emilia: ^ My| ~ Hus|{band}?
othello: What needs this itt
e
rance, Woman? I say, thy husband.
emilia: O Mistris, vill
a
ny hath made mockes with loue:
My Hus|band say| sh
e
was false|?
othello: ^ He|, ^ Wo|{man};
I say thy Husband: Dost vnderstand the word?
My friend, thy Husband; honest, honest
I
ago.
(OTH 5.2.13954)
Thus by experimenting with silent beats and silent off-beats Shakespeare
arrived at a kind of line that was ideal for the oral medium of the the-
atre: supple, naturalistic, varied, capable of striking expressive and mimetic
effects and yet retaining all the advantages of metered verse. It is true that
short pentameters remain relatively rare in Shakespeares verse, probably
because they are parasitic upon the normative syllabic regulation of the
line; too many of them would undermine our sense of the decasyllabic
norm against which they must be perceived. Shakespeare himself scrupu-
lously avoided them in his non-dramatic verse, in obedience to the domi-
nant prescriptive belief that the line must be decasyllabic; it was only in the
more relaxed oral mode of the dramatic pentameter, it seems, that he felt
at liberty simply to record the rhythmic structures he imagined, pauses and
gestural beats included, without worrying about counting syllables.
Notes
1
The process begins as early as the First Folio itself (see OConnor 1977).
2
It is true that metrical and prosodic variations cannot mean directly, but they can
certainly seem an Eccho to the Sense (Pope, Essay on Criticism 365).
3
In what follows I shall use <!> to indicate a silent beat, ^ to indicate a jolt (a silent
off-beat that falls in a syntactic break), ~ for the drag (a silent off-beat that comes
between words not separated by a syntactic break), | to indicate the end of a foot,
<> to enclose a reversed foot, [] to enclose a swap (see below), { to indicate a fem-
inine ending, superscript letters (
e
) to indicate syncopation or elision, and to
indicate where a metrical line-ending is for some reason (such as compositorial
convenience) not registered in F.
4
This foot has two independents and so may be realized either as a normal or as a
reversed foot, which is the meaning of the < | notation.
5
Compare That croakes| the fa|tall en|t[e]rance| of Dun|{can}, Macbeth 1.5.39. Four
out of the nineteen verse-occurrences of entrance in Shakespeare are trisyllabic.
6
Macbeth 2.1.57 as it appears in F1 needs relineation; since Rowe, however, editors
have routinely (and mistakenly) relineated the entire sequence 2.1.27.
7
Most editions print a full stop after passes, as in F1, but while the verb could be
used intransitively to mean excels this leaves the silent beat unexplained (and
in any case He passes. seems far too laconic for this speaker when he could say
something like He passes the extolment of all virtues, / And . . . ).
8
Editors follow F1 in placing No, Percy, thou art dust after Lies on my tongue,
and stranding and food for as a fragment. My arrangement arguably makes
better sense, with an expressive pause after Lies on my tongue and the grunt or
death-rattle expressed as the fnal rest of Hotspurs fnal line.
References
Quotations are from Hinman, Charlton (ed.) 1968. The Norton Facsimile: The
First Folio of Shakespeare. New York: W. W. Norton & Co., Inc., except where
otherwise indicated; where a line is shared between speakers, all segments
of that line after the frst are indented, as in modern edited texts, in order
to clarify the metrical structure of the whole line. Line-numberings refer to
Evans, G. Blakemore and J. J. M. Tobin (eds) 1974. The Riverside Shakespeare.
Boston: Houghton Miffin Company.
Attridge, Derek 1982. The Rhythms of English Poetry. London: Longman.
Bowers, Fredson 1980. Establishing Shakespeares text: notes on short lines and the
problem of verse division, Studies in Bibliography 33: 74130.
Furness, Howard H. (ed.) 1873. Macbeth, vol. 2 of A New Variorum Edition of Shakespeare.
3rd edition. London & Philadelphia: J. B Lippincott Co.
Gascoigne, George 1575. Certayne notes of instruction concerning the making of
verse or ryme in English in: The Posies of George Gascoigne Esquire. London: For
Richard Smith.
Groves, Peter L. 1998. Strange Music: The Metre of the English Heroic Line. ELS
Monograph Series 74. Victoria, B.C.: University of Victoria.
Groves, Peter L. 2005. Finding his feet: Wyatt and the founding of English metre,
Versifcation: An Electronic Journal of Literary Prosody 4 (www.arsversifcandi.net/cur-
rent/groves.html)
Groves, Peter L. 2007. Shakespeares pentameter and the end of editing, Shakespeare
(Journal of the British Shakespeare Association) 3: 12642.
Marlowe, Christopher 1973. The Complete Works of Christopher Marlowe, 2 vols, Fredson
Bowers (ed.), vol. 1. Cambridge: Cambridge University Press.
OConnor, John 1977. A qualitative analysis of compositors C and D in the
Shakespeare First Folio, Studies in Bibliography 30: 5774.
Sicherman, Carol 1982. Meter and meaning in Shakespeare, Language and Style
15: 16992.
Sicherman, Carol 1984. Short lines and interpretation: the case of Julius Caesar,
Shakespeare Quarterly 35: 18095.
Ure, Peter (ed.) 1961. King Richard II, 5th edition. The Arden Shakespeare, London:
Methuen & Co. Ltd.
Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery (eds)
1986. William Shakespeare: The Complete Works. Oxford: Clarendon Press.
Wright, George 1988, Shakespeares Metrical Art. Berkeley: University of California
Press.
Further reading
Groves, Peter L. 2001. What music lies in the cold print: Larkins experimental
metric. Style 35: 70323.
Chapter 7
Wholes and Holes in the Study of
Shakespeares Wordplay
Dirk Delabastita
1
1. Introduction
Nineteenth-century philologists felt confdent enough to put an exact num-
ber on Shakespeares puns (in 1887 Dr F. A. Bather counted exactly 1,062
of them) and to contain their often anarchic abundance within strictly
systematic taxonomies (a game at which Leopold Wurths 1895 study still
remains to be beaten). More recently the study of puns has moved a very
long way indeed from restrictive positions like these. Some of the new criti-
cal paths have led as far as the Derrida-inspired type of position which holds
wordplay to be a principle of unbridled plurality and free linguistic play.
Regardless of whether one agrees with them or not, the historical reality
and indeed the intellectual interest of such radically opposed critical posi-
tions need to be acknowledged. It is easy enough to dismiss them, from
your own vantage point, as the fallacies of a naively self-confdent past or
those of a perversely over-theorized present. But we make our work irrel-
evant to the historical reality of how many of us are now interpreting and
how we used to interpret Shakespeares wordplay if our theoretical models
of the pun cannot somehow accommodate positions as wide apart as those
of Dr Bather and modern poststructuralism.
In such a spirit of comprehensiveness, the approach to Shakespeares
wordplay advocated in this chapter aims to be wide-ranging, fexible and
multiperspectival. We shall propose a whole set of critical distinctions that
need to be considered in the discussion of puns, and, crucially, we shall
argue that these are gradual rather than radical distinctions. No less cru-
cially, our approach is historical and thus fairly relativistic. Among other
points, we shall emphasize the historicity of interpretation, since, after all, it
is our interpretative strategies that make or mar the puns that could poten-
tially be attributed to Shakespeares texts.
2. Defning wordplay
I am offering my defnition as a comprehensive whole that covers various
types of wordplay and links them meaningfully and systematically to vari-
ous types of non-wordplay. But, far from being totalitarian or deterministic,
the model will also provide for the existence of the many holes gaps,
questions, indeterminacies, doubts that keep appearing when we try to
understand and conceptualize Shakespeares punning and its history. In
their legitimate search for descriptively rigorous and theoretically coherent
frameworks or wholes, linguistic approaches to the pun have tended to
turn a blind eye to such problematic hermeneutic holes. Conversely, in
its acute and equally legitimate sensitivity to the latter, postmodern liter-
ary criticism has tended to forget about the importance of the former. If
all goes well, the integrated model presented below should amount to a
whole showing the holes in their right perspective.
Our basic descriptive parameters derive from the following defnition of
wordplay. I suggest that wordplay is best seen as an umbrella term covering
the various discursive phenomena in which certain features inherent in the
structure of the language(s)
2
used are mobilised to produce a communica-
tively signifcant, (near-)simultaneous confrontation of at least two linguistic
units with more or less dissimilar meanings and more or less similar forms.
As in many defnitions of this type, the wording is dense, technical and
inelegant, but a few examples will help to clarify the defnitions key points
and implications. The following short fragment from the punning match in
scene 2.4 of Romeo and Juliet provides us with a suitable starting point:
(1) benvolio Stop there, stop there!
mercutio Thou desirest me to stop in my tale against the hair.
3
There are actually three distinct puns here working together to produce
the joke:
(1a) stop: s1 discontinue
stop in: s2 put in, insert (in order to close or plug an opening)
(1b) tale: s1 account, story
tail: s2 penis
(1c) against the hair: s1 (fguratively) against my inclination or senti-
ment, against the grain
against the hair: s2 (literally) against the (female pubic) hair.
Wholes and Holes in the Study of Shakespeares Wordplay 141
We shall now in the following paragraphs revisit the different conditions
stipulated in the defnition above and apply them one by one to this exam-
ple, beginning perhaps with the issue of communicative signifcance, which
in the case of this integrated sequence of puns is perfectly obvious. Their
function is quite evidently to sustain Benvolios and Mercutios bawdy ban-
ter. The puns must have been intended by the author. They are also meant
to be seen as deliberate on the part of the characters within the fctional
world of the play, which is not always or necessarily the case. Sometimes a
pun is intended by the author over the head of the character who deliv-
ers it blissfully unaware of the double meaning, which may or may not be
picked up by the other characters, either immediately or later. This pro-
duces an effect known as dramatic irony. In their own, less subtle way mal-
apropian puns such as in example (7) below fall under this category too.
There may thus be different embedded levels of intentionality and under-
standing which can produce a range of ironic effects.
Let us continue the exemplifcation of the defnition. The respective
units involved in sequence (1) are formally identical in the cases of puns
(1a) and (1c). They are not identical but highly similar in the case of (1b),
where the components tale and tail show identical sounds but different
spellings. In this respect pun (1b) presents an instance of what is called
homophony, which is traditionally opposed to homography (same spelling
but different sound), homonymy (same sound, same spelling) and paronymy
(both sound and spelling show certain differences against a background of similari-
ties). Needless to say, the exact signifcance of these distinctions requires
further scrutiny when we consider semiliterate cultures, let alone oral ones,
or when we attend to linguistic situations where both pronunciation and
spelling are subject to change and variation, as happened to be the case in
Shakespeares England.
The formal similarities in each of the three puns serve to highlight the
very different meanings contrasted in each of them. Witness the seman-
tic distance between the respective meanings indicated as s1 and s2 in
the glosses under (1a), (1b) and (1c) above. But what really matters of
course is the cumulative effect they have in producing two very different
readings of the same sound sequence [stp n ma tl]. The more or less
striking combination of different meanings contained within the same
or nearly the same forms is widely seen as the main distinctive feature of
wordplay.
The confrontation of forms and meanings can be simultaneous, as in (1b)
and (1c), where the two meanings are superimposed, one on top of the
other so to speak, within a single formal sequence. Using a spatial metaphor,
such wordplay may be called vertical wordplay. Invoking the Saussurean
distinction between paradigmatic and syntagmatic relations in language,
we might also call it paradigmatic wordplay.
4
Pun (1a), on the other hand,
presents the formal constituents of the pun in a near- simultaneous manner.
There are different occurrences of the word stop involving a shift of mean-
ing; they come at the heels of each other rather than appearing together.
While the pun components appear consecutively, they have to do so within
a textual span that is suffciently short to still enable their cognitive linkage
to take place. This may be called horizontal or syntagmatic wordplay.
5
The linking of forms and meanings brought about by the pun is made
possible by the mobilization of certain linguistic principles, alone or in
combination. Thus, pun (1a) exploits the polysemy of the verb to stop,
with stop in possibly being on its way to becoming a phrasal verb. Pun
(1c), on the other hand, presents a literal, compositional reading (s2) of
an idiom (against the hair), which is set off against the meaning that the
idiom has as the indivisible unit which it normally is (s1). A comparable
example of an idiom is to fall in love, which we understand as a single
phrase with a meaning of its own which is different from the sum total of
the meanings of its components, but when Julia asks Lucetta in The Two
Gentlemen of Verona if she would counsel [her] to fall in love, she receives a
punning reply which invokes the literal sense of the idiom: Ay, madam, so
you stumble not unheedfully (1.2.23).
Linguists would argue that polysemy and idiom-formation are closely
related, both being lexical-semantic phenomena. Pun (1b) depends on
phonological structure and that is an altogether different level of language.
More exactly, pun (1b) owes its existence to the fact that English, like any
other language, operates with a limited number of phonemes, which can
moreover occur in certain combinations only. Very similar or even identical
phonological pairs such as tale and tail are therefore bound to occur,
despite their being totally independent words in terms of meaning and ety-
mology. At yet another level of language structure, puns may furthermore
exploit the loopholes of grammar. Some possible uses of morphology and
syntax are illustrated below by examples (2) and (3) respectively. In the
case of morphology-based wordplay, words deriving from compounding or
derivation are decomposed or analysed in an incorrect manner, as in:
(2) antipholus of syracuse Where stood Belgia, the Netherlands?
dromio of syracuse O, sir, I did not look so low.
(The Comedy of Errors 3.2.1378)
s1: name of a country
s2: the nether regions (of her body).
Syntax-based puns involve phrases that can lend themselves to two possi-
ble parsings, as in the following double-edged prophecy:
(3) The duke yet lives that Henry shall depose, (2 Henry VI 1.4.30)
s1: Henry will depose a duke
s2: a duke will depose Henry.
These linguistic mechanisms which are creatively exploited by punsters
are known by all language users inasmuch as we need them to produce ver-
bal utterances and to understand those produced by others. But they nor-
mally operate under the radar of our consciousness. It is only in metalingual
statements in language about language, for example, grammar books or
comments such as these that we become fully aware of phenomena such
as grammatical ambiguity, polysemy or phonemic structure. However sub-
tly, wordplay too could be said to raise our awareness of the mechanisms of
language and of the verbal materiality of the specifc phrase in question. In
other words, wordplay endows language with a certain level of metalingual
information. It causes language to bend back upon itself, reminding us that
meaning comes from language and that languages have their own stubborn
ways of putting things.
Last but not least, for all their dependence on the structural properties
of the language system as recorded in grammars and dictionaries, puns are
discursive phenomena. This means that puns exist as language-in-action.
They occur in concrete utterances, and their meanings and effects depend
crucially on their interplay with many specifc factors the surrounding text,
genre conventions, pragmatic rules and contexts, intertextual networks,
ideologies, psychological dispositions and so on which condition mean-
ing-formation at both the point of production and the point of reception.
The fact that the two points of production and reception are so far removed
from each other with older writers such as Shakespeare makes it doubly
important to adopt a strictly historical and context-sensitive approach.
3. Areas of transition
As we have just suggested, there are no context-free puns foating free in
abstracto. Puns exist in discourses, or they dont exist at all. This part at least
of the defnition given above leaves no room for doubt or disagreement.
But for all the other criteria that make up the defnition, it is possible and
indeed unavoidable to describe puns in terms of degrees to which they
meet its several conditions. In this way, wordplay turns out to be gradually
different from various sets of other, more or less related, discursive phe-
nomena, rather than constituting a totally autonomous and homogeneous
domain. In this sense we could see wordplay and its various types in terms
of prototypical categories. This fundamental fact is worth emphasizing, as
it has all too often been obscured by the tendency of rhetoricians and lin-
guists alike to pigeonhole wordplay and its various possible textual manifes-
tations in systems of discrete categories and neat distinctions.
6
Take the following question: how far can we stretch the near in the con-
dition (near-)simultaneous confrontation of at least two linguistic units
before the breaking point is reached and the components of a horizontal
pun have drifted too far apart for their dissimilar meanings and similar
forms to still refer back to each other? Cases such as pun (1a) above, or
such as Pistols notorious
(4) To England will I steal, and there Ill steal; (Henry V 5.1.86)
present us with textbook examples of horizontal wordplay; the proximity of
their respective pun components plays an important part in making their
confrontation strikingly direct and effective.
7
Now compare these puns with
the following example taken from M. M. Mahoods (1957) classic study
of Shakespeares wordplay. Mahood refers to the end of Macbeths great
monologue in Act 1 scene 7:
(5a) . . . I have no spur
To prick the sides of my intent, but only
Vaulting ambition, which oerleaps itself
And falls on thother (Macbeth 1.7.258)
Mahood argues that [t]he fnal image from horsemanship is so vivid that
it makes possible a kind of long-distance pun in Macbeths words after the
murder (144):
(5b) The wine of life is drawn, and the mere lees
Is left this vault to brag of. (2.3.934)
There may be a vertical pun in the last line on vault as meaning wine-
cellar or sky, but Mahoods point here specifcally concerns the possibility
of horizontal wordplay linking the vault in this line to the much earlier
use of vaulting as leaping in line 1.7.27. This reading is very attractive.
But then, there is no denying that the solid 292 lines which separate the
vaulting ambition (5a) from this vault to brag of (5b) create a very long
textual distance indeed the kind of distance which can only be spanned
by a reading strategy that is prepared to stretch the notion of horizontal
wordplay from an inch narrow to an ell broad.
Mere textual distance is not all that counts, though. See the following
instance:
(6) The more shame for ye. Holy men I thought ye,
Upon my soul, two reverend cardinal virtues
But cardinal sins and hollow hearts I fear ye.
Mendem for shame, my lords. . . (Henry VIII 3.1.1025)
Queen Katherine applies the word cardinal both to the ecclesiastic sta-
tus of the men she is addressing and to their virtuousness or rather lack
thereof. Her speech also contains a horizontal pun on the words holy and
hollow and this is what we are now concerned with. The components of
the pun are separated by 15 intervening words and roughly the equivalent
of two lines. This is much less than the 292 lines in Mahoods example
above, but surely more than in the average horizontal pun. Moreover, the
actual sound similarity between the pun components holy and hollow is
far from being complete. The puns strong semantic effect, then, is largely
due to the complex and tight structural organization of the entire pas-
sage, which places both words in strongly equivalent positions. Consider
the following features: the adjectival character of holy and hollow (both
belong to the same word-class); the semantic and syntactic parallelism of
holy men and hollow hearts (as well as of cardinal virtues and cardi-
nal sins); the syntactic and metrical parallelism of holy men I thought ye
and hollow hearts I fear ye, and the similar relative position of each word
group in its respective line; and, fnally, the chiastic construction of the
entire passage (shame-holy-cardinal versus cardinal-hollow-shame). Critical
attention given to such structural patterning is easily tarred with the brush
of formalism these days, but such attention seems vital to understand the
operation of many a pun.
If examples (4) (steal/steal), (6) (hollow/holy) and, more controversially,
(5) (vaulting/vault) show instances of horizontal wordplay, we cannot be
blind to their mutual differences. Distances between the pun components
in this set of examples range widely between three or four intervening
words to 292 lines; we sense that when the distance gets too long, the hori-
zontal pun is liable to dissolve into non-punning discourse; but we have also
seen how grammatical patterning may play a crucial part in foregrounding
a punning connection. In other words, distinctions between puns and non-
puns are gradual and careful contextualization is necessary. I contend that
what is true in these examples for the distances between punning compo-
nents in horizontal wordplay applies across the board to the other distinc-
tive features in the defnition above, and thus to wordplay generally.
It is worth spelling out the major implications of this argument for what is
the ultimate condition in the defnition, namely the communicative signif-
cance of puns. As we have seen in example (5), the distance between pun
components in a horizontal wordplay can be measured quite accurately. But
it is an altogether more delicate interpretative manoeuvre to assess whether
and to what extent the repetition of vault in Macbeth would have struck
Shakespeare himself and/or the people around him as being in any way
punningly signifcant. It might be safest to assume that the jury is still out on
this one, as well as in the case of many hundreds of other potential puns in
Shakespeare. The jury would at best comprise sensitive and intelligent critics
of the calibre of William Empson, M. M. Mahood, Stephen Booth or Patricia
Parker. But, to continue the legal metaphor, I would argue that very often
there is simply not enough hard evidence to settle the case for or against
wordplay, leaving us somewhere in the zone of incertitude with different
shadings of grey between the obvious puns and the obvious non puns.
Explicit metalingual signals such as Can sick men play so nicely with their
names? (Richard II 2.1.84) will as a rule leave little doubt as to the com-
municative pertinence of puns. But then, such overt cuing is comparatively
rare. Horizontal puns such as illustrated by examples (4) and (6) also tend
to belong to the category of fairly obvious puns inasmuch as they advertise
themselves and their rhetorical intent by the formal repetitions involved in
them.
8
Other unusual features in a text besides repetition may similarly
acquire signal value and help us recognize a meaningful wordplay when
we see one. Such telltale features would include idiosyncratic spellings or
pronunciations, strange collocations, apparent problems with textual cohe-
sion and so on. For example, the reply of Dromio of Syracuse in fragment
(2) above introduces what appears to be a non sequitur until, after a split
second and much to our gratifcation, we realize that Netherlands lends
itself to a double reading that perfectly matches the double context of geo-
graphy and sexual anatomy. Similarly, along with the wider dramatic setting,
it is the logical-semantic and collocational incongruity of Bottoms famous
malapropism
(7) Thisbe, the fowers of odious savours sweet (A Midsummer Nights Dream
3.1.77)
that alerts us to the underlying punning substitution of odious for
odorous.
However, not all textual anomalies have suffcient signal value to guar-
antee the level of communicative signifcance we tend to require of word-
play. Sometimes an anomaly in the text turns out to be just an infelicity of
expression or perhaps some error in the production or transmission of the
text, with no ulterior signifcance attached to it. And sometimes seeming
anomalies or incongruities may be interpreted either way, that is as either
having signalling value or not. It is worth recalling here that Shakespearean
text editors have hit upon several instances of textual variants being formally
similar and contextually appropriate, quite in the manner of vertical word-
play. In such cases one may then wonder if the textual hesitations in the
transmission of the play could be regarded as traces of deliberate wordplay.
Or is the textual crux no more than that an accident in the copying and
printing history of the text? A notorious example of this type of problem is
Hamlets
(8) O that this too too
?
sullied
?
solid
?
sallied fesh would melt, (Hamlet 1.2.129)
in which sullied, solid and sallied are not only textual variants vying for
editorial validation; following Mahoods suggestion (1957), we could also
consider the possibility that all the versions of the phrase may participate in
the complex semantics of Hamlets famous monologue, no longer as mere
textual variants of course, but as intrinsic components of a subtle but artis-
tically effective subdued form of wordplay (Mahood: 15) or portmanteau
word (16).
In the fnal analysis such critical hesitations between Shakespearean
puns and textual cruxes are not very numerous. Even though they may
constitute a subset of it, they are certainly very far indeed from making up
the largest category of potential wordplay where the question of commu-
nicative signifcance is problematically at stake. That troublesome cate-
gory consists mainly of the multitude of potential vertical puns that come
without explicit metalingual cues (compare with the quote from Richard II
above), that are monological (compare with example (1) from Romeo and
Juliet, where the pun serves as a pivot in a dialogical exchange) and that
are not prompted by any particular semantic, cohesive, collocational, stylistic or
other anomaly in the text (compare with Bottoms blatant verbal blundering
in example (7) above), so that, all things considered, the double mean-
ings come as a semantic surplus to a textual sequence that already appears
quite well-formed and meaningful without it. Let us look at one example
that could stand for many. This is Othello speaking, expressing his trust
in Iago:
(9) Why did I marry?
This honest creature doubtless
Sees and knows more much more than he unfolds. (Othello 3.3.2457)
Patricia Parker (1996: 2367) relates this passage to the wider context of
the language that was used to refer to the sexual parts of women:
The privity or lap itself was understood as something folded and hence
needing to be unfolded in order to be available for showa sense of
unfolding exploited . . . in Othellos suspicion that his informer Sees and
knows more, much more, than he unfolds.
Without actually using the term pun or wordplay, Parker suggests that the
semantics of unfold in example (9) may be unpacked as follows:
s1: (fguratively) reveal, disclose (a secret)
s2: (literally) open, display (a womans private parts).
The critic embeds this reading within an intriguing network of intertex-
tual references taken from the same play and from other plays (especially
Hamlet) but also from a range of anatomical and other contemporary dis-
courses outside the Shakespeare canon all of which seem to agree in con-
ceptualizing the female genitals as something hidden, closed and secret,
to be opened, dilated and displayed. True, the convergence of all these
discursive contexts as they are made to crowd in on the above fragment
from Othello lends considerable credibility to the pun that Parker reads
in unfold. But when one (re)situates the phrase in the plays immedi-
ate logical, grammatical and dramatic context rather than in the persua-
sive momentum of Parkers erudite and sophisticated criticism, one may
fnd the case for s2 somewhat less convincing. Isnt Othellos phrase per-
fectly meaningful and coherent linguistically, dramatically, situationally
and psychologically if one simply confnes the meaning of unfold to
s1? Doesnt perhaps projecting s2 onto the phrase create more problems
than it solves? How to account for the mid-sentence shift from the naive
and sentimental indefniteness of Othellos belief in Iagos innate good-
ness and tactful discretion (this honest creature) to the lurid anatomical
precision of a reference to vaginal dilation? And which woman is it anyway
that Iago would have sexually unfolded to Othello to the dissatisfaction
of the latter, who would want to see much more of it? When would such a
scene have taken place? Some readers at least will be left wondering if this
passage in Shakespeares play is really elucidated by the references to the
discursive networks that prompt s2 in Parkers reading. Could it be that the
elucidation rather happens in the opposite direction, with the virtual but
unrealized ambiguity of the word unfold in the fragment being used to
give substance and credence to the critics argument about discourses and
conceptualizations of female sexuality in Early Modern England?
My intention in asking these questions is not per se to challenge Parkers
reading of this passage. Rather, I have wanted to illustrate how potential word-
play of this type vertical, monological, lacking explicit metalingual cues and
occurring in contexts that show no special textual anomalies whatever that
would have required resolution by some double meaning to be unearthed
is bound to be more vulnerable to critical doubt and controversy.
That critical decisions about the signifcance of wordplay are taken
along a spectrum of possibilities and increasing probabilities rather than
in terms of simple either/or positions is already indicated by the subtle
discriminations we fnd in the phraseology of our best wordplay critics. To
quote a short sample of variously hedged pronouncements on the commu-
nicative strength of puns, Mahood speaks of subtle and subdued forms of
poetic wordplay (123), unconscious wordplay (13), punning which is
unintentional (17) or not wholly unintentional (21), submerged puns
(23), unspoken puns, whether conscious or unconscious (25), puns too
dependent on the readers fckle responsiveness to be counted and cata-
logued (51), concealed pun[s] (81), and so forth and so on. Empson
in his Seven Types of Ambiguity identifes latent puns and Stephen Booth
(1977) fnds a range of ideational puns in the Sonnets, as well as many
overtones and echoes (adding that, when he uses words like these, I
mean no more than I say and have no intention of denying or offering
a substitute for the immediately available sense that most of the sonnets
make, 369). Patricia Parker, too, in her Shakespeare from the Margins (1996)
has recourse to a wide range of subtly modulated phrases to acknowledge
that not all double meanings in Shakespeare will strike us as unambig-
uously pertinent because they may not involve suffciently direct reso-
nances or acknowledged subtext[s] (10). Indeed, very often the talk in
Parkers book is of mere resonances, affnities and echoes (many of
which are heard sotto voce), or of unwitting echoes (59), marginal
allusions (75), unnoticed interstitial or marginal links (187), and so on.
Taking care not to overplay her critical cards, Parker identifes many of the
Shakespearean puns she has spotted in the rather uncommittal language of
semantic effects becoming part of a suggestibility (3) and not necessarily
enabling a particular deciphering (15). Critical circumspection may thus
entail a circumlocutionary approach allowing puns to play hide-and-seek
with the reader, as on page 252, where a phrase is said to hid[e] a language
that has lurking within it other meanings (my emphasis).
The various other conditions included in our defnition need to be taken
on board, because the question of the signifcance of potential wordplay is
predicated on them, but it is this last issue that is ultimately both the most
decisive and also the most elusive one. We shall therefore have to return to
it once more at the end of this chapter, where a further complication will
emerge. But meanwhile, the argument about the need for multiple contin-
uums in defning and identifying wordplay should be clear. Much of this
argument may be visualized by the following diagram:
I am emphatically not arguing that rhymes, word repetitions, irony, acci-
dental ambiguities, etc. have to be regarded as instances of wordplay, only
that there tend to be grey zones between the prototypically clear cases of
wordplay in the centre and the various clear instances of non-wordplay on
the border. This is so because the distinctive features summed up in the def-
inition of wordplay (in the diagram printed in italics) operate in a gradual
and context-sensitive manner. Thus, on the side of formal similarity, the
diagram accommodates the notion that wordplay may in some instances
become hard to distinguish from rhyme, assonance and the like. On the
polyptoton and other figures of word repetition
semantic
dissimilarity
formal
similarity
WORD
PLAY
based on
language
structure
rhyme
alliteration
assonance
consonance
jingle
and other
figures of
sound
repetition
communicatively
significant
irony
speech-act
ambiguity
allusion
allegory
metaphor
metonymy
referential
ambiguity
referential
vagueness
accidental ambiguities, unintentional sound echoes, Freudian slips...
single-reading sentences
Figure 7.1 Defning wordplay at a crossroads of four continuums
side of semantic dissimilarity, the diagram refects our experience that
Shakespeare sometimes plays with semantic nuances which may become so
subtle as to be barely distinguishable. And on the side of communicative
signifcance it acknowledges the diffculties inherent in assessing the rele-
vance of perceived potential wordplay.
All of this should be familiar ground by now, but on the side of language
structure the diagrams insistence on the idea of a gradual scale of possi-
bilities challenges what was for a very long time, and may still be for some
linguists, a basic tenet, backed by the authority of Saussures programmatic
separation of langue and parole, namely that a frm line can and should be
drawn between language structure on the one hand and language use on
the other. To be sure, there are many convincing examples on either side
of the distinction which suggest that such a clear separation makes perfect
sense. We have already indicated how the threefold pun in sample (1)
above depends quite decisively on linguistic structure. At the other end
of the spectrum, one fnds countless instances of doubleness of meaning
whose special effect owes nothing to linguistic structure as such. One such
example is the Third Apparitions riddling prophecy in Macbeth:
(10) Macbeth shall never vanquishd be, until
Great Birnam wood to high Dunsinane hill
Shall come against him. (4.1.924)
s1: Great Birnam Wood
s2: soldiers having camoufaged themselves with boughs cut in
Great Birnam Wood.
For all its ambiguity, this phrase indisputably exceeds our defnition of
wordplay, as the Apparitions equivocation hinges on a misleadingly origi-
nal metonymic extension of the word wood and the name derived from it
(Great Birnam Wood), and certainly not on the intrinsic linguistic prop-
erties of that word or name itself. Or consider the famous metaphor in the
fourth line of Sonnet 73:
(11) That time of year thou mayst in me behold,
When yellow leaves, or none, or few do hang
Upon those boughs which shake against the cold,
Bare ruined choirs where late the sweet birds sang;
s1: (vehicle) choir as that part of a church where the service is
sung
s2: (tenor) branches of a tree.
Needless to say, there is neither polysemy, nor any other lexical, grammati-
cal or phonological mechanism at work to generate the ambiguity between
s1 and s2. The latter does not depend on language structure and therefore
the ambiguity cannot qualify as a pun. We are dealing with a quite unique
metaphorical blend which results from creative language use.
9
The ambiguities in examples (10) and (11) result from metonymy and
metaphor respectively and it would be misleading to describe them as word-
plays. The picture that emerges from clear examples such as these connects
wordplay quite specifcally with language structure (of which it cleverly
exploits certain intrinsic features, such as polysemy, etc.), while the effects
of phenomena such as metaphor or metonymy are situated in the realm of
language use only (since, beyond the general productivity of metonymic
or metaphorical association, no specifc linguistic structures are triggered
into operation to achieve the interpretative hesitation). It appears that this
binary logic works well in a good many cases, and certainly well enough to
satisfy the taxonomically minded.
But historical semantics has of course demonstrated one or two funda-
mental facts about polysemy that spoil the binary neatness of the picture.
Polysemy as an intrinsic feature of lexemic structure usually results from
meaning shifts such as metaphorical or metonymic extension, and these
shifts tend to start as more or less occasional or creative instances of lan-
guage use before progressively becoming more common and being inte-
grated into the structures of the lexicon. This process of lexicalization is
given something like an offcial confrmation when the word in question is
declared to be polysemous by lexicographers and starts appearing in dictio-
naries as a complex entry with its diverse senses spelt out distinctly. This is
not necessarily the fnal stage, as in some cases the process of semantic dif-
ferentiation will go further and lead to the point where linguistic conscious-
ness will reinterpret (historical) polysemes as (semantically unrelated)
homonyms. Thus, we treat the words pupil (ward, disciple, child receiving
tuition) and pupil (circular opening at the centre of the iris in the eye)
as unrelated homonyms, whereas they can be traced back historically, via
French and Latin, to a common polysemous root. In fact, Shakespeare may
well have intended a slight pun on pupil in Romeo and Juliet:
(12) romeo Thou chidst me oft for loving Rosaline.
friar laurence For doting, not for loving, pupil mine.
(2.3.778)
s1: disciple
s2: opening in eye; (literally but also fguratively) apple of the (my) eye.
The main context that would support the secondary sense s2 would be
Friar Laurences preceding speech, which enlarges quite elaborately on the
all too common mistake of young men to love in their eyes rather than
truly in their hearts (line 64). Another example would be Shakespeares
more frequent and also more obvious puns on metal and mettle, two
words that have even come to be differentiated by diverging spellings but
which developed from a common lexical root via metaphorical extension.
In cases like these we should beware of underplaying the earlier sense of
semantic connectedness in Early Modern English between such latter-day
homonyms or homophones, reading the semantic realities of Shakespeares
generation through the lexical grid of early twenty-frst-century English.
The crucial point about these lexical processes is precisely that they are
processes occurring along the gradual distinction between language use and
language structure rather than happening overnight at the fick of some
switch. Without wishing to collapse the distinction, we have to conclude
therefore that there is some essential continuity between phenomena such
as creative metaphors or metonymies, on the one hand, and wordplay, on
the other. This continuity was noted by critics such as William Empson
(1984 [1930]), partly because they were hardly interested in the ortho-
doxies and classifcatory niceties of linguistic description anyway, but also
because they had a keen intuitive sense of the deeper cognitive effects of
language. Empsons beautifully cadenced comment on Shakespeares bare
ruined choirs metaphor see (11) above is worth quoting here:
the comparison holds for many reasons; because ruined monastery choirs
are places in which to sing, because they involve sitting in a row, because
they are made of wood, are carved into knots and so forth, because they
used to be surrounded by a sheltering building crystallised out of the like-
ness of a forest, and coloured with stained glass and painting like fowers
and leaves, because they are now abandoned by all but the great walls
coloured like the skies of winter, because the cold and Narcissistic charm
suggested by choir-boys suits well with Shakespeares feeling for the object
of the Sonnets, and for various sociological and historical reasons (the
protestant destruction of monasteries; fear of puritanism), which it would
be hard now to trace out in their proportions; these reasons, and many
more relating the simile to its place in the Sonnet, must all combine to
give the line its beauty, and there is a sort of ambiguity in not knowing
which of them to hold most clearly in mind. (Empson: 23)
For Empson meaning-formation depends on a combination of many fac-
tors over and beyond dictionary meanings, as auditive and visual sensations
become part of the semantic equation as well as a range of contexts and
encyclopaedic (e.g. biographical, historical, political) frames. As a result
the process of understanding can hardly be other than complex, dynamic
and ultimately characterized by a degree of doubt and hesitation; there-
fore, and also as a result of sheer historical distance (hard now to trace out
in their proportions), it may at times defeat the critics best efforts.
Cognitive linguistics has done much to enhance our understanding of
semantic evolution and polysemy, emphasizing how the multiple meanings
of words and thus the use of polysemy in wordplay presuppose complex
forms of cultural or encyclopaedic knowledge and not just the features of
a supposedly autonomous lexical structure. It could even be argued that
monosemy and polysemy cannot be defnitively separated like all catego-
ries, these also have fuzzy boundaries (John Taylor, paraphrased in Crane
2001: 28). The full implication of this may be that our constant recourse
to that most valuable of lexicographical resources the OED may not
quite suffce for an historical understanding of Shakespeares words and
wordplay. We also need to look at Shakespearean words in the many other
discursive contexts in which they occur in order to have a better grasp of
what is at stake beyond their dictionary defnition when Shakespeare puts
them to punning use: who else besides the Bard was using them; how often;
in which senses; within which contexts and institutional settings; with which
range of social or cultural implications, etc.? Such a cognitively inspired
attention to social and cultural meanings conveyed by words and puns char-
acterizes Mary Thomas Cranes Shakespeares Brain. Reading with Cognitive
Theory (2001). Without the explicitly cognitive framework, it is also a key fea-
ture of Patricia Parkers work on wordplay, which frmly broadens language
into discourse and discursive contexts, and therefore routinely oversteps
the descriptive perimeter of the OED curtly described as useful in cer-
tain instances, in spite of its obvious biases and critical omissions (Parker
1996: 17) to consider intertextual semantic networks that emerge in lan-
guage use. Needless to say this type of research is facilitated and further
encouraged by the increased availability of large varieties of Early Modern
English texts in linguistic corpora permitting context-sensitive searches.
4. Historical variations
The model presented here with its multiple descriptive scales helps us see
both differences and continuities between various forms of wordplay and
various forms of non-wordplay. Crucially, the model may also heighten
our awareness of historical change and help us in modelling it, since it is
along the descriptive axes we have identifed that various types of shifts may
occur which have the effect of obscuring wordplay or of changing its impact
for later generations of readers and spectators.
Thus, phonetic changes some of them associated with the fnal stages
of the Great Vowel Shift may play a large part in blocking our access to
Shakespearean wordplay. Classic examples include the names Ajax (Loves
Labours Lost) and Jaques (As You Like It); the phonetic resemblance of both
with jakes (privy, latrine) enabled bawdy jokes to be made which are no longer
directly available to our generation. Here is another textbook example:
(13) [jaques] And so from hour to hour, we ripe, and ripe,
And then from hour to hour, we rot, and rot,
And thereby hangs a tale . . . (As You Like It 2.7.268)
Not until Helge Kkeritzs Shakespeares Pronunciation (1953: 589) did it
become clear why these lame pseudo-philosophical contemplations would
have induced Jaques to laugh, sans intermission, / An hour by his dial
(lines 323). It turns out that the passage involves a bawdy subtext involv-
ing puns on hour/whore, ripe (meaning either ripen or search),
rot/rut and tale/tail. A crucial key to unlock this sustained double
entendre lies in the phonetic likeness of hour and whore, which has
subsequently been weakened by sound evolution beyond unprompted
recognition. The student of the Shakespearean pun will therefore bene-
ft from having frequent recourse to works such as Helge Kkeritzs afore-
mentioned book or Fausto Cercignanis Shakespeares Works and Elizabethan
Pronunciation (1981).
Phonetic dictionaries or philological footnotes will admittedly be of little
use to theatregoers. This could be a reason to perform Shakespeare today
in Elizabethan English pronunciation, or as close a modern approximation
of it as we can manage. Coached by linguist David Crystal, who prepared
phonetic scripts for the actors, the London Globe theatre did experiments
precisely along those lines with Romeo and Juliet and Troilus and Cressida in
their 2003 and 2004 seasons, and one of the interesting side effects of this
exercise in historical couleur locale was indeed to bring some of lost rhymes
and puns to life again.
The effect of semantic change upon Shakespeares wordplay is even more
decisive than that of phonetic evolution. In an article on the puns in The
Comedy of Errors and Twelfth Night which are now falling on deaf ears as a
result of linguistic (mainly semantic) evolution, Gillian West (1990: 6) suggests
that the scale of the loss is suffcient to cause the reader some consternation.
This encouraged West to extend her research and publish her Dictionary of
Shakespeares Semantic Wordplay in 1998. Ambiguities and subtleties in Early
Modern English which are no longer directly understandable today fll up
a large proportion of the indispensable annotations in our critical editions,
while semantic change over the past four centuries has also created a market
for reference works such as David and Ben Crystals Shakespeares Words (2002)
and Eugene F. Shewmakers Shakespeares Language (2008 [1996]). None of
these works, however, suffces and neither does the OED, to cover the lexical
underbelly of Early Modern English that enormous feld of bawdy slang
and allusion, which always features prominently in punning. The semantics
of Shakespeares taboo language is explored in time-honoured studies such
as Eric Partridges Shakespeares Bawdy (1968 [1947]) and E. A. M. Colmans
Dramatic Use of Bawdy in Shakespeare (1974). More recent guides, which gen-
erally show even fewer inhibitions towards sexual taboo, include Frankie
Rubinsteins Dictionary of Shakespeares Sexual Puns and their Signifcance (1984),
Gordon Williams Dictionary of Sexual Language and Imagery in Shakespearean and
Stuart Literature (1994) and Pauline Kiernans juicily titled Filthy Shakespeare:
Shakespeares Most Outrageous Sexual Puns (2006).
These phonetic and semantic issues should always be viewed within a proper
historical sociolinguistic perspective. The social values ascribed in our own
epoch to certain regional varieties of English, to foreign accents, to the use of
certain foreign languages, to the observance of a standard of correct English,
or to rude or obscene language must not be projected back onto the socio-
linguistic context in which Shakespeare worked. The English language was
in a state of rapid fux, much like London itself, which was attracting many
foreigners as well as provincials coming from Britains various regions, to pro-
duce a fairly unregulated Babylonian mix of accents, registers, speech varieties
and even foreign tongues. All of this matters for our purpose. For example,
in order to build certain puns Shakespeare used regional words and pro-
nunciations that nowadays have dialectal or social connotations attached to
them that would not have applied in his age. Similarly, he frequently puns on
hair, heir and air, but it would be a sociolinguistic anachronism to attrib-
ute any social signifcance to such cases of aitch-dropping, as the absence of
the word-initial h would not have been the marker of social inferiority it
was to become in later periods. Historical sociolinguistics and dialectology
are therefore relevant disciplines for the student of Shakespearean wordplay,
who may already beneft from an older and now largely forgotten work such
as Morgans Study in the Warwickshire Dialect (1900).
Conversely, historical change can also produce puns for a later gener-
ation which may not have been intended or meaningful originally. Thus,
the semantics of modern English will highlight the anatomical, scatological
or sexual meanings in words such as bottom, excrement, incontinent,
gay or make love, but such meanings were either absent in Early Modern
English or much less prominent than today, so that we have to mistrust
these words as historical false friends liable to lead us astray in the histori-
cal interpretation of wordplays.
Finally, the need for an historical approach applies also, at a very funda-
mental level, to the way in which language generally connects forms and
meanings, since, here too, the danger of linguistic anachronism is always
just around the corner. We routinely understand Shakespeare through a
linguistic prism that involves a fairly clear sense of individual words and
their correct forms and meanings. This modern awareness of sharply
delineated words is the combined result of several developments which
started to take full effect after Shakespeares age: the growing impact of a
print culture with increasingly uniform spellings and printing conventions;
rising levels of literacy; the progress of linguistic standardization, leading to
clearer distinctions between (and class-bound stratifcations of) language
varieties and involving a clearer sense of what was good and bad usage;
the drive towards greater semantic order, both symbolized and boosted by
the work of lexicographers such as Dr Johnson, possibly as a way to refect
and strengthen the striving for social order; the consolidation of all the
above by the education system; and so on. But we should try to keep in mind
that in the greater fuidity of Shakespeares verbal universe, associative link-
ages could still happen more freely. There was perhaps even a lesser sense
that something specifc like wordplay was taking place, inasmuch as there
was a lesser awareness to begin with of the discreteness of the language ele-
ments that entered into the punning game. In the terms of our defnition
above, the confrontation of the different words and phrases may thus on
the whole have produced a less pointed effect that it might for us today.
5. Interpretative communities
As Rosaline reminds us in Loves Labours Lost (5.2.8524), [a] jests prosperity
lies in the ear / Of him that hears it, never in the tongue / Of him that makes
it. These words could make us think again about the question of the signif-
icance of wordplay and wonder if the notion of a scale of possibilities which
we have advocated is actually good enough to get a grip on the wide range
of responses to Shakespeares punning. Rosaline brings to mind an idea that
centuries later became a cornerstone of reader-response theory, namely that
meanings be they single ones, jestingly double ones or hesitations between
single and double ones are always realized at the point of reception, and
the full implications of this for the understanding of Shakespeares puns are
too often overlooked. Yet, their importance already leaps to the eye when one
looks at that most democratic and least sophisticated form of literary criticism
these days, namely Amazon.coms online customer reviews. Commenting on
Pauline Kiernans Filthy Shakespeare, one very appreciative reader applauds the
authors discussion of sexual puns as a serious dramatic device for important
issues such as morality, politics, and war, calling this an important book which
will be appreciated by playgoers, readers, actors and directors alike (posted by
CB on 2 November 2007). A rather less enthusiastic reader begs to differ:
the book appears to be well-researched and accurate if youve never read
any of these plays. Some of the puns are obvious enough, but Kiernan really
goes over the top on the rest (mostly taking passages out of context) . . . The
book is self-consciously vulgar, attempting to surprise rather than to inform.
Sure, Shakespeare had an incredibly flthy mind, but it seems that Kiernan
has one far flthier (posted by Enter Pirates on 21 November 2008).
10
Both readers are referring to the same text corpus, and it is surely not, or
not primarily, the level of their Shakespearean expertise that distinguishes
them. Rather, they seem to have read their Shakespeare with very different
semantic seismometers, one being super-sensitive and responding eagerly
to the slightest hint of a possible sexual vibration below the textual surface,
while the other will register nothing short of semantic shocks.
The idea of the pertinence of double meanings forming a spectrum
between obvious puns and obvious non-puns does not stop making sense,
but, clearly, the added complication is that we often fail to agree on how
to calibrate the scale. Most critics naturally assume that they occupy the
neutral ground between the two extremes of over-reading and under-
reading. They believe to be alert and receptive to the punning that is really
there in Shakespeare, while wisely refraining from historically irresponsi-
ble pun-hunting. This self-calibration at the sensible centre shows even in
the theoretically refned wordplay criticism of someone like Patricia Parker.
Parker lucidly acknowledges an awareness of her own critical position,
methodological assumptions and theoretical allegiances, but she also seems
quite adamant that her approach to Shakespeares puns is a truly historical
one and free from anachronistic bias. In her actual readings of Shakespeare
she lets the puns mean and sound for themselves: however subtle and sotto
voce their resonance may be, the wordplays are already there, waiting to
be picked up by the critics fnely tuned ear.
It is decidedly not my argument that we should give up trying to read
Shakespeares puns historically and from a maximally neutral or balanced
position. The point is rather that for many potential puns no amount of
historical attention lavished on them will enable us to settle the question of
their signifcance, because neither the texts themselves nor their contexts
provide suffciently clear information on the matter. The authors conscious
intentions and his/her more hidden or unconscious motives are lost in the
mists of time. The same applies to levels of awareness and responsiveness in
the contemporaneous readers as well as in the actors on the stage and the
spectators in the playhouses. Indeed, given the virtual absence of theatre
criticism or other detailed records, the theatrical realization of the poten-
tial puns is poorly documented. How were the lines delivered in terms of
intonation and body language, and how, if at all, did they become part of
the dynamics of the scene?
11
Therefore, for the countless puns in the grey
zone along the scale of communicative signifcance, critics are well advised
to adopt a due measure of modesty, caution and scepticism, even if these
hardly count as strong selling points in the academic marketplace.
The point where we run up against the intrinsic limitations of the avail-
able evidence is also the point where the decisive importance of our
interpretative choices becomes fully visible and where we can therefore no
longer avoid addressing the historicity of interpretation. In other words,
studying puns historically also has to mean recognizing the historicity of
our interpretative strategies. Puns do not simply either exist or fail to exist;
they have a history, and that history depends on reading strategies. Certain
generations or groups of readers (interpretive communities, as Stanley
Fish might say) are more responsive to semantic slippage or doubleness
than others, and will more readily rediscover, discover or perhaps invent
Shakespearean puns by endowing potential double readings and verbal
associations with a semantic substance, a communicative value, a produc-
tivity and/or a form of intentionality they did not possess before, perhaps
not even in the mind of the original author or of his most immediate audi-
ences. Other interpretative communities may again be much less alive to
semantic plurality, if not downright hostile to it. Against the backdrop of
such broad patterns of readerly habits and discursive preferences, a single
infuential critical edition or commentary may then accomplish the feat of
initiating or changing the course of a particular puns history, for example
by securing a place for hitherto undetected double readings in the canon
of orthodox interpretations, or conversely (and more rarely) by dismiss-
ing an accepted ambiguity from that canon as the result of contextually
unfounded pun-hunting. To quote one example among hundreds, Helge
Kkeritz has to be and is widely credited for the recognition of the pun in
example (13) above.
The task of self-critically taking on board our own interpretative strat-
egies adds an extra dimension of interest and complexity to discussions
of Shakespeares wordplay. I would like to conclude by briefy mention-
ing two arguments that merit further consideration in this context. First,
responses to Shakespeares puns are bound to show great variety not only
if we consider diachronic evolutions in the understanding and apprecia-
tion of wordplay, but even if we make the point from a synchronic view-
point, looking, for example, at Shakespeares own age. Even intuitively it
seems evident that the potential puns must have worked out very differ-
ently for Shakespeares original readers (on the one hand) and for those
who saw the plays in performance (on the other), with theatrical shows
potentially offering a greater density of situational context but also impos-
ing a much faster textual pace than any reading act. For the readers a fur-
ther difference may perhaps have to be made between those who saw the
shows and bought the quarto editions, and the later readers of the 1623
Folio, a book which presents the plays at a further remove from the the-
atrical environment which had initially engendered them, giving them a
somewhat greater autonomy as literary texts. If we consider the spectators
in the playhouses, they, too, must have shown a wide range of responses,
even if we agree to make abstraction of the uniqueness of every single per-
formance and the singularity of each individuals theatrical experience.
Efforts have been made to analyse the composition of Shakespeares audi-
ences in terms of their social background and tastes, and then to con-
nect certain types of wordplay with the interests of the subgroups in the
audience; witness the commonly heard argument that Shakespeare would
have indulged in bawdy punning and knockabout humour to please the
palate of the groundlings, while his more refned poetic puns would have
been destined for the educated members occupying the more expensive
places. However, such sociological reconstructions of Shakespeares audi-
ence turn out to be quite problematic, being based on rather unreliable
sources and often refecting the critics own political bias (Boecker 2008).
What we are left with is the sobering conclusion that any statement on
Shakespeares original audience and on how it would have perceived the
wordplay implies a great deal of speculative abstraction.
Second, and fnally, we need to consider the discursive materiality of our
responses to the wordplay in Shakespeare. A readers or spectators response
to a pun may happen with different levels of awareness and of physical and
verbal explicitness. It may, or may not, be externalized by a nod of recognition,
a smile, a chuckle or a ft of belly-laughter. The acknowledgement of a pun
may, or may not, be articulated verbally. When it occurs, such metalingual
awareness may be expressed privately in ones own thoughts or more overtly
in some public form. In an absolutely overwhelming majority of cases, our
responses remain unreported, but when an understanding of a pun is cap-
tured and expressed in some published critical report, the latter is not neces-
sarily an accurate representation of the initial mental reality that it pretends
to put into words. There may indeed be a signifcant gap between the critics
initial understanding of Shakespeares puns and his/her more refective critical
writing on the puns, which is done in accordance with the discursive norms
that happen to prevail in the critics intellectual and social milieu.
Thus, it is too easily assumed that Molly Mahood was not a perceptive
reader of Shakespeares bawdy wordplay, on which she has admittedly little
specifc to say. But let us not confuse her critical writing with the intellectual
reality of her understanding of the puns. As Mahood (19961997: 136) was
to indicate in a reappraisal of her Shakespeares Wordplay four decades later,
I was up against another limitation of time and place: young women tutors,
around 1950, did not explicate Shakespeares bawdy, which could be left
to Eric Partridge. Wordplay critics today work under totally different stric-
tures. Filthy puns stopped being a source of embarrassment years ago and
have actually become a strong pole of intellectual and even commercial
attraction. This is one of the reasons why it is more likely nowadays for extra
puns to be produced by the discursive demands of critical practice than for
puns to be suppressed by them (as happened with Mahood).
There are more such reasons. Consider the more fexible way in which
many wordplay critics have come to defne relevant contexts for double
meanings, extending the sense of context far beyond short individual dra-
matic and verbal sequences, possibly to include all of Shakespeares works
and even their biographical, historical and discursive backgrounds; in other
words, the lesser extent to which recent wordplay critics feel that the onus
is on them to demonstrate the immediate contextual plausibility of a pun.
Consider furthermore: their lesser concern with psycholinguistic realities
restricting the amount of vertical ambiguity that the mind can handle, or
with the limits that our memory span imposes on long-distance horizontal
punning; the zest with which critics mingle their self-produced puns with
the Shakespearean wordplay under discussion; the critics self-granted free-
dom to overrule the linearity of normal readings and performances and
to zigzag across fragments, texts and contexts, thereby assuming simultanei-
ties of allusion and reference. Let us accept that these are more than mere
interpretative strategies enabling us to fnd wordplay in Shakespeare, as they
can also become, in variable degrees, discursive strategies that generate the
puns within the essential context of the critic s writing.
Notes
1
Many thanks are due to the anonymous referees and to Mireille Ravassat for their
useful feedback on an earlier version of this chapter.
2
We cannot enlarge here on the special but not uncommon case of multilingual
wordplay; for this specifc variety of punning, see Parker (2001) and Delabastita
(2002, 2005). But it might be noted that here again, as for other distinctions in the
defnition, the model of a continuum is preferable to that of a binary opposition.
Given the fact that the integration of loanwords into the lexicon of a language is
a gradual process, wordplay involving loanwords can be said to occur in the grey
zone between monolingual and bilingual punning.
3
In this and later Shakespearean quotations, the emphasis is mine.
4
I frst encountered the terms vertical and horizontal wordplay in a few interesting
German studies on Wortspiel from the 1960s and 1970s (Haussman 1974, Wagen-
knecht 1965) which time and the language barrier should not allow us to forget.
5
For the rather special case of implicit wordplays, where the confrontation takes
place neither paradigmatically nor syntagmatically but in absentia, see Mahood
(1957: 24): Sometimes a word, the various meanings of which offer the poet a
range of images, itself remains unexpressed. Booth (1977: 134, passim) provides
examples from the Sonnets.
6
With the help of the model it is of course possible to describe and compare the
various historical, conventionalized typologies of wordplay (e.g. rhetorical ones)
one may encounter. In this chapter there is room only for a presentation of the
model in its generality. The fact that specifc varieties of wordplay were recognized
as having a specifc form and effect is obviously of major historical signifcance,
but I do not believe that we should necessarily follow the terminology and clas-
sifcations of classical rhetoric in our scholarly descriptions of wordplay.
7
Classical and Renaissance rhetoricians called this type of wordplay antanaclasis.
See also my previous note.
8
Inasmuch as horizontal puns are of the long distance variety as with example
(5) this auto-signalling effect tends to dissolve. This has to do with cognitive limi-
tations inherent in our attention span and retention capacity. There is, moreover,
a statistical argument. Repetitions of identical or similar discourse units in a short
text segment are likely to stand out and to catch our attention, but inasmuch as
we consider larger stretches of discourse, such repetitions are bound to become
more frequent and their foregrounding effect will diminish correspondingly.
9
Stephen Booth in his edition of the Sonnets (1977: 259) notes that choirs can
also refer to groups of singers, anticipating the sweet birds later in the same
line. This association, says Booth, gives an extra-logical solidity both to the image
and to the very vague assertion that the speaker (a poet, a singer), is like a season
(ibid.). This effect would qualify as a pun inasmuch as the metonymically based
polysemy of choir (s1: place in a church where singing takes place; s2: group of
singers in a church) became a lexical fact long before Shakespeare had ambiguous
recourse to it.
10
Source: www.amazon.com/Filthy-Shakespeare-Shakespeares-Outrageous-
Sexual/dp/B0014E92O6/ref=sr_1_1?ie=UTF8&s=books&qid=1235772996&
sr=11 (accessed 27 February 2009).
11
For more recent performances, the resources of theatre criticism and especially
the advent of modern recording technologies have made it increasingly possible
to turn to effective interpretative use the fact that situational and pragmatic con-
texts on the stage perform a crucial part in either realizing potential wordplays
or blocking them from the semantics of the performance. We can indeed learn a
great deal from studying how various actors and directors have managed to make
a scene work on the stage with or without the potential wordplays being dra-
matically activated, and how audiences have responded to this. Such theatrical
contexts can and should be read as highly valuable glosses on potential word-
play; as such they are no less useful than the kind of information one may glean
from critical works or annotations in text editions.
References
The text used throughout is The Arden Shakespeare: Complete Works (revised
edition), edited by Richard Proudfoot, Ann Thompson and David Scott
Kastan (London: Thomson Learning, 2001).
Bather, Dr F. A. 1887. The puns of Shakespeare, in: Hawkins, Charles Halford
(ed.), Noctes Shaksperianae. Papers by late and present members [of the Winchester College
Shakspere Society]. Winchester: Warren & Son, 6991.
Boecker, Bettina 2008. Groundling, gallants, grocers: Shakespeares Elizabethan
audience and the political agendas of Shakespeare criticism, in: Dirk Delabastita,
Jozef De Vos and Paul Franssen (eds), Shakespeare and European Politics. Newark:
University of Delaware Press, 22033.
Booth, Stephen (ed.) 1977. Shakespeares sonnets. Edited with analytic commentary by
Stephen Booth. New Haven and London: Yale University Press.
Cercignani, Fausto 1981. Shakespeares Works and Elizabethan Pronunciation. Oxford:
Clarendon Press.
Colman, E. A. M. 1974. The Dramatic Use of Bawdy in Shakespeare. London: Longman.
Crane, Mary Thomas 2001. Shakespeares Brain: Reading with Cognitive Theory.
Princeton and Oxford: Princeton University Press.
Crystal, David and Ben 2002. Shakespeares Words: A Glossary & Language Companion.
London: Penguin Books.
Delabastita, Dirk 2002. A great feast of languages. Shakespeares multilingual com-
edy in: King Henry V and the Translator, The Translator 8(2): 30340.
Delabastita, Dirk 2005. Cross-language comedy in Shakespeare, Humor: International
Journal of Humor Research 18(2): 16184.
Empson, William 1984 [1930]. Seven Types of Ambiguity. London: the Hogarth Press.
Fish, Stanley 1980. Is There A Text in This Class? The Authority of Interpretive Communities.
Cambridge, MA: Harvard University Press.
Haussman, Franz J. 1974. Studien zu einer Linguistik des Wortspiels. Das Wortspiel im
Canard enchan. Tbingen: Max Niemeyer.
Kiernan, Pauline 2006. Filthy Shakespeare: Shakespeares Most Outrageous Sexual Puns.
New York: Gotham Books.
Kkeritz, Helge 1953. Shakespeares Pronunciation. New Haven: Yale University Press.
Mahood, M. M. 1957. Shakespeares Wordplay. London: Methuen.
Mahood, M. M. 19961997. Shakespeares wordplay some reappraisals: a reply,
Connotations 6(2): 1357.
Morgan, Appleton 1900. A Study in the Warwickshire Dialect, with a Glossary and Notes
Touching the Edward the Sixth Grammar Schools and the Elizabethan Pronunciation as
Deduced from the Puns in Shakespeares Plays, and as to Infuences Which may have
Shaped the Shakespeare Vocabulary. 4th edition. New York: The Shakespeare Press.
Parker, Patricia 1996. Shakespeare from the Margins: Language, Culture, Context. Chicago
and London: The University of Chicago Press.
Parker, Patricia 2001. The novelty of different tongues: polyglot punning in
Shakespeare (and others) in: Franois Laroque and Franck Lessay (eds), Esthtiques
de la nouveaut la Renaissance. Paris: Presses de la Sorbonne Nouvelle, 4158.
Partridge, Eric 1968 [1947]. Shakespeares Bawdy: A Literary and Psychological Essay and
a Comprehensive Glossary. Revised and enlarged edition. London: Routledge &
Kegan Paul.
Rubinstein, Frankie 1984. A Dictionary of Shakespeares Sexual Puns and their Signifcance.
London and Basingstoke: Macmillan.
Shewmaker, Eugene F. 2008 [1996]. Shakespeares Language: A Glossary of Unfamiliar
Words in His Plays and Poems. 2nd edition. New York: Facts on File.
Wagenknecht, Christian J. 1965. Das Wortspiel bei Karl Kraus. Gttingen:
Vandenhoeck & Ruprecht.
West, Gillian 1990. Lost humour in The Comedy of Errors and Twelfth Night, English
Studies 71(1): 615.
West, Gillian 1998. A Dictionary of Shakespeares Semantic Wordplay. Lewiston,
Queenston and Lampeter: The Edwin Mellen Press.
Williams, Gordon 1994. Dictionary of Sexual Language and Imagery in Shakespearean
and Stuart Literature. London: The Athlone Press.
Wurth, Leopold 1895. Das Wortspiel bei Shakspere (Wiener Beitrge zur Englischen
Philologie, 1). Wien & Leipzig: Wilhelm Braumller.
Further reading
Delabastita, Dirk 1993. Theres a Double Tongue: An Investigation into the Translation
of Shakespeares Wordplay, with special reference to Hamlet. Amsterdam and Atlanta:
Rodopi.
Delabastita, Dirk (ed.) 1997. Traductio: Essays on Punning and Translation. Manchester:
St Jerome; Namur: Presses Universitaires de Namur.
Lopez, Jeremy 2003. Theatrical Convention and Audience Response in Early Modern
Drama. Cambridge University Press.
Ravassat, Mireille 2007. Assessing and translating the ambiguities of wordplay
in Shakespeares Macbeth, Bulletin de la Socit de Stylistique Anglaise 29: 5160.
http://stylistique-anglaise.org/document.php?id=64
Chapter 8
a thing inseparate/Divides more wider
than the sky and earth of Oxymoron in
Shakespeares Sonnets
1
Mireille Ravassat
Some outstanding scholars have carried out most illuminating investiga-
tions into the nature and meaning of oxymoron in Shakespeares Sonnets,
but no comprehensive and systematic study of this fgure of speech and
thought in this specifc literary work has yet been undertaken. Furthermore,
oxymoron is often seen as the signature trope of Petrarchism, which makes
Shakespeares highly personal use of it all the more intriguing and chal-
lenging and the relative neglect of it in his work all the more puzzling. The
present chapter therefore intends to demonstrate how oxymoron, albeit far
less obtrusive
2
in the sequence than the distantly close fgures of antithesis
and paradox, represents the touchstone for this poetic collection. Indeed
the 21 oxymoric paradigms in the Sonnets display the pattern of difference
within similarity and similarity within difference
3
that has always appealed
to Shakespeare as dramatist and poet. Also, such ever-renewed stylistic com-
binations adumbrate both the baroque vein and the modern vision in this
specifc body of texts.
A biformed concept, to take up the term coined by Francis Bacon,
informed thinking patterns, rhetorical modes, like the habit of arguing
on both sides argumentum in utramque partem or the medical tradition
founded on contradictory humours, thus highlighting a distinctively Early
Modern kind of mental promiscuity in spawning contraries (Clark 1997:
47/50).
4
The culture of Early Modern England, and of Early Modern
Europe as a whole, was pre-eminently a rhetorical one. As Rosalie L. Colie
puts it in Paradoxia Epidemica (1966: 304) the world was then a discordia
concors, a composition to which oxymoron was the most appropriate fgure
of rhetoric. In that era the hallmark of which was [t]he interaction of
contraries (Grudin 1979: 14), Shakespeare was born, took up and revived
a time- honoured literary genre, the sonnet form, and infused it with his
inbred habit of seeing the obverse sides of things manifesting an unparal-
leled ability to give us contrasting things without the slightest diminution
of either (Hubler 1952: 101/48).
Faced with the task of tracing and accounting for the presence of oxy-
mora in the Sonnets, the stylisticians initial assumption is that such a chal-
lenging body of poetry, that discloses as much as it conceals, exhibits
Shakespeares usual bifocal concern with recapturing an undivided state
of wholeness which can only be reached by probing an initial confation
of opposites, at once male and female, weighty and volatile, terrestrial
and celestial . . . An apt example is the quotation from Troilus and Cressida
(5.2.1556) which gives its title to the present chapter. Revealing and defn-
ing the essence of this very moment when things are seen with parted eye, /
When everything seems double, as Hermia says in A Midsummer Nights
Dream (4.1.1889), is one of Shakespeares central issues. Preoccupation
with joining, as Patricia Parker has it, is everywhere in Shakespeare, from
the twain made one of The Phoenix and the Turtle . . . to the explorations of
the implications of marriage as one fesh . . . (1996: 1089)
5
.
More often than not, in his dramatic production, Shakespeare probes
into such instances of joining versus sundering or rather of joining by
means of sundering [f]or nothing can be sole or whole / That has not
been rent, as W. B. Yeats puts it in one of the Crazy Jane poems (1987
[1933]: 295). Murdring impossibility, to make / What cannot be, slight
work (Coriolanus 5.3.612), Shakespeare tirelessly examines the unity of
contraries (The Rape of Lucrece line 1558) exposing and exploring such
rationally improbable phenomena and mechanisms as a miraculous ship-
wreck in Twelfth Night, the chiaroscuro of the tragic psyche in Macbeth, wise
folly and the rhetoric of silence in King Lear, the pilgrimage of hate in Timon
of Athens, the problematic of natural art versus artistic nature in The Winters
Tale, avatars of the black pastoral in Titus Andronicus and Cymbeline, to name
a few. Now, it turns out that the Shakespeare of the Sonnets instead devises
compounds strange (Sonnet 76, line 4), themselves bespeaking a twofold
truth (Sonnet 41, line 12) more often than not a cleft one coming short
of a self-proclaimed mutual render (Sonnet 125, line 12). The Sonnets ulti-
mately leave the protagonists of the love/hate triangle, especially the poet/
young man pair, not so much twinned as twinning in the sense of going
asunder. Initially intended fusion ends up in dissolution.
Behind the protean persona of the poet-speaker, Shakespeare traces and
captures the similitude in dissimilitude, and dissimilitude in similitude
(Wordsworth 1961 [1904/1778]: 740) inherent to the fuctuating moods
and facets of love and desire. Amorous moods, like discourse itself, are
Oxymoron in Shakespeares Sonnets 167
changeable like taffeta as Feste has it in Twelfth Night (2.4.74). Thus the
sonnets straddle the extremities of erotic experience from exhilaration to
utter despondency, from loves delight (36.8) to loathd delight already
explored in The Rape of Lucrece (line 742). Like Du Bellay, Shakespeare has
nearly forgotten the art of petrarchizing
6
, and consequently his poems
do not rehearse standard worn-out phrases such as the topos of the lady as
[s]weet warriour in Spensers Amoretti (Sonnet 57.1: 138)
7
. Shakespeares
oxymora are no mere adornments or colours of rhetoric to paint the blackest
face of woe, to use Sidneys own self-refexive satiric phrase in Astrophel and
Stella (1.5: 2). Instead I will argue that the 21 oxymoric paradigms to be found
in this puzzling collection can be said to map out the narrative of the fendish-
angelical triangle. The structure of the Sonnets could be said to be oxymoric
throughout with a present absent fgure ruling the various intertwined
plots. Indeed the three main protagonists of the Sonnets are engaged in an
ontological progress bearing the stamp of the carnal and the metaphysical
a process of psychomachia highlighted in Sonnet 144. The young man is a
master mistress (Sonnet 20, line 2), one among a crowd of pitiful thriv-
ers (Sonnet 125, line 8), the equally mysterious female protagonist is a
black beauty (Sonnet 127, line 3). As for the poet-speaker, plagued by a
[b]ifold authority (Troilus and Cressida, 5.2.151), he is a willing patient
(Sonnet 111, line 9) wallowing in masochism. Even the moon is mortal
(Sonnet 107, line 5) symbolizing as it does mans existential progress under
the sign of mutability and transience. Finally the narrative strategy at work
in the Sonnets is that of a dialogue of one (Magnusson 2006 [2004]: 635).
Consequently I will argue that oxymoron could very well provide us with the
narrative thread intimately connecting a poetic sequence that might just
otherwise appear disconnected.
.
I take the term narrative in its most basic
sense of telling a story. For even if the plotline of the Sonnets is defnitely a
stylized one and we get neither a physical portrayal of the three protagonists
nor any precisely delineated environment for their actions, still the poems
are no pure philosophical or metaphysical epistles and meditations.
This study means to focus more precisely on the process of oxymoronic
clashing (Jeffries 2010: 125), on the fgure of speech enacting a clash of
contiguity, what Brian Vickers (1984: 70) referring to antithesis, calls a
collision
8
of opposites, sold[ring] close impossibilities, / And [making]
them kiss (Timon of Athens 4.3.3901). Oxymoron is by defnition a close
encounter, and ultimately a welding together, of apparently irreconcilable
opposites admitting no syntactic expansion unlike paradox

although it is
partly attuned to it, acting on one single object, or referent, whereas antith-
esis involves an unreconciled juxtaposition of contraries.
.
As John Porter
Houston explains: oxymoron is a yoking of contrary adjectives and nouns
(1983: 313), an only partly accurate defnition though, as the present absent
oxymoron of Sonnet 45 shows. Such a defnition as is to be found in Sister
Miriam Josephs Shakespeares Use of the Arts of Language (1949: 135), is more
satisfactory. She speaks of a composition of contraries, stimulat[ing] atten-
tion by the seeming incompatibility of the terms it unites. But the major
point I wish to emphasize here is that oxymoron in the Sonnets, defnitely
much more low-key than either paradox, the fgure of the subversion of doxa,
or antithesis, fagging unresolved antinomic parallels, strikingly enough, in
this specifc body of texts, violates its ethos of reconciliation in favour of divi-
sion and estrangement. Very much in a baroque fashion, what oxymoron
does, it undoes
9
thereby expressing a stance of ontological vacillation, but
also the use of this particular fgure in the Sonnets testifes to the emergence
of a modern self, a modern subjectivity. Shakespeares Sonnets, locating
themselves within a literary tradition whose contents and contours they both
revise and supplant articulate the very modern subjectivity effect in western
literary tradition that Joel Fineman explored in an eponymous collection of
texts published posthumously, and this they manage by demonstrating the
intentionality of the literary subject, i.e. his desire . . . motivated by strictly psy-
chological, rather than by biological or by theological, motives (1991: 222).
Or rather, to qualify Finemans statement, Shakespeares Sonnets articulate a
bifocal, partly organic orientation itself informed by the tenets of Galenic
medical theory and a distinctively individualized discourse of interiority
most aptly mirrored in the self-enclosed units of the sonnet form.
* * *
In the very frst of the series of 17 sonnets, to be notably construed as
Shakespeares own secular rewriting of the Biblical exhortation to procre-
ation, the structure follows a series of paradoxical and antithetical state-
ments paving the way for the tender churl oxymoron of line 12:
But thou, contracted to thine own bright eyes,
Feedst thy lights fame with self-substantial fuel,
Making a famine where abundance lies,
Thyself thy foe, to thy sweet self too cruel.
. . .
Within thine own bud buriest thy content,
And, tender churl, makst waste in niggarding.
(Sonnet 1, 58/112) (italics mine)
As for line 8 Thyself thy foe, to thy sweet self too cruel the antithe-
sis enhances the polar nature of the young mans selfhood. When we come
to line 12, the fair friends youthful carefree attitude makes him spend, or
rather, squander his beauty even as he tries to preserve it. He is termed a
tender churl, in other words a youthful miser and a niggard prodigal (The
Rape of Lucrece line 79).
10
For Joel Fineman the sonnets to the young male
dedicatee rehearse the tradition of the poetry of praise, or epideictic poetics,
whereas those to the dark lady, an unusually promiscuous and accessible crea-
ture, at once emulate and imitate this particular poetic genre making it para-
doxical. Fineman construes Sonnet 1 within the framework of Christian and
Neo-platonic self-fulfllment and unity (1986: 246), but perceives the whole
development as a distorted mirror image of a time-honoured orthodox ethos.
Indeed the young man, perceived through the synecdochic prism of his own
bright eyes (line 5) contracted with a pun on betrothed to himself and
restricted to the specular orb of his gaze is defnitely an enemy to his best,
sweet self (line 8), rejecting as he does the exhortation to increase (line 1)
and perpetuation. Nevertheless he remains the paragon of earthly beauty and
as such is worthy of respect and admiration. The young man in the speakers
discourse turns out to be in chiastic terms an embodiment of pure impiety
and impious purity (Much Ado About Nothing 4.1.103).
* * *
In Sonnet 4, the poet harps again on the idea of beauty as a treasure that
should be invested for proft and on the fair friends mission of the perpetu-
ation of beauty:
Then, beauteous niggard, why dost thou abuse
The bounteous largesse given thee to give?
Proftless usurer, why dost thou use
So great a sum of sums, yet canst not live?
For having traffc with thyself alone,
Thou of thyself thy sweet self dost deceive;
(Sonnet 4, 510)
Here the young dedicatees refusal to beget a child is likened to his
wasting an inheritance on self-centred pleasures. Sonnet 4 makes use
of the vocative vein of indictment that characterizes a good number of
the Sonnets.
11
The poet-speakers tone is that of a homily that has been
secularized (Vendler 1999 [1997]: 62). The two oxymora of lines 5 and 7,
beauteous niggard and [p]roftless usurer, come up in vocative form,
which imparts an isomorphic structure to the second quatrain. In the frst
instance, beauteous niggard is to be understood as a miser with a fair
appearance, to be perceived as a combination of antithetical terms, but
the oxymoric tenor is still more obvious, line 7, with [p]roftless usurer.
The poet thereby paradoxically designates a moneylender who makes no
proft. In other words, the friend uses up, exhausts natures bounteous
loan instead of using it to make either money or children. Through his
deliberate choice of non-reproductive sexuality, expressed by the mastur-
batory activity of line 9, the fair friend cuts himself off from any possibility
of a future-oriented identity.
Moreover, contrary to the mitigated violence of churl thanks to the use
of the affectionate epithet tender in Sonnet 1, the beauteous niggard of
Sonnet 4, line 5 testifes to sheer provocation on the part of the older man.
As John Blades argues:
At frst it sounds like an insult, a positive and a negative held in tension.
Beauteous takes up the fattering metonymy of loveliness from the frst
line but niggard drives the poem in a new direction. In the opening line
the youth is Unthrifty, a spendthrift of his beauty, but niggard is par-
adoxically the opposite, a miser. (2007: 8)
This riddle could be solved by arguing that the young man indulges in
bootless sensual pleasures and is thereby unthrifty, but he is also niggardly
in that he privileges carnal indulgence over procreation. This is one among
the instances of oxymora in the Sonnets relevant to the itself oxymoric rhe-
torical strategy of negative politeness probed into by Lynne Magnusson
(2006 [1999]: 50):
.
Negative politeness in its extreme forms is a rhetoric
of contradictions, for it works in such a way as to simultaneously do and
undo the speech actions it undertakes. This particular self-defeating rhe-
torical strategy is explicitly reminiscent of the Antony and Cleopatra quote
(2.2.215); see note 9. Moreover, by avoiding coercing or controlling the
action of the other (Magnusson 2006 [1999]: 50), the poet-speaker of the
Sonnets allows an ever-yawning chasm to divide his own psyche.
* * *
Sonnet 125 is the last before one poem addressed to the young dedicatee.
It is quite dense both in form and content and implicitly harks back to the
procreation poems:
Have I not seen dwellers on form and favour
Lose all, and more, by paying too much rent,
For compound sweet forgoing simple savour,
Pitiful thrivers, in their gazing spent?
(Sonnet 125, 58)
It starts by distinguishing the poet from the crowd of those who crave his
friends love by relying on mere appearances those are dwellers on form
and favour (line 5) but they are bound to fail in their efforts and have to
express their admiration for the young man from a distance, a conclusion
crystallized in the [p]itiful thrivers oxymoron. On the contrary, the poet-
speakers devotion is loyal and heartfelt as opposed to the nothing-gift of
differing multitudes (Cymbeline 3.7.58).
The oxymoric [p]itiful thrivers, as Thomas M. Greene explains, take
their place in a line of disappointed or misguided would-be thrivers distrib-
uted throughout the work, the onanistic friend himself being a case par
excellence, which sheds new light on the proftless usury and the beauteous
stinginess, amounting to barrenness, of the procreation sonnets (2003 [1985]:
231).
12
As Greene further argues, the barrenness of the friends attitude, par-
adoxically [m]aking a famine where abundance lies (line 7), is translated
from the literal, fnancial and biological, level to the fgurative realm of literary
creation: [i]n this econonomic system, all value seems to reside in the friend,
or in thoughts of the friend, and the poet seems to be a leaky vessel constantly in
need of replenishing, his personal and linguistic poverty never defnitely abol-
ished (2003 [1985]: 232).

In other words, in a global structural perspective, the
series of [p]itiful thrivers, beauteous niggard, proftless usurer oxymora
can be interpreted as the threat of both a seminal and semantic bankruptcy
precariously, if triumphantly, counterbalanced by the metastylistic assertions of
Sonnet 76. As usual Shakespeares apprehension of things and events is dual.
* * *
In Sonnet 8, although a new feld of imagery is introduced, as in Sonnets 1
and 4, the young mans insistently solipsistic stance makes him stand above
the fray as a very incarnation of ever jarring-concord and discord-dulcet
(Alls Well that Ends Well 1.1.172):
If the true concord of well-tuned sounds
By unions married, do offend thine ear,
They do but sweetly chide thee, who confounds
In singleness the parts that thou shouldst bear:
Mark how one string, sweet husband to another,
Strikes each in each by mutual ordering,
Resembling sire, and child, and happy mother,
Who all in one, one pleasing note do sing:
Whose speechless song being many, seeming one,
Sings this to thee: Thou single wilt prove none.
(Sonnet 8, 514)
Here, the poet observes the young man listening, Orsino-like, to music
sadly (line 1) and concludes that the fair friend hears in the instruments
individual yet united harmony a condemnation for his refusing to take a part
in the general concord of sire, and child, and happy mother (line 11). The
poet is engaged, oxymorically speaking, in sweetly chid[ing] (line 7) that
is mellifuously rebuk[ing] (Burrow 2002: 396) the tone-deaf handsome
fair friend for his refusal to marry. The apparent simplicity, even navety, of
the gnomic stance of the closing line fnds itself rejuvenated by the pres-
ence of the philosophical issue of the Many and the One, the very idea of
his Oneness having to merge into Manyness which in fact amounts to both
a mathematical and philosophical impossibility
13
being met with a sharp
rebuff by the young man.
Almost all the editors of the Sonnets focus attention on frequent instances
of deliberate point counterpoint echo between subject and numerol-
ogy. Paul Edmondson and Stanley Wells (2004: 60) endorse this very
idea, thus implicitly emphasizing the overall oxymoric substance of the
poem:
The numerological perspective of Shakespeares sonnets, which can often
seem to be an imperative which is proved by degrees of coincidences,
here seems utterly appropriate as Sonnet 8, the number of musical notes
in a complete octave, makes the sweet and bitter concords and discords
of desire attune to the procreation of children.
* * *
Sonnet 18 strikes a very different note, broaching as it does the central
issue of the collection, that is vicarious self-perpetuation by means of liter-
ary creation:
Sometime too hot the eye of heaven shines,
And often is his gold complexion dimmed;
And every fair from fair sometime declines,
By chance, or natures changing course, untrimmed:
But thy eternal summer shall not fade,
Nor lose possession of that fair thou owst,
(Sonnet 18, 510)
The poet thereby solves the dilemma of memory versus oblivion by boldly
suggest[ing], as Michael Schoenfeldt argues that the climatological impos-
sibility of an eternal summer can in fact be attained in the idealizing verse of
the poet (2007: 137). However triumphant it may seem, the self-confdent
closing line of Sonnet 18: So long lives this, and this gives life to thee, the
performative vow of staving off the ravages of Time, fnds itself jeopardized
by the very next line in the sequence, Sonnet 19, line 1 wherein the threat
of the arch destroyer is seen to mangle the lover like a lion. This example
among many highlights the artistry of the whole as a progressive contradic-
tory experience characterized by ambiguous resolutions. Similarly, in a typ-
ically dialectical fashion, poetrys best is to be a tomb, / Which hides your
life, and shows not half your parts (Sonnet 17, lines 34) whereas in the
following sonnet, the literary monument, or eulogistic tribute, becomes the
eternal lines that warrant the young mans immortality, promising to get
him out of the grip of deaths shade (Sonnet 18, lines 12/11), thereby turn-
ing him into an oxymoric boy eternal (The Winters Tale 1.2.64), as Robert
Ellrodt puts it (2007: 34).
* * *
The experience of the Sonnets is defnitely placed under the sign of muta-
bility in very much an Ovidian fashion. Sonnets 35 and 40 should now be
considered as a pair, notably because they feature an oxymoric vision of
robbery in matters of love:
No more be grieved at that which thou hast done;
Roses have thorns, and silver fountains mud;
Clouds and eclipses stain both moon and sun,
And loathsome canker lives in sweetest bud.
All men make faults, and even I, in this,
Authorizing thy trespass with compare,
Myself corrupting, salving thy amiss,
. . .
Thy adverse party is thy advocate,
And gainst myself a lawful plea commence:
Such civil war is in my love and hate
That I an accessory needs must be
To that sweet thief which sourly robs from me.
(Sonnet 35, 17/104)
I do forgive thy robbry, gentle thief,
Although thou steal thee all my poverty;
And yet love knows it is a greater grief
To bear loves wrong, than hates known injury.
Lascivious grace, in whom all ill well shows,
Kill me with spites; yet we must not be foes.
(Sonnet 40, 914)
As for the [s]weet thief of Sonnet 99, line 2, it will be dealt with on its own,
having a totally different thematic and semantic dimension. Both Sonnets
35 and 40 are poems of indictment, or rather mainly of self-indictment in
the case of the former. In Sonnet 35, the poet exculpates the fair friend by
making a catalogue of exempla of degraded beautiful natural objects. He then
accuses himself of being contaminated through the very process of excus-
ing his beloveds faults. The frontiers of the poet-speakers self have become
porous and through a process of mute acquiescence [l]ike the dyers hand
to which he himself refers in Sonnet 111 he takes on the faults of those he
loves (Dubrow 1987: 212). The ambiance is a gloomy one. The poem shifts
from proverbial inspiration in the octave to legal terminology in the sestet,
but the dominating image is that of the civil, or internal, war of line 12,
an insurrection within the state of man, to take up the phrase from Julius
Caesar (2.1.69/67). What has been snatched away from the poet in 35 is more
abstract and conceptual than is the case in Sonnet 40. Whereas in Sonnet 35
the poet may have been deprived of his reputation, of his friends company,
or of both, the theme of Sonnet 40 is the poets accusation against the young
man for having stolen the poets mistress, which becomes still more explicit in
Sonnets 41 and 42. Acknowledging his own corruption as he does in Sonnet
35, the poet-speaker states he has become an accomplice he uses the spe-
cifc legal term accessory (line 13) in the friends sin. As the intestine war
metaphor referred to above shows, Sonnet 35 is eminently a poem evincing
a bifocal perception of things, as can be seen again in the speakers bitter
examination of poetic commonplaces, all of them amounting to the poets
bootless attempts at exonerating himself from blame. As a matter of fact, the
whole structure of Sonnet 35 is cleft. It is, in fact, as though in the course
of a dialogue of one mentioned earlier the poetic voice of Sonnet 35
underwent a splitting of personality with the speaker of the second quatrain
chastising the speaker of the frst for misbehaviour, hence again the civil war
metaphor. The image of civil strife, the sweet-sour, or Mel and Sal , opposi-
tion (Colie 1974: 68134) epitomizing the theme of divided selfhood, cul-
minates in what Helen Vendler calls the love-hate voice the speaker adopts
in speaking of the sweet thief that sourly robs (1999 [1997]: 188), a metaphor
which crops up again in Sonnet 40. Heather Dubrow comments thus upon
this odi et amo trend
14
:
[The poet-speakers] rhetorical equivocations stem from psychological
ones. . . . [The sequences] frequent references to civil war, its repeated
allusions to a loss of identity, all suggest a character who is losing touch
with his own emotions and with the language intended to express
them. . . . It is possible to see much of the rhetoric in the sonnets as . . . the
product not of a skilled sophister but of a mind diseased. (1987: 201)
As such Sonnet 35 adumbrates the willing patient pathological oxymo-
ron of Sonnet 111.
* * *
Sonnet 40 yet evinces a far heavier and gloomier mood than Sonnet 35,
signalling as it does a frst climactic manifestation of the poet-speakers
masochistic wallowing in the abjectness of the love feeling, a masochism
crystallized in the lascivious grace oxymoron of line 13 combining the
spiritual and the carnal with a sense of lustful nobleman vestigially regis-
tered (Burrow 2002: 460). This oxymoron shows a poet-speaker helplessly
enthralled by beauty, for whom aesthetics is the very be-all and end-all of
any human experience worthy of the name. Whatever the infdelity at stake,
infatuation prevails and in taking all [his] loves (line 1), the addressee has
managed to rob the poet of what he paradoxically already lacks, hence his
being a gentle thief. This is a perfect example of what Heather Dubrow,
ascribing a mourning function to oxymoron, explains in the following way:
[t]he rhetorical fgure syneciosis . . . ties together opposites and thus at
once gives and snatches away (1999: 12). Also this oxymoron encapsulates
a process of revival of the medieval concept of psychomachia informing
Shakespeares major tragedies. Thus like Cleopatra, the young dedicatee,
even when wanton, is simultaneously to be condemned and praised for vil-
est things / Become themselves in him, just as they do in her (Antony and
Cleopatra 2.2.2489).
Moreover the lascivious grace oxymoron, with its specifc blend of the
religious and the unchaste, paves the way, in the poet-speakers hell-bound
heavenly progress, for the willing patient oxymoric posture of Sonnet 111,
line 9 at the hands of both the pitiful thriver, the young man, and the
gentle cheater, the dark lady, of 151.3:
Pity me, then, and wish I were renewed,
Whilst like a willing patient I will drink
Potions of eisellgainst my strong infection;
(Sonnet 111, 810)
Randall McLeod comments thus on such an oxymoron as the willing
pacient (line 9):
it is a paradox . . . which we perceive when we remember that the Latin ety-
mon of pacient is pati, to suffer. . . . The transformational desire to suf-
fer is the prerequisite of penance (line 12) and of submitting to a cure
(line 14). This contrasts with the usual fare of static and unredemptive
paradoxes in love sonnets of hot ice and cold fre. This dynamic para-
dox of patience must be seen to transvalue the initial and partial assump-
tion of masochism. (1981: 91)
Such an unobtrusive alliance of polar twins, showing the poet-speaker
trapped in crippling contraries, is certainly a far cry from either Spensers
showy antithesis: My love is lyke to yse, and I to fyre: (Amoretti 30.1: 126)
or Draytons combination of paradox and antithesis in the closing couplet
of his Idea, 62: I have, I want; Despaire, and yet Desire, / Burnd in a Sea
of yce, and drownd amidst a fre (113). The poet-speaker of Shakespeares
Sonnets drinks most delicious poison (Antony and Cleopatra 1.5.28) and expe-
riences a loathd delight, which amounts to the Sonnets distinctive mad-
dening experience. The physiological terminology of Sonnet 111, just as
the explicit humoural lexicon of Renaissance medicine in the twin sonnets,
44 and 45, underscore [a] portrait of desire as a disease that threatens the
physical and mental health of [the] self; [l]yric poet and medical doctor,
then, are both students of inwardness (Schoenfeldt 1999: 75).
* * *
Continuing from Sonnet 71, in Sonnet 72 the poet-speaker again indulges
in a masochistic mood of self-indictment under the sway of loves own
sweet constraint (Alls Well that Ends Well 4.2.16) explaining that the fair
friend can be paradoxically morally justifed by speaking falsely, by giving
the poet more credit than he deserves:
For you in me can nothing worthy prove;
Unless you would devise some virtuous lie
To do more for me than mine own desert,
(Sonnet 72, 46)
The general idea of the poem is that the beloved is urged to forget the
poet once he is dead, a statement fnding itself retrospectively contradicted
in Sonnet 71 by the ambiguous exhortation of line 11: Do not so much as
my poor name rehearse, wherein rehearse is to be perceived as a subtle
antiphrastic pun meaning do not care to repeat my poor name, but also
do not bury it again in the tomb of forgetfulness.
* * *
In Sonnet 99, the oxymoron sweet thief of line 2 is only formally attuned
to the sweet thief of Sonnet 35 and the gentle thief of Sonnet 40:
The forward violet thus did I chide:
Sweet thief, whence didst thou steal thy sweet that smells,
If not from my loves breath? The purple pride
Which on thy soft cheek for complexion dwells
In my loves veins thou hast too grossly dyed.
(Sonnet 99, 15)
Contrary to what happens in the two earlier bleaker sonnets, the tone
is much more light-hearted, and the fair friend, instead of being berated
by the lover-poet for misconduct, fnds himself the object of unmitigated
praise. In this poem, the poet again meditates on the beloveds absence.
Here he accuses spring fowers and herbs of stealing their attractive colours
and odoriferous fragrance from the beloved. The violet in particular is
accused of being a sweet thief, amounting to a hypallage meaning either
fragrant thief or thief of fragrance. At stake is a myth of origin: the cou-
plet concludes that all the fowers stole their attributes, whether sweet, or
colour (line 15), from some aspect of the beloved.
In other words, Sonnet 99 is oxymorically speaking a chaste blazon,
itself a syncretic palimpsest of classical, Biblical and Petrarchan traditions
of comparing the beloveds body to a catalogue of fowers. Sonnet 99
offers one striking example of rejuvenated literary recycling of new and
old discourse. However close in subject to Petrarchs Rime Sparse, Henry
Constables Diana and to Campions lyric, There is a garden in her face
and though [i]t treats a timeworn theme, as Colin Burrow reminds us
(2002: 578), Sonnet 99 encapsulates Shakespeares typical attitude when
it comes to reading the language and style of his predecessors. Here the
specifc wit consists in substituting a literal to a fgurative use of language
in the elaboration of a conceit, traditional in the context of Petrarchan
praise. As such the speaker presents himself as one who really believes the
totally irrational assertions he is putting forward in making the fair friend
the supreme natural nonpareil.
But above all, the meaning of Sonnet 99 cannot be understood fully with-
out reference to the Ovidian intertext of metamorphic transformations call-
ing up the indissociable realities of Eros and Thanatos. The poem is frst and
foremost about transmutation through the process of dying in both senses
of the word. In that perspective lines 35 are reminiscent of the allegory of
the little fower once white, now purple with loves wound, in A Midsummer
Nights Dream (2.1.167) an apt emblem for the cruelty of love.
* * *
Against a background of shining nights (Loves Labours Lost 1.1.90),
Sonnets 27, 43 and 61 form a trilogy wherein the poet fnds himself trans-
fxed by phantasmatic obsession with the beloved. All three poems are about
the projected image of the lover keeping the febrile poet awake at night
.
,
in a state of [s]till-waking sleep (Romeo and Juliet 1.1.181), but among the
trio only Sonnet 43 makes use of two oxymora exploring the chiaroscuro
of presence in absence. Here, as Paul Hammond puts it, Shakespeare gets
away from tradition ascribing no private space to Poet and Boy, be it pas-
toral arbour or limpid stream. Instead, [t]he spaces in which they come
together are dream spaces, imaginary places in which the Poet contrives to
fashion some form of the Boys presence, only to leave us (and, often, him-
self) all the more aware of the Boys actual absence, (2008: 274):
When most I wink, then do mine eyes best see;
For all the day they view things unrespected,
But when I sleep, in dreams they look on thee,
And darkly bright, are bright in dark directed.
Then thou whose shadow shadows doth make bright,
How would thy shadows form form happy show
To the clear day with thy much clearer light,
When to unseeing eyes thy shade shines so?
(Sonnet 43, 18)
Sonnet 43 is inspired at once by the Renaissance belief that eyes create the
light allowing sight and by Shakespeares own omnipresent concern with
the distinction between appearance and essence. Vivid nocturnal dreams,
inevitably part as they are of lifes ftful fever, as Macbeth puts it (3.2.23),
are devoutly to be wished compared to the empty real day. But would the
day beneft from the presence of the fair friend, the real night would seem
but a bootless, bloodless counterfeit. The darkly bright oxymoron of line 4
enhances the paradoxical vein of lines 1 and 2. Shining darkly amounts to a
sheer impossibility in the prosaic rational world, but such an incompatibil-
ity fnds itself resolved in the realm of poetry. Furthermore darkly bright
should be construed as a pun involving a single oxymoric layer. Indeed
the frst sense of darkly bright is luminous behind the eyelids darkened
ocular globes whereas the second meaning is blind but seeing, hence oxy-
moric. The shade/shines oxymoric association of line 8 itself is only to
be interpreted as such if shade is understood as an echo of the shadow
of line 5, and not as a mere synonym of image.
In this specifc sonnet, explicitly characterized by copia, hinging on an
accumulative rhetoric of binary oppositions and balanced isocolons, what
stands out above all is the parodic emphasis on subversion of the norm.
As Joel Fineman has it: it is because Sonnet 43 thus deliberately overstyl-
izes its Petrarchist style and overthematizes its conventionally oxymoronic
themes that we do not really read the poem as a solemn inquiry into opti-
cal epistemology (1988 [1986]: 237). In the end, here again, like in the
very frst sonnet in the procreation sequence, the young man appears as
but a distorted refection of the ideal world of Ideas so that the poem fore-
grounds an ever-widening discrepancy between a moot, ambiguous, ideal
and an oxymoric real exposing the poet-speakers self-division and the ded-
icatees alike. Conceit, in the Shakespearean sense of (brooding) imagi-
nation, is more than ever in the Sonnets a double-edged experience, both
a comfort and an injury (The Comedy of Errors 4.2.65), wherein the self
is submerged, swallowed up, or obliterated by the other, which amounts
to a [p]sychic death, itself equivalent to the loss of personhood as the
psychoanalyst Arnold H. Modell puts it (1996 [1993]: 102).
* * *
When it comes to the two-part poem made up of Sonnets 44 and 45,
the poet-speakers and the young mans separation abides and fies
(Antony and Cleopatra 1.3.104). Both texts trace the eagerly yearned for
progress from duality to indivision. In Sonnet 44 the poet wishes that he
were thought rather than fesh so that he could constantly abide with the
beloved. However the poet, being mortal, is made up of the four elements
and the dullest of these, earth and water, constitute the dominant part of
his being, leaving him desperately in the doldrums, condemned to weep
heavy tears. The twin poem, Sonnet 45, features the poets thoughts and
desires as the noblest elements, those of air and fre:
The other two, slight air, and purging fre,
Are both with thee, wherever I abide:
The frst my thought, the other my desire,
These, present absent, with swift motion slide;
(Sonnet 45, 14)
When the poets thoughts and desires dwell with the beloved, the poet,
reduced to the base elements, earth and water, sinks into melancholy. But such
a dire state is mended when his thoughts and desires return, assuring him of
the friends fair health (line 12). The agony of separation from the beloved is
a literary topos if any, but Shakespeares oxymoron, present absent (line 4),
somewhat rejuvenates an old story. The idea is that the volatile elements in
the poet, those of air and fre, have been dispatched from the poet to his
friend, leaving the former in a state of dejection. The oxymoric combination
in present absent encapsulates this very process of transference. Harping as
it does on the dual, implicitly chiastic, interplay between body and substance
and body and thought, Sonnet 45 signals the failure of the poets mercurial,
Ariel-like, yearning. The devoutly wished for disincarnation as a warrant of
ubiquitous reunion with the beloved only indicates one more stage in the
process of the poets twofold, ever-more splitting personality.
* * *
Sonnet 76 is itself the touchstone for Shakespeares artistic intent in the
Sonnets, devising new meanings by means of oxymoron. Here the poet raises
the issue of why his poetry never alters, but keeps repeating the same lan-
guage and stylistic techniques:
So all my best is dressing old words new,
Spending again what is already spent:
For as the sun is daily new and old,
So is my love still telling what is told.
(Sonnet 76, 114)
The answer, he explains, is that his theme remains ever one and the same,
always rhapsodizing as he does on the beloved and on love. Positioned
almost halfway through Shakespeares collection, Sonnet 76 however offers
an illuminating comment on the authors actually ever-renewed methods
of poetic variation. As such line 2 most aptly exposes the poets false mod-
esty: since quick change is itself synonymous with variation, this line is
itself an instance of variation in choice of expression, and thus shows that
his verse is all but barren or devoid of inspiration. Here Shakespeare is
dressing, not only old words, but also ideas, poetic topoi and common-
places new (line 11).
Sonnet 76 is intensely self-refexive and its purpose is defnitely a metasty-
listic one, full as it is of rhetorical terms and devices designating poetic rules
and verbal arrangements of all kinds (lines 16): invention (line 6), that
is poetic creation, is what the poem is fundamentally about. But the poem,
and even the collection of Sonnets at large derive their meaning from line 11.
What the poet does, despite the self-confessed monotony of the theme (line
10), is dress with a pun on the French dresser, erect (Vendler 1999
[1997]: 345) a new stylistic system designed to clothe old thoughts in a
new garb, to take up a rhetorical topos of the period. Hackneyed as it may
seem, the poets argument, or theme, thus fnds itself, Phoenix-like, reju-
venated like the sun which daily rises and sets, a poetic topos for mans ter-
restrial progress (line 13). Here the coordinator and, far from lessening
the impact of oxymoron, acts as a warrant of simultaneity (Ravassat 2007).
Lesley Jeffries comments thus on one of the functions of the conjunction
and in introducing created opposites: it is a means to
put forward the notion that despite being opposites, and thus stereotyp-
ically mutually exclusive, there is sometimes a paradoxical co-existence
of the two extremes: . . . This is not the same as categorizing them as con-
verses, since they are not simply two perspectives on the same phenome-
non but are conficting but co-existent states that logically should not be
able to co-exist. (2010: 445)
Since the lexicon of any given language is limited by defnition, the poet
has to devise new means to say old things and to dress old topoi new
(Ravassat 2008). Such is the case in Sonnet 1, wherein Shakespeare takes
the conventional Petrarchan praise of chaste female beauty and turns it
upside down. There the poet-speaker defeats the readers expectations by
neither engaging nor indulging in commonplace eulogy of the beloved,
but instead accuses a young man of ravenous self-consumption in his
refusal to beget a tender heir in a position to perpetuate beautys leg-
acy to the world of men beyond the grave. Another notable example is
that of the notorious dark lady. The latter is referred to in terms of her
black beauty (Sonnet 127, line 3) she is an icon of oxymoric black aes-
thetics, that is the dissonant counterpart of the Petrarchan blazon, the
lyrical itinerary of female beauty (Callaghan 2007: 52). Sonnet 127 itself
ends with a paradoxical epigram turning blackness, rather hitherto the
badge of hell (Loves Labours Lost 4.3.250), into the very quintessence of
beauty. Referring to the dark ladys eyes, Shakespeare writes: Yet so they
mourn, becoming of their woe, / That every tongue says beauty should
look so (lines 134). In both cases, whether the black beauty oxymo-
ron or the paradoxical ending of Sonnet 127, Shakespeare defnitely goes
beyond either Sidneys own praise of the oxymoric black beames (47.2:
20) emanating from Stellas dark eyes or Berownes eulogy of the brunette,
Rosaline, born to make black fair (Loves Labours Lost 4.3.257).
* * *
At this stage, my point is to connect the procreation sonnets and the meta-
stylistic sonnet par excellence, Sonnet 76: since the young dedicatee will
not reproduce his fair image in the fesh, the poet has to devise new-found
methods (Sonnet 76, line 4) in order to perpetuate his memory. Of course,
as is a well known fact, the Sonnets are primarily meant for that, but more
implicitly, oxymoron, the Crosse-couple, in Puttenhams (1936 [1589]: 206)
own jocular and vernacular renaming of syneciosis or contrapositum
15
, turns
out to be a mongrel fgure born of a form of semantic miscegenation. And
that very peculiar, prodigious fgure in turn produces incongruous crea-
tures, compounds strange (Sonnet 76, line 4) purely literary and stylistic
artefacts such as an implicit fair black potential agent of subversion or the
explicit master mistress oxymoron of Sonnet 20.
Margreta de Grazia highlights the spectacular watershed that is Sonnet
127, reversing as it does, in a social perspective, the ideal of the perpetua-
tion of an only fair (blond, light-complexioned) commendable race. She
tackles the issue of miscegenation, the creation of a new cross-coupled race,
a virtually fair black one, itself jeopardizing the very ideological edifce of
the Sonnets:
. . . after sonnet 17, it is through his own poetic lines rather than the
youths generational loins that fairs lineaments are to be reproduced,
fairs lineage extended. The fair line ends, however, at 127 with the
shocking declaration that now is blacke beauties successive heire. As if
a black child had been born of a fair parent, a miscegenating successor
is announced, one who razes fairs lineage (And Beautie slandered with
a bastard shame) and seizes fairs language (beauty hath no name)
genealogy and etymology. . . . In praising the youths fair lineaments, social
distinction had been maintained; in praising the mistresss dark colours,
social distinction is confounded. (de Grazia 2001: 159)
Both oxymora the implicitly derived fair black one and the actual
master mistress one however eminently different they may seem, in
fact have in common a facet of incongruity which is one of the hall-
marks of oxymoron.
16
The master mistress of Sonnet 20, the most
famous oxymoron in the whole collection, itself an arresting fusion of
polar twins with a distinctly Ovidian flavour, harking back to the myth
of Pygmalion, is construed by John Blades in the following way: The
young man is a siren for both sexes . . . a misfit, an oxymoron, a lapse
of doting Nature. (2007: 25) The master mistress of Sonnet 20 sug-
gests a play on double gender and feudal service, a recurring feature
of Shakespeares production, also a superimposition of discourses of
androgyny and of misogyny alike. As David Schalkwyk demonstrates
(2007 [2002]: 224):
[t]he poem . . . oscillates between male and female desire and desirability,
in both its position of representation and in the position being repre-
sented. The object of desire is both male and female; but so is the desir-
ing subject. Only at the very end of the poem is the chiastic confusion of
sex and gender naturalised through the transformative power of dot-
ing Nature.
In fact, it could be argued that the master mistress oxymoron of Sonnet
20 feshes out the essential chiastic structure of the Sonnets as a whole cul-
minating in a poem like Sonnet 42 wherein the poet-speaker striving to con-
sole himself for the loss of both the young friend and the mistress, depicts
himself as crucifed (line 12) by faintly oxymoric [l]oving offenders (line 5)
making him fuctuate between unbearable extremes of loss and gain:
If I lose thee, my loss is my loves gain,
And losing her, my friend hath found that loss;
Both fnd each other, and I lose both twain,
And both for my sake lay me on this cross:
(Sonnet 42, 912)
* * *
The present chapter has throughout aimed at showing that whatever the
persona he hides behind, the moralist, the masochist, the mercurial poet . . .
[r]hetoric matters in Shakespeares poems not least because it is the prin-
cipal means of creating variety (Roe 2007: 33), which is the very point of
Sonnet 76. Cross-coupling in totally innovative ways, the oxymoric paradigms
in Shakespeares Sonnets breed new senses, a unique performative language
of action, as well as a totally new mode of narrative approach. Finally, oxymo-
ron in the Sonnets is far less often than in Shakespeares dramatic production
a warrant of either coincidentia oppositorum or reconciliation of opposites.
The Sonnets display a deliberately plain vocabulary, the explicitly metasty-
listic russet yeas and honest kersey noes advocated by Berowne in the early
Loves Labours Lost (5.2.413). Beyond that, as was demonstrated, Shakespeare
innovated in more ways than one. Notably, he managed, to paraphrase his
metastylistic Sonnet 76, to dress an old fgure new. This implies a redefni-
tion of the sonnet genre proper as well as of its speech quirks and manner-
isms, including discarding the too often mawkish Petrarchan posture of
languishing. Also, defnitely a hallmark of Shakespeares Sonnets is the poet-
speakers thriving on contradictions and oppositions, notably in the poems
wherein his solipsistic nature appears most clearly. In such a perspective
Shakespeares Sonnets tirelessly explore the interface between novelty and
tradition, reve[l] in them, infusing the sonnet sequence with unexpected
skepticism and satire, bawdiness and bitterness (Roberts 2007: 172). But
no author albeit one of the greatest is an island, immune from (inter-
textual) infuences and born as he was in a literary culture that valued and
cultivated imitation, Shakespeare followed in his predecessors footsteps.
As Jonathan Bate reminds us, [t]hat good imitation involves difference
as well as similarity, is a cardinal principle of Renaissance poetics (Bate
1993: 87). Refning the Art of Precedent, Shakespeare explores new and
old oxymoric paradigms, infusing fresh blood into stale topoi, in doing
so he relies on the play of linguistic contraries not just for ornament,
but for it fgures the psychology of contrariness (Bate 1993: 69). Also, the
very taut and rigid structure of the Shakespearean sonnet form octave
+ sestet encapsulates a vocabulary of extravagant longing (Schoenfeldt
1999: 76). As such the content and very architectonics of Shakespeares
sonnets signal a sharp oxymoric contrast. In fact, I would argue, relying on
Michael Schoenfeldts analysis, that Shakespeares Sonnets notably capture
the very spirit of that sweet constraint of desire, a dialectic wherein the
Early Modern period fnds itself caught, between a fetish of control
the very ethos of the era most forcibly embodied by Sonnet 94 and a
sense of pre-baroque explosive instability (Schoenfeldt 1999: 18) which
Shakespeares updated use of oxymoron brilliantly illustrates. Oxymoron
not only enacts the collision of opposites mentioned above, but also an
explosion of semantic categories. In that, oxymoron both contains and
overfows the very limits of the language of the self, thereby bridging, if
only partially so, the gap between the Early Modern and the modern self.
Above all what I have emphasized is that Shakespeares oxymora do
have an organic, structural, function, substantiating the poet-speakers dis-
course. Shakespeare is openly anti-Petrarchan in Sonnet 130, less meta-
Petrarchan (Cousins 2000: 117) than Sidney who explicitly mocks
the trite evocations of unrequited lovers, experiencing living deaths,
plagued by deare wounds and enduring faire stormes and freesing fres
(Astrophel and Stella 6.4: 4)
17
but defnitely what Shakespeare achieves in
his own Sonnets is creating a distinctively jarring strange harmony which
must be expressed in . . . discords (Hoskins 1935 [ca. 1599]: 36) by means
of oxymoron.
This study has sought to demonstrate that, far beyond the (meta-)
Petrarchan paraphernalia, the young man appears as an at once ambig-
uously perfect and fawed refection of an ideal of unity which in fact
conceals a circumscribed bisection of the self-contained (Fineman 1988
[1986]: 248), an avatar of the bifold authority ruling Cressida in Troilus
perturbed psyche. This pattern of difference within similarity informs the
character of the young man defeating in a way the essence of reconcilia-
tion which is the very hallmark of oxymoron. This impossible unity, this
solipsistic, narcissistic self-absorption of the young man in turn generates a
sense of ever-shrinking embedded circularity accounting for the obsessive
leitmotiv of withinness that Fineman traces in a number of sonnets
18
and
he applies to the poet-speakers own stance even outside the yoked pair
of sonnets, 44 and 45: this same chiasmicized withinness . . . governs the
poets present-absent, hither-thither relationship to the young man (248).
Fineman applies the same train of thought to Sonnet 48:
Thee have I not locked up in any chest,
Save where thou art not, though I feel thou art,
Within the gentle closure of my breast,
From whence at pleasure thou mayst come and part;
(Sonnet 48, 912)
The poet-speaker further becomes ensnared within a pattern of oxy-
moronic yoking (Fineman 1988 [1986]: 249); his own identity fractured
under the infuence of his master mistress, he can, like a blasphemous
Christ, but drink [p]otions of eisell like an oxymoric willing patient
(Sonnet 111, lines 10/9). Doing that the poets only sure uncertainty
(The Comedy of Errors 2.2.184) is that of the dear divorce (Timon of Athens
4.3.384) of himself and himself and the sweet friend. This Shakespeare
achieved in an unprecented way, at once in a pre-baroque spirit and on
the cusp of modernity, acutely aware as he was of the variations of identity
caught in a process of ever-fuctuating experience. This notably results in
a pre-Hegelian master and slave relationship emphasizing the asymmetry
of self and other itself resulting from the asymmetry of desire (Modell
1996 [1993]: 100), a notion highlighting the lord/vassal relationship rul-
ing the Sonnets. Also when the poet-speaker of Sonnet 121 solemnly asserts
I am that I am (line 9), duplicating the divine ontological statement par
excellence, he could just as well say I am not what I am (Othello 1.1.64),
although not in a Machiavellian way like Iago, but rather in a vein of onto-
logical vacillation. There lies the rub.
A different sameness (Fineman 1988 [1986]: 259) plagues the young
man and accounts for his cleft identity and in turn for the poet-speakers
devoted but excruciatingly painful attachment to him. As such the young
man fails to be an embodiment of concordia discors contrary to, for example,
in a dramatic context, Cleopatra reconciling art and nature in Enobarbus
ekphrastic evocation of her foating on a barge on the river Nile, or Antony
referred to as a heavenly mingle (1.5.62) a possible defnition of oxymo-
ron compounded as he is of human and superhuman parts and attributes,
and beyond that being an incarnation of the resolved struggle for the inf-
nite within the fnite (Davis 2007: 97 n3), thereby adumbrating the radiant
achievement of the romances.
Notes
1
I am grateful to Heather Dubrow, Robert Ellrodt and Jonathan Culpeper for their
precious feedback on this chapter.
2
A paradox in itself since oxymoron is by defnition the show-off among fgures of
speech, as Helen Vendler puts it (cited by Raymond Gibbs 2002 [1994]: 95).
3
The present development will notably highlight the intimate connection between
oxymoron and chiasmus. As Joel Fineman puts it: Rhetoric gives many names
to this troping of trope syneciosis, antimetabole, metathesis, contrapostum, conjunc-
tio, commistio, chiasmus all of which refer to some way that language manages
noticeably to redouble with a difference the complementary similarities of a fg-
urality based on likeness (1988 [1986]: 37).
4
For the distinctively dual Weltanschauung ruling the Early Modern period at
all levels, see Stuart Clarks extensive analysis (2005 [1997]) and also Robert
Grudin (1979). For the rhetorical strategy of argumentum in utramque partem see
also Stephen Greenblatt (2005 [1980]: 2301).
5
For a qualifed assessment of twain in the Sonnets, see Hammonds commentary
of Sonnet 36 (2008: 2757).
6
De voz doulceurs, ce nest que sucre & miel, / De voz rigueurs nest qualos &
fel . . . uvres potiques dition Chamard, iv, 206, cited by Rosalie L. Colie in
Shakespeares Living Art (note 53, p. 92).
7
As K. K. Ruthven puts it: The conceits themselves are ancient, for the idea of love
as warfare stems originally from Ovid (Amores, I.ix), and the burning and freezing
go right back to Sappho. Both occur in Petrarch and later Italian Petrarchizers,
from whom English writers were able to inherit a ready-made vocabulary of love
(Ruthven 1969: 20).
8
Brian Vickers writes: . . . .........if he often unites contraries, Shakespeare much more
frequently opposes them. . . . This tremendous density of opposites points less to
the binary structure of the human mind than to the endless possibilities of gen-
erating energy from collision (1984: 70). Such an energy resembles the heat
generated by the interaction of two bodies that rub or jar against one another,
as Stephen Booth argues (1997: 6).
9
To paraphrase Enobarbus pronouncement in his evocation of Cleopatra sailing
on the river Nile, her cheeks simultaneously glowed and cooled by her atten-
dants fans: And what they undid did (2.2.215).
10
Stephen Booth points out the existence of a double-barrelled oxymoron, frst
tender churl meaning gentle boor with tender echoing and playing on
tender meaning young in line 4, and he goes on to explain that . . . [s]ince
tender carries a reminder of tendering, offering, giving tender churl embodies sug-
gestion of another oxymoron: generous miser (2000 [1977]: 136).
11
See David Schalkwyk on the interplay of the contradictory and indissociable poet-
ics of praise and of blame in the Sonnets and Twelfth Night (2008: 138). Also the
use of the vocative here evinces a sense of immediacy, of the poet-speaker being
intent on buttonholing his addressee thereby eliciting a performative speech act,
which is part of the ethos of oxymoron, and thereby matches the performative
nature of the Sonnets underscored by David Schalkwyk (2007 [2002]: 12).
12
For similar conclusions, see Bradin Cormack: Sonnet 126 takes leave of the
beloved, while in sonnet 125 the speaker takes leave of his own desire for the
beloved. . . . The gaze of the thrivers, although outwardly directed, is self-absorbed
and self-destructive. Line 8 thus returns us to sonnet 1 . . . The phrase pitifull
thriuors suggests abundance and lack (2009: 139).
13
Indeed as Paula Blank explains in Shakespeare and the Mismeasure of Renaissance
Man (2006: 51): The proverb one is no number derived from the Pythagorean
idea that One, as the antithesis of Many, cannot itself be subdivided; it cannot be
a number because it contains no plurality.
14
In the combined civil war and odi et amo vein, see also the pretty wrongs oxymo-
ron in Sonnet 41 (line 1) explained in the following terms by Heather Dubrow:
The oxymoron that opens the poem may indirectly prepare us for the internal-
ized oxymoron in the speaker: to describe wrongs as pretty (1) can be the result
of an attraction to loving hate or even . . . of a tendency toward another oxymo-
ron, praising blame (1987: 208).
15
In Puttenhams defnition, the Crosse-couple takes me two contrary words, and
tieth them as it were in a paire of couples, and so makes them agree like good
fellowes, as I saw once in Fraunce a wolfe coupled with a mastiffe, and a foxe with
a hounde (1936 [1589]: 206). As for Hoskins, he defnes the scheme syneciosis,
also known as contrapositum, as [a] composition of contraries, and by both words
intimateth the meaning of neither precisely but a moderation and mediocrity of
both; (1935 [ca. 1599]: 36).
16
See Raymond Gibbs: [o]ne of the most visible fgures of thought and speech
that refect our incongruous understanding of experience is the oxymoron
(2002 [1994]: 394) and also Javier Herrero Ruiz: In paradox and oxymoron, the
incongruity between two or more terms is solved via a projection space that arises
from integration and accomodation operations in such a way that the conceptual
structure of the apparently contrasting terms is possible in a given cognitive envi-
ronment (2009: 264).
17
Shakespeare happens to be explicitly meta-Petrarchan himself in his portrayal
of the pre-Juliet Romeo or of Orsino in the throes of the sweet pangs of love
(Twelfth Night 2.4.16); see Ravassat (2006).
18
See within thine own bud buriest thy content, (1) . . . within thine own deep-
sunken eyes (2,7) or else [t]hy gifts, thy tables, are within my brain (122)
(Fineman 1988 [1986]: 248).
References
Quotations from the Sonnets are from Katherine Duncan-Jones edition, 2005
[1997], Shakespeares Sonnets. The Arden Shakespeare, Third Series. London:
Thomson. Other Shakespeare quotations are from The Complete Works 2007
[2001] The Arden Shakespeare. Revised edition, Richard Proudfoot, Ann
Thompson and David Scott Kastan (eds). London: Thomson Learning.
Bate, Jonathan 1993. Shakespeare and Ovid. Oxford: Clarendon Press.
Blades, John 2007. Shakespeare The Sonnets, Analysing Texts series. Basingstoke and
New York: Palgrave Macmillan.
Blank, Paula 2006. Shakespeare and the Mismeasure of Renaissance Man. London: Ithaca
and London, Cornell University Press.
Booth, Stephen (ed.) 2000 [1977]. Shakespeares sonnets Edited with analytic commen-
tary. Yale: Nota Bene Books, Yale University Press.
Booth, Stephen 1997. Shakespeares language and the language of Shakespeares
time, in: Stanley Wells (ed.), Shakespeare Survey 50, Shakespeare and Language.
Cambridge: Cambridge University Press, 117.
Burrow, Colin (ed.) 2002. The Complete Sonnets and Poems. Oxford: Oxford Worlds
Classics, Oxford University Press.
Callaghan, Dympna 2007. Shakespeares Sonnets. Malden MA, Oxford UK, Carlton,
Victoria, Australia: Blackwell Publishing.
Clark, Stuart 2005 [1997]. Thinking with Demons: The Idea of Witchcraft in Early Modern
Europe. Oxford: Oxford University Press.
Colie, Rosalie L. 1966. Paradoxia Epidemica: The Renaissance Tradition of Paradox.
Princeton: Princeton University Press.
Colie, Rosalie L. 1974. Shakespeares Living Art. Princeton: Princeton University
Press.
Cormack, Bradin 2009. Shakespeares Narcissus, Sonnets Echo in: Leonard
Barkan, Bradin Cormack and Sean Keilen (eds), The Forms of Renaissance Thought.
Basingstoke and New York: Palgrave Macmillan, 12749.
Cousins, A. D. 2000. Shakespeares Sonnets and Narrative Poems. Harlow: Longman.
Davis, Philip 2007. Shakespeare Thinking. London: Continuum.
De Grazia, Margreta 2001. The scandal of Shakespeares Sonnets, in: Catherine
M. S. Alexander and Stanley Wells (eds), Shakespeare and Sexuality. Cambridge:
Cambridge University Press, 14667. First published in Shakespeare Survey
46(1994): 3549.
Dubrow, Heather 1987. Captive Victors: Shakespeares Narrative Poems and Sonnets. New
York: Cornell University Press.
Dubrow, Heather 1999. Shakespeare and Domestic Loss: Forms of Deprivation, Mourning,
and Recuperation. Cambridge: Cambridge University Press.
Edmondson Paul and Stanley Wells 2004. Shakespeares Sonnets. Oxford: Oxford
Shakespeare Topics, Oxford University Press.
Ellrodt, Robert 2007 (editor and translator). William Shakespeare: Sonnets. dition
bilingue, Babel. Arles: Actes Sud.
Evans, Maurice (ed.) 1984 [1977]. Elizabethan Sonnets. London and Melbourne:
Dent and Sons Ltd, Everymans Library.
Fineman, Joel 1988 [1986]. Shakespeares Perjurd Eye: The Invention of Poetic Subjectivity
in the Sonnets. Berkeley and Los Angeles: University of California Press.
Fineman, Joel 1991. The Subjectivity Effect in Western Literary Tradition: Essays Toward
the Release of Shakespeares Will (includes: Introduction: Joel Finemans Will by
Stephen Greenblatt). Cambridge MA and London UK: The MIT Press, An
October Book.
Gibbs, Raymond Jr. 2002 [1994]. The Poetics of Mind: Figuration, Thought, Language,
and Understanding. Cambridge: Cambridge University Press.
Greenblatt, Stephen 2005 [1980]. Renaissance Self-Fashioning: From More to Shakespeare.
Chicago: The University of Chicago Press.
Greene, Thomas M. 2003 [1985]. Pitiful thrivers: failed husbandry in the Sonnets,
in: Patricia Parker and Geoffrey Hartman (eds), Shakespeare and the Question of
Theory. London: Routledge, 23044.
Grudin, Robert 1979. Mighty Opposites: Shakespeare and Renaissance Contrariety.
Berkeley, Los Angeles and London: University of California Press.
Hammond, Paul 2008. Shakespeares male utopias, in: tudes Anglaises 613. Paris:
Klincksieck, 26678.
Hoskins, John 1935 [ca. 1599]. Directions for Speech and Style, Hoyt H. Hudson (ed.).
Hubler, Edward 1952. The Sense of Shakespeares Sonnets. New York: Hill and Wang.
Jeffries, Lesley 2010. Opposition in Discourse: The Construction of Oppositional Meaning.
Advances in Stylistics (series editor: Daniel McIntyre). London and New York:
Continuum.
Joseph, Miriam, Sister 1949. Shakespeares Use of the Arts of Language. New York:
Colombia University Press.
Magnusson, Lynne 2006 [1999]. Shakespeare and Social Dialogue: Dramatic Language
and Elizabethan Letters. Cambridge: Cambridge University Press.
Magnusson, Lynne 2006 [2004]. Shakespeares Sonnets: A modern perspective, in:
Barbara A. Mowat and Paul Werstine (eds), Shakespeares Sonnets and Poems. New
York: Folger Shakespeare Library, Washington Square Press.
McLeod, Randall 1981. Unemending Shakespeares Sonnet 111, Studies in English
Literature 21. Houston: Rice University, Johns Hopkins University Press, 7596.
Modell, Arnold H. 1996 [1993]. The Private Self. Cambridge MA and London:
Harvard University Press.
Parker, Patricia 1996. Shakespeare from the Margins: Language, Culture, Context. Chicago
and London: The University of Chicago Press.
Porter Houston, John 1983. The Rhetoric of Poetry in the Renaissance and Seventeenth
century. Baton Rouge: Louisiana State University Press.
Puttenham, George 1936 [1589]. The Arte of English Poesie, Gladys Doidge Willcock
and Alice Walker (eds). Cambridge: Cambridge University Press.
Ravassat, Mireille 2006. The pangs of disprizd love on some discourses of amorous
languor and melancholy in Shakespeare in: Max Duperray, Adrian Harding and
Joanny Moulin (eds), Discourses of Melancholy. Aix-en-Provence: e-rea, 4.1, spring
2006: 518. Available at: http://erea.revues.org/index402.html
Ravassat Mireille 2007. Oxymoron, hendiadys and coordinate structures
Shakespeare from duality to indivision, in: Wilfrid Rotg (ed.), Bulletin de la
Socit de Stylistique Anglaise 28, Paris, 95110. Also available at: http://stylistique-
anglaise.org/document.php?id=548.
Ravassat, Mireille 2008. So all my best is dressing old words new lart de
Shakespeare dans les Sonnets, in: Pierre Kapitaniak and Jean-Michel Dprats
(eds), Costume et dguisement dans le thtre de Shakespeare et de ses contemporains.
Paris: Socit Franaise Shakespeare, www.societefrancaiseshakespeare.org/doc-
ument.php?id=1467
Roberts, Sasha 2007. Shakespeares Sonnets and English Sonnet sequences, in:
Patrick Cheney, Andrew Hadfeld and Garrett A. Sullivan, Jr. (eds), Early Modern
English Poetry: A Critical Companion. Oxford: Oxford University Press, 17283.
Roe, John 2007. Rhetoric, style, and poetic form, in: Patrick Cheney (ed.), The
Cambridge Companion to Shakespeares Poetry. Cambridge: Cambridge University
Press, 3353.
Ruiz, Javier Herrero 2009. Understanding Tropes at the Cross-Roads between Pragmatics
and Cognition. Frankfurt am Main: Peter Lang.
Ruthven, K. K. 1969. The Conceit. London: Methuen.
Schalkwyk, David 2007 [2002]. Speech and Performance in Shakespeares Sonnets and
Plays. Cambridge: Cambridge University Press.
Schalkwyk, David 2008. Shakespeare, Love and Service. Cambridge: Cambridge
University Press.
Schoenfeldt, Michael 1999. Bodies and Selves in Early Modern England: Physiology and
Inwardness in Spenser, Shakespeare, Herbert, and Milton. Cambridge: Cambridge
University Press.
Schoenfeldt, Michael 2007. The Sonnets, in: Patrick Cheney (ed.), The Cambridge
Companion to Shakespeares Poetry. Cambridge: Cambridge University Press,
12543.
Vendler, Helen 1999 [1997]. The Art of Shakespeares Sonnets. Harvard: The Belknap
Press of Harvard University Press.
Vickers, Brian 1984. Rhetoric and feeling in Shakespeares Sonnets, in: Keir Elam
(ed.), Shakespeare Today: Directions and Methods of Research. Florence: La Casa
Usher, 5398.
Wordsworth, William 1961 [1904/1778]. The Poetical Works of Wordsworth, Preface to
the second edition of the Lyrical Ballads, Thomas Hutchinson (ed.). London,
New York, Toronto: Oxford University Press.
Yeats, William Butler 1987 [1933]. Crazy Jane talks with the Bishop, Words for
Music Perhaps, VI in The Collected Poems of William Butler Yeats (Papermac).
London and Basingstoke: Macmillan Publishers.
Chapter 9
Rue with a difference: a Computational
Stylistic Analysis of the Rhetoric of Suicide
in Hamlet
Thomas Anderson and Scott Crossley
1. Introduction
Suicide or self-slaughter, as Shakespeare describes it in Hamlet (ca. 1601)
is the plays motivating force.
1
Hamlets famous deliberation over his own
fate in Act 3, scene 1 To be, or not to be; that is the question (58)
is the most famous contemplation of suicide that Shakespeare offers the
Early Modern audience who would have had conficting opinions about
self-murder because of the shift in the cultural signifcance of the act taking
place at the turn of the seventeenth century. In response to Hamlets death,
Horatio briefy considers suicide in Act 5 when he grabs the poisoned cup:
Never believe it. / I am more an antique Roman than a Dane. / Heres yet
some liquor left (5.2.2824). Horatios affliation with Roman stoicism that
tolerated self-murder as heroic action is in marked contrast to the plays
concern with Early Modern accounts of suicide, which describe it as a most
heinous sin an offence against God, against the king, and against Nature
(Dalton 1626: 234) that results in certain damnation. Other characters
consider the cost of suicide in a culture that prohibited it. In Act 5, a grave-
digger poses a question about the religious fate of Ophelia in reference
to her drowning: Is [Ophelia] to be buried in Christian burial that wil-
fully seeks her own salvation? (5.1.12); the other gravedigger responds
that the coroner . . . fnds [Ophelias interment] Christian burial (45). In
Early Modern England, the coroner headed a jury that tried suicides, and
if the death was ruled self-murder, then the victim and her/his family were
subject to severe punishment, including a paupers burial and loss of family
assets (MacDonald and Murphy 1990: 17). Hamlet himself underscores the
tension in the play between the desire to represent suicide outside of the
religious and legal doctrines that regulate it and the Christian tradition that
The Rhetoric of Suicide in Hamlet 193
precludes this impulse. In Act 1, scene 2, he wishes that the Everlasting had
not fxed / His canon gainst self-slaughter (1312).
This chapter examines the language of suicide, mourning and grief of
both Hamlet and Ophelia within the context of the shifting signifcation of
suicide within Early Modern English culture that alters the meaning of the
act transforming a sin into psychological estrangement. While the plays
depiction of Ophelias apparent suicide registers the rhetoric of religious
sin that offers a narrative of appropriate punishment by Renaissance stan-
dards, Shakespeares representation of Hamlets suicidal impulses and his
fnal enactment of rage highlights the evolving cultural meaning of suicide
during the period, removing it from its religious narrative and placing it
within an emerging discourse of affect (Shneidman 1996) and societal dis-
engagement (Durkheim 1951, Prezant et al. 1988).
In the process of our argument, we demonstrate how a stylistic inter-
pretation can enrich literary approaches to Hamlet and how literary inter-
pretation complements linguistic inquiry. We will use lexico-semantic and
corpus approaches to analyse Hamlets and Ophelias dialogue for suicidal
discourse in order to reveal the texts semantic prosodies and multi-word
meaning structures that have remained under-examined by literary critics
(Hoover 2007). Central to this analysis will be the computational tool LIWC
(Linguistic Inquiry and Word Count; Pennebaker, Francis, and Booth 2001),
which provides an effcient method for analysing the emotional, cognitive,
structural and processing components of discourse. Of interest to our study
is the use of pronominal references, social factors, communicative func-
tions and lexicon related to death, sexuality and religion. A corpus analy-
sis of suicidal discourse can provide an understanding of character that is
more complete and underscores the plays representation of the cultural
tension engendered by shifting social attitudes towards suicide. By compar-
ing the linguistic elements of characters discourse of suicide, we hope to
convey a clear sense of the plays interest in the emerging secular subject
defned by nascent psychological discourses that challenge the hegemony
of religious narratives.
While implicitly arguing for an expanded notion of textuality that
includes the play itself, legal and religious discourses contemporaneous
with the plays frst performance, and a lexico-semantic corpus comprised
of Early Modern and modern affective indicators of suicidal desire, the
aim of this chapter is to suggest that Shakespeare considers suicide ambiva-
lently. MacDonald makes a similar observation about suicide in the play as
it pertains to Hamlet and Horatio, arguing that [t]he emergence of new
and more tolerant ideas about self-destruction is signifcant. . . . For these
ideas created a new ambivalence about suicide that Shakespeare exploits
in Hamlet (MacDonald 1986: 315). He uses the shifting notion of Early
Modern suicide as a case study to demonstrate what he calls Shakespeares
prismatic imagination (317) that adds complicated hues and shades to
seemingly transparent cultural artefacts and concepts. The corpus analysis
that helps shape our own interpretations of suicide in Hamlet, we hope,
offers even more evidence for Shakespeares prismatic imagination
an imagination that includes linguistic discovery as well as literary and
historical inquiry. This chapter asks what happens when we wed literary,
cultural and linguistic analysis in order to prove that the rhetoric of sui-
cide in Hamlet at once retains the theological and legal implications of a
heinous sin while simultaneously registering an emergent discourse that
makes self-slaughter less a sin and more a private action with psychological
signifcation.
By acknowledging that Gods law (the fxed . . . canon [1.2.1312])
frmly prohibits suicide, Hamlet underscores that the church was the pri-
mary social force regulating self-murder. However, not only was Gods law
frmly against suicide, but also as the debate over Ophelias interment sug-
gests, Tudor and Stuart England had developed a legal system that posthu-
mously regulated acts of self-murder, determining which ones could legally
be considered criminal acts. MacDonald and Murphys (1990) account of
suicide in Renaissance England provides a detailed picture of the legal
attempts to control how suicide was understood. According to their study
of self-murder between 1485 and 1659, suicides in England peaked between
the years 1600 and 1610 around the time that Hamlet was frst performed
(30). Those who killed themselves in madness, as ruled by the coroners
jury, were judged to be innocent of a crime and could, therefore, receive
a Christian burial. A victim of self-murder was legally categorized as non
compos mentis, not of sound mind; a suicide, however, who the jury found
to be sane and capable of premeditating the act was categorized as felo de
se, a felon himself. Drawing on statistics culled from the Public Records
Offce, MacDonald and Murphy argue that those who committed suicide
were rarely exonerated for their crime: Over 95 per cent of the men and
women who killed themselves between 1485 and 1660 were convicted as
felones de se; fewer than 2 per cent were excused as persons non compos men-
tis. They conclude that the rigour with which the law against suicide was
enforced . . . distinguishes this period from the centuries before and after-
wards. Before the Early Modern period, laws against suicide were infre-
quently enforced, and after 1660, suicide was increasingly secularized and
decriminalized (16).
Despite the draconian enforcement of the laws governing suicide in
England, Hamlet captures what Williams (1977) might describe as an emer-
gent discourse about suicide that counters the dominant legal and religious
doctrines that attempt to regulate the action. Wymer (1986) has identi-
fed the emergence of this competing discourse as a response to the new
humanism that sought out valid ideals of human conduct (2) outside of
ossifying religious categories. The nascent discourse created a new ethic of
suicide that would not be fully articulate until the mid-seventeenth century,
but early evidence of this new ethic, for example, appears in John Donnes
Biathanatos (2000 [ca. 1608/ca. 1647]), with the subtitle, A Declaration of
that Paradox of Thesis, that Self-Homicide is not so naturally Sin that it may
never be otherwise. Targoff points out, however, that Biathanatos was by no
means an unqualifed endorsement of suicide (2006: 218). Indeed, even
within Donnes new ethic regarding self-murder, it was a suicide motivated
by the desire to embody Gods glory and not self- interest that he consid-
ered theologically and legally acceptable. For Targoff, what is critically new
in Donnes defence of the act is his insistence that the desire to end ones
life is not against the laws of nature (219). As an indication that the mean-
ing of self-murder was evolving in ways that Donne perhaps initiated in
his writings about death, Wymer argues that the term suicide, which frst
appears as a description of the act of self-murder in Sir Thomas Brownes
Religio Medici in 1635, is evidence of a new outlook that views suicide as
morally neutral. Wymer writes, Unlike the older terms, self-murder and
self-slaughter, [the term suicide] avoids the implication of violence and
criminality (2).
The character Hamlet, like his eponymous play, appears on the precipice
of this discursive shift.
2
Read within this cultural context, Hamlets most
famous contemplative moment, To be, or not to be, is also a comment
on the social, theological and legal forces that he knows will attempt to
defne him posthumously a concern that appears most pressing in Act 5,
after his death is assured. Concerned about his legacy in a society that can
legally control how his wounded name (2.286) is remembered, Hamlet
enjoins Horatio to tell my story (2.291). Upon hearing the warlike noise
(2.291) of Fortinbrass arrival, Hamlet implores his surrogate Horatio to give
Fortinbras his support and, along with the vote of confdence, the occurents,
more and less, / Which have solicited. The rest is silence (299300).
Hamlets concern that his dying voice (298) still has the force to defne his
legacy during the impending political transition underscores the inescap-
able reality that in death, he surrenders this ability. His assertion that the
rest is silence is a powerful rebuke of the theo-juridical forces that seek
posthumously to regulate the signifcance of Hamlets suicide by proxy in
Act 5.
Central to our comparison of the language of suicide and despair in the
play is the argument that Hamlets death is, in fact, self-slaughter what we
term suicide by proxy. In his analysis of Early Modern melancholy, Trevor
(2005) suggests that Early Modern scholars such as John Donne or the fc-
tional Hamlet fnd themselves in the curious . . . predicament of wanting to
be sad at the same time that they recognize that such sadness imperils their
lives . . . because it can prompt thought of self-annihilation (9). For Trevor,
the melancholics awareness that sadness can generate painful psycholog-
ical trauma points to a type of introspection that challenges the assump-
tion that, for instance, there is nothing to Hamlets interior. The depths of
Hamlets interiority might be measured, then, by the extent to which he
pursues a suicidal end within a culture that prohibits it. Hamlet acknowl-
edges as much in the plays frst act:
O that this too too solid fesh would melt,
Thaw, and resolve itself into a dew,
Or that the Everlasting had not fxed
His canon gainst self-slaughter! O God, O God,
(1.2.12932)
3
Hamlet expresses the tension between his desire to commit suicide that
his solid fesh would melt and the theological injunction against it. Yet,
by Act 5, Hamlets confusion has vanished. Suicide emerges as an option
within a performance the fencing match that will disguise his intentions
and effectively silence the regulatory power of the church and state, allow-
ing him to defne his legacy after death through Horatios retelling. Pollin
(1965) terms Hamlets death a successful suicide, noting that Hamlets
intellect, sensitivity, and moral and religious scruples caused him to seek
the least reproachable means of terminating an intolerable burden of exis-
tence (240). Pollin observes that Hamlet repeatedly and intentionally
exposes himself to peril: he returns to Denmarks hostile shore naked
(4.7.42), or unarmed and without followers. Hamlet also plays into the
hands (254) of his two adversaries by accepting the challenge from Laertes
of the fencing match, although Laertes has already tried to kill him in the
graveyard. Pollin notes that Hamlet refuses to postpone the match twice,
even as he admits to Horatio: thou wouldst not think how all here about
my heartbut it is no matter (5.2.14951). About Hamlets choosing
frst the rapier and dagger, Pollin concludes, In a sense Hamlet chooses
the instrument and Claudius through his henchman [Laertes] effects the
release which Hamlets moral and religious scruples prevent him from
grasping himself (254). Attaching surrogacy to the relationship between
Claudius and Laertes, Pollins conclusion insinuates that in the fnal fenc-
ing match the play enacts self-murder by proxy.
Far from being the nave dupe in the Claudius-Laertes plot to kill
him, as Pollin suggests, Hamlet willfully directs this play, as he does The
Mousetrap in Act 2. Where the earlier playlet exposes the conscience
of the King (2.2.582), Hamlets fnal production in Act 5 disguises his
own suicidal intentions and obscures his reservations about engaging
in self-murder. Suicide by proxy as Laertes ushers in Hamlets demise
with poison is the plays loophole that enables Hamlets inaugural self-
authorizing act, ironically requiring a second proxy: by telling Hamlets
story, Horatio arbitrates his friends legacy and effectively silences the
theo-political discourse against suicide by rendering it moot the rest is
silence.
Although the play does not explicitly show that Ophelias death is sui-
cide, there is ample evidence to suggest that she, too, commits self-slaugh-
ter. From the churchyard debate over her proper burial to her retreat
into insanity that presages her demise, the play hints that her death was
suicide. Gertrudes description of her death personifes her garments
(4.7.152), granting agency to the clothes that [p]ulled the poor wretch
from her melodious lay / To muddy death (1534). Her description that
redirects agency away from Ophelias will necessarily protects Ophelia
from the explicit charge of self-murder and allows for the possibility of a
Christian burial; however, before her drowning Ophelias concerns about
her own post-suicidal legacy attest to her contemplation of suicide. As she
gives various plants to several on-stage spectators, including her brother,
Ophelia offers rue among the many plants that have symbolic value to her:
Theres rue for you, and heres some for me. We may call it herb-grace
o Sundays. O, you must wear your rue with a difference (4.5.1779). In
her madness, Ophelia articulates the plays truth that during the cultural
moment of its production, suicide produces different types of regret:
Ophelia rues the prospect of her suicide through a theo-juridical lens
and Hamlet rues his through an emerging personal or psychological lens
that subverts theological and legal forms of regulation. Hamlet wants his
death to be remembered differently than he assumes the theo-juridical
mandates of the court will allow. Ophelia, however, appears resigned to
the fact that her suicide is subject to the condemnation that comes with
self-murder. Put otherwise, the representation of Ophelias apparent
suicide includes the threat of legal and theological condemnation, while
Hamlets death embodies a new cultural understanding of suicide that
avoids the stain of criminality and sin in Ophelias words, they rue with
a difference.
2. Methods
To demonstrate how a stylistic interpretation can enrich literary
approaches to Hamlet, we examined the discourse of Hamlet and of
Ophelia for lexico-semantic implications of differences in suicidal rhe-
toric using LIWC (Pennebaker et al. 2001), which provides an effcient
method for analysing the emotional, personal, communicative, sexual
and metaphysical components of language. We developed a corpus of dis-
courses for Hamlet and Ophelia that included pre-suicidal and suicidal
discourses as well as two control corpora selected from non-suicidal char-
acters (Horatio and Laertes). We predict that the discourse of Hamlet will
exhibit lexico- semantic patterns that beft modern narratives of suicide
(e.g. more personal pronouns, lack of communicative and social features),
while the discourse of Ophelia will exhibit more lexico-semantic patterns
that denote a rhetoric of religious sin (e.g. more plural pronouns, words
related to death, sex, religion).
3. Linguistic inquiry and word count (LIWC)
LIWC was developed as a tool to examine the cognitive, emotional and
structural processes found in writing and speech. The tool is a text analysis
program that uses word counts to search text for lexical and semantic fea-
tures that correlate with standard language categories (e.g. articles, prep-
ositions, pronouns), psychological processes (e.g. positive and negative
emotion words), relativity-related words (e.g. time, verb tense, spatiality)
and content dimensions (e.g. death, sex, home, occupation) (Pennebaker
et al. 2001). The theoretical underpinnings of the tool are premised on
the notion that words individuals use when writing and speaking provide
access to their emotional and cognitive states (Gottschalk and Gleser 1969,
Rosenberg and Tucker 1978, Stiles 1992). The words selected for the LIWC
categories are based on emotional rating scales, thesauri and dictionaries.
These words were then evaluated by independent judges and the words
were placed into aligned categories if agreement was met (ranging from
0.86 for the category optimism to 1.00 for relatives). The selected words
then came to form the LIWC database, which capture, on average, 80 per
cent of the words found in writing and speech.
The external validity of LIWC as a general measure of emotional and cog-
nitive lexical and semantic features was explored in the Pennebaker and
Francis (1999) study that examined the psychological ratings of frst-year
college student essays written about the affective toll of the college experi-
ence. These essays were frst evaluated by human judges in consideration
of the psychological, emotional and cognitive categories analysed by LIWC
and later by the LIWC tool. The evaluations made by the human judges and
LIWC were highly correlated.
LIWC has also been used to analyse suicidal vocabularies (Stirman and
Pennebaker 2001). In this study, the writing styles and characteristics
of writers who had committed suicide were compared to those that had
not with the prediction that suicidal writers would exhibit more words
of hopelessness in their writings than non-suicidal writers. The LIWC
tool was used to analyse 300 poems written by suicidal and non-suicidal
poets. Statistical analyses of the LIWC output supported the notion that
suicidal writers and non-suicidal writers differed in their use of words
indicating the desire for social integration and indicating a sense of
hopelessness. Specifcally, suicidal poets used more frst person pro-
nouns and words related to death and sexuality and less words related
to communication.
4. Corpus
To collect our corpus, we used Stephen Greenblatts Norton edition of
Hamlet (1997), considered a standard, scholarly edition, as well as a hyper-
textual version of the play from MITs edition of the complete works.
From these editions of Hamlet, six corpora were created. These were the
suicidal discourses for Hamlet and Ophelia, the non-suicidal discourses
for Hamlet and Ophelia, and the non-suicidal discourses of Horatio and
Laertes. We demarcated the suicidal from the non-suicidal discourses of
Hamlet and Ophelia based on the critical opinion that Ophelias fall into
despair in Act 4 is a prelude to her death by drowning (MacDonald 1986,
Wymer 1986) and that Hamlets own deliberation about his fate in Act
3 reveals the desire that self-murder governs his actions for the remain-
der of the play (Spinrad 2005). Ophelias discourse in Act 4, scene 5
was considered suicidal. Her discourse before this act was considered
non-suicidal. However, the number of lines assigned to Ophelia limited
the size of the analysis. Thus, her discourse was broken into 100-word seg-
ments to ensure enough data points were available for a statistical analy-
sis. These segments were then compared with 100-word segments taken
from Hamlets suicidal discourse in Act 5, scene 2 and non-suicidal dis-
course taken from Act 1, scene 2. For controls, we selected the beginning
scenes for the characters Horatio and Laertes, both non-suicidal charac-
ters. Descriptive statistics for the corpora can be found in Table 9.1. The
text segments for all the characters were then processed through the
LIWC tool and normalized lexico-semantic values were selected for lin-
guistic categories related to personal, communicative, sexual and meta-
physical lexical components.
In applying LIWCs modern database of words related to suicide, hope-
lessness, and estrangement to an Early Modern corpus, we made minor,
but important, alterations to the plays corpus. We made these alterations
to the corpora in order to align the language used in the original work to
that examined by LIWC. Principal among these was the modifcation of
pronouns (thee, thou, and thy) to the modern equivalents. In addition, the
word lord, when used to refer to an aristocrat, was changed to the aristo-
crats title (e.g. prince or king). This was to control for confounds within the
religious semantic category.
5. Statistical analysis
To compare for differences among the characters, the LIWC values were
analysed using Multivariate Analyses of Variance (MANOVA) and follow-up
Analyses of Variance (ANOVA) to compare individual differences. These
statistical analyses were conducted to demonstrate that differences existed
between Ophelias and Hamlets suicidal discourses and non-suicidal
discourses.
Table 9.1 Corpora descriptive statistics
Character Discourse Act. Scene Number of texts Average number of words
Hamlet Non-suicidal 1.2 7 113.00
Hamlet Suicidal 5.2 13 113.38
Ophelia Non-suicidal 1.32.1 5 119.80
Ophelia Suicidal 4.5 5 99.20
Horatio Non-suicidal 1.1 6 97.17
Laertes Non-suicidal 1.3 4 96.25
6. Analysis
6.1 Suicidal discourse: Hamlet and Ophelia
We predicted that the suicidal discourse of Hamlet would exhibit more lex-
ico-semantic patterns that beft modern narratives of suicide (e.g. personal
pronouns, less communicative functions), while the suicidal discourse of
Ophelia would exhibit more lexico-semantic patterns that denote a rhetoric
of religious sin (e.g. words related to death, sex and religion). The statistical
analyses support these predictions with the suicidal discourses of Hamlet
and Ophelia exhibiting signifcant difference: F(12,5) = 6.78, p < 0.05 (where
F represents the measure of distance between individual word choices, fol-
lowed by the degrees of freedom [between groups = 12 and within groups
= 5] and fnally by p, which is the confdence level of signifcant difference
Table 9.2 Mean, standard deviation, and F scores for LIWC values (Hamlets
versus Ophelias suicidal discourse)
Category Character Mean Standard deviation F Score
I Hamlet 6.09 1.98 6.58*
Ophelia 3.26 2.40
We Hamlet 1.14 1.96 0.09
Ophelia 1.41 0.56
Self Hamlet 7.23 3.10 1.84
Ophelia 5.05 2.92
You Hamlet 2.19 1.65 10.05*
Ophelia 4.96 1.70
Social Hamlet 10.66 4.54 7.88*
Ophelia 16.69 2.21
Communication Hamlet 1.13 0.67 6.05*
Ophelia 1.98 0.59
Family Hamlet 0.12 0.30 13.85**
Ophelia 0.80 0.45
Metaphysical Hamlet 0.82 0.85 9.93*
Ophelia 4.75 4.51
Religion Hamlet 0.60 0.73 7.70*
Ophelia 3.33 3.53
Death Hamlet 0.22 0.42 6.91*
Ophelia 1.42 1.58
Body Hamlet 1.14 0.92 2.79
Ophelia 0.40 0.55
Sexual Hamlet 0.15 0.36 5.48*
Ophelia 0.78 0.83
* p < 0.05 ** p < 0.001.
between two means). Specifcally, Hamlets suicidal discourse contains more
instances of words related to personal discourse and less to communicative
discourse. On the other hand, Ophelias suicidal discourse exhibits signif-
cantly more words related to religion, death and sexuality (see Table 9.2 for
additional information). The increased frequency of these words is evidence
that, at the linguistic level, Ophelia articulates the anxiety that suicide has
theo-juridical consequences common in Early Modern England.
The category metaphysical in Table 9.2 and in the other Tables below
is a superordinate category that includes within its subordinate categories
(religion, death, body and sexuality).
6.2 Suicidal and pre-suicidal discourse: Ophelia
When considering Ophelias pre-suicidal and suicidal discourses, no sig-
nifcant differences were noted in the MANOVA analysis F(9,1) = 4.83,
Table 9.3 Mean, standard deviation, and F scores for LIWC values (Ophelias
suicidal discourse versus non-suicidal discourse)
I Pre-suicidal 8.71 3.37 9.11*
Suicidal 3.26 2.40
We Pre-suicidal 0.17 0.42 18.14**
Suicidal 1.42 0.56
Self Pre-suicidal 8.88 3.43 5.68*
Suicidal 4.68 2.11
You Pre-suicidal 2.19 2.59 4.23
Suicidal 4.98 1.70
Social Pre-suicidal 12.43 5.10 3.07
Suicidal 16.76 2.24
Communication Pre-suicidal 0.50 0.55 18.46*
Suicidal 1.99 0.60
Family Pre-suicidal 1.84 2.64 0.74
Suicidal 0.81 0.45
Metaphysical Pre-suicidal 1.00 0.63 4.15
Suicidal 4.75 4.51
Religion Pre-suicidal 1.00 0.63 2.58
Suicidal 3.33 3.53
Death Pre-suicidal 0.00 0.00 4.99*
Suicidal 1.42 1.58
Body Pre-suicidal 3.01 2.13 6.98*
Suicidal 0.41 0.56
Sexual Pre-suicidal 0.00 0.00 5.55*
Suicidal 0.79 0.83
* p < 0.05 ** p < 0.001.
p > 0.05; however, the associated ANOVAs demonstrate that Ophelias
pre-suicidal language contains more personal discourse and less occur-
rences of words related to death, body and sexuality (see Table 9.3 for
additional information). This subtle variation in her pre-suicidal and
suicidal rhetoric shows that Ophelias concerns about death and sexual-
ity emerge as signifcant once her suicide becomes her primary motiva-
tion. As we will discuss in more detail below, our prediction that Ophelia
would use more words pertaining to the body in her suicidal language
was not statistically signifcant. While the mean value for religious words
was higher in Ophelias suicidal discourse (3.33) as compared with her
pre-suicidal discourse (1.00), the differences only approached signif-
cance (p =.063). The mean value for words related to the body was sig-
nifcantly higher in pre-suicidal discourse (3.01) than suicidal (0.41).
Rhetoric about the body may indeed emerge as meaningful using
non-statistical analysis; however, this meaningfulness is demonstrated
through metaphorical analysis, which proves diffcult for computational
tools like LIWC.
6.3 Suicidal and pre-suicidal discourse: Hamlet
When Hamlets pre-suicidal discourse is compared to his suicidal discourse,
signifcant fndings do emerge, F(14, 5) = 5.77, p < 0.05, with Hamlets earlier
discourse containing more incidences of words relating to the metaphysical
and the body (see Table 9.4 for additional information). This supports the
assumption that the change registered in Hamlets suicidal rhetoric corre-
sponds to an emergent cultural understanding of suicide that subverts the
formerly dominant model registered in Ophelias theo-juridical concerns
about the fate of her body and soul.
6.4 Suicidal and pre-suicidal discourse: Ophelia and Hamlet to controls
Comparisons between pre-suicidal Hamlet and Ophelia to the controls,
Horatio and Laertes, demonstrate no signifcant differences in word choices,
F(15,5) = 1.32, p > 0.05, while comparisons between suicidal Hamlet and
Ophelia to the controls exhibited signifcant differences in word choices,
F(15,12) = 9.57, p <0.001. These fndings support the notion that Ophelias
and Hamlets non-suicidal discourse is no different from that of other non-
suicidal characters, while there is a marked difference between their sui-
cidal discourse and that of the non-suicidal controls.
7. Discussion
The fndings from the lexical analysis of the rhetoric of suicide suggest that
both Ophelia and Hamlet contemplate different meanings of suicide prior
to their deaths. This analysis demonstrates that there is a difference between
how their deaths are described and strongly suggests that the play consid-
ers the two suicides differently with contrasting cultural valences. Further,
by contrasting their suicidal and non-suicidal discourse to that of Laertes
and Horatio, two non-suicidal characters, the fndings of this investigation
exhibit that the differences in rhetoric between Hamlet and Ophelia are
most likely the result of suicidal considerations.
Because Hamlet commits a type of suicide by proxy, literary critics have
been slow to notice how his suicidal discourse works in relation to Ophelias.
In looking closely, however, at the plays fnal act with the language of suicide
Table 9.4 Mean, standard deviation, and F scores for LIWC values (Hamlets
suicidal discourse versus non-suicidal discourse)
I Pre-suicidal 7.41 3.32 1.25
Suicidal 6.09 1.98
We Pre-suicidal 0.26 0.44 1.35
Suicidal 1.14 1.96
Self Pre-suicidal 7.67 3.36 0.08
Suicidal 7.23 3.10
You Pre-suicidal 3.75 3.08 2.21
Suicidal 2.19 1.65
Social Pre-suicidal 10.58 4.82 0.00
suicidal 10.66 4.54
Communication Pre-suicidal 1.45 1.62 0.39
Suicidal 1.13 0.67
Family Pre-suicidal 1.29 0.86 20.03**
Suicidal 0.12 0.30
Metaphysical Pre-suicidal 1.88 0.78 7.53*
Suicidal 0.82 0.85
Religion Pre-suicidal 1.38 1.02 4.01*
Suicidal 0.60 0.73
Death Pre-suicidal 0.51 0.48 1.90
Suicidal 0.22 0.42
Body Pre-suicidal 2.30 1.04 6.68*
Suicidal 1.14 0.92
Sexual Pre-suicidal 0.48 0.63 2.46
Suicidal 0.14 0.34
*p < 0.05 ** p < 0.001.
framing our analysis, an image of a character emerges for whom suicide is
not a religious nor even a legal act, but one more personal and psycholog-
ical. While Hamlet expresses dismay at the religious prohibition on suicide
in Act 1, wishing that God had not fxed canon law against self-murder, his
subsequent language when compared to Ophelias, once they both commit
to suicidal action, reveals two important discoveries: for her, the ethics of
the act of suicide are mired in traditional regulatory signifcance; while for
him, the consequence of suicide is no longer subject to posthumous theo-
juridical control.
When the difference in Ophelias suicidal rhetoric is compared with the
difference in Hamlets suicidal rhetoric, a statistical fnding emerges that
demonstrates that the difference in the language between the two charac-
ters is the result of suicidal tendencies and their cultural, legal and theo-
logical implications. I the pronoun most associated with proclamations
of self-identity appears signifcantly more often in Hamlets suicidal dis-
course when compared to Ophelias. However, the pronoun appears sig-
nifcantly more often in Ophelias pre-suicidal discourse when compared
to her suicidal discourse. This fnding is important if understood in context
of how thoughts of suicide affect the different characters. Ophelias selfsh
concerns correspond with a greater frequency in the use of I only when
her desires are defned by earthy pleasures such as her relationship with
her lover Hamlet or her domineering brother Laertes. In expressing these
concerns, Ophelia resembles Hamlets psychologically self-centred attitude
that only gets more pronounced over the course of the play. Where Ophelia
turns away from expressions of selfhood refected in the use of the pronoun
I as she considers the ramifcations of suicide, Hamlet solidifes his con-
cern for self, and in comparison to Ophelia, his sustained use of I is sta-
tistically signifcant in relation to Ophelias diminished use of the pronoun
during her suicidal discourse.
Equally as important in the same comparison, statistically signifcant fnd-
ings emerge when we look at the use of words that express relationships to
family, the metaphysical, religion, death, the body and sex. In each of these
categories the mean is higher in a way that indicates the critical impact
that the suicidal tendencies of both characters have on the play. Hamlets
references to family are fewer in his suicidal condition than in Ophelias
suicidal rhetoric. This suggests that Hamlets solipsism was already well-
developed, resulting from his narcissistic grief due to the loss of his father,
while Ophelias relationship to family only became an issue once suicide
became an option. In each of the other categories metaphysical, religion,
death and sexual references to key indicators demonstrating her abiding
concerns are signifcantly higher in Ophelias suicidal discourse, suggesting
again that her anxieties about how her suicide will affect her within existing
social networks are profound when compared to Hamlets fading concerns
about those same consequences.
A computational analysis of Ophelias language before Act 4, scene 5
demonstrates that she uses pronouns such as I, and self, that are most
associated with a self-centred image. The fndings represent a signifcant
difference when compared to her use of the collective pronoun we,
and you which become more frequent
4
in the discourse of suicide that
emerges in Act 4, scene 5. Likewise, Ophelias language reveals a signif-
cant new interest in death and religion, as words such as God, shroud,
death-bed, died and grave appear with greater frequency in her sui-
cidal discourse. The increased frequency of these words associated with the
religious consequences of death is evidence that at the linguistic level, she,
perhaps covertly, articulates her anxiety that her suicide has a theo-juridical
effect; her concern with how her suicide might affect others is registered in
the statistically signifcant increase in the use of words indicating her desire
to connect verbally: words enacting the desire to engage in some form of
human contact such as mark, come and counsel appear more often in
the suicidal discourse. Moreover, Ophelias concern over the status of her
virginity registered in increased sexual discourse is amplifed in the sui-
cidal discourse.
The statistical analysis of Ophelias rhetoric for the body (suicidal versus
pre-suicidal) did not match our predictions; however, the central image
memorializing her death Gertrudes announcement in Act 4 creates the
impression that Ophelias body has indeed become a privileged signifer
that articulates her suicidal desires. This scene, when considered outside
of our stylistic analysis of the language of suicide in the play, offers compel-
ling evidence that Ophelias body becomes a contested site on which are
written anxieties over self-murder. Indeed, the madness that drives Ophelia
to suicide is written in her sexuality, which is described as a document
in madness (4.5.175) and articulated in bodily rhetoric. Salkeld describes
Ophelias new relationship to her own body, which is on public display
and like her speech is now fgural and promiscuous (1993: 95). He con-
tinues, No longer closeted and sewing, passively obedient to the men who
owned and subjected her, she roams the palace grounds. Ophelia is fol-
lowed because no one dare touch her (95). Her physical availability dur-
ing this suicidal phase in Act 4 mirrors the fndings that emerge from a
linguistic analysis of the scene that shows her concern over sexual fate to
be increasing due to the theological and legal consequences of suicide.
Although the fndings from the body category do not support this read-
ing, the results from the sexuality analysis confrm her increasing concern
over her bodys fate. For Ophelia, Gertrudes famous description of her
drowned body in the willowy brook enacts the theological and religious
verdict that dooms her:
Therewith fantastic garlands did she make
Of crow-fowers, nettles, daisies, and long purples,
That liberal shepherds give a grosser name,
. . .
There on the pendent boughs her crownet weeds
Clambring to hang, an envious sliver broke,
When down the weedy trophies and herself
Fell in the weeping brook. Her clothes spread wide,
And mermaid-like a while they bore her up;
Which time she chanted snatches of old tunes,
As one incapable of her own distress,
Or like a creature native and endued
Unto that element. But long it could not be
Till that her garments, heavy with their drink,
Pulled the poor wretch from her melodious lay
To muddy death.
(4.7.13941/14354)
Her body, now reduced to pure discourse after her suicide, is erased
from the stage and rendered immutable in Gertrudes description that
immortalizes her.
5
Gertrudes description of her body functions like a bla-
zon a literary convention with metaphorical and allusive signifcance: its
meaning in Hamlet is contingent on its relationship to the poetic device
found in Petrarch and imitated in works from many Early Modern poets
and playwrights.
6
A statistical account of Ophelias suicidal, bodily rhe-
toric does not register the impact of poetic devices such as the blazon to
impart meaning through intertextual metaphor and poetic convention
7

and therefore presents one problem in trying to wed literal interpreta-
tion supported by corpus analysis with the metaphorical that characterizes
traditional literary analysis. In her pre-suicidal discourse, the more fre-
quent references to her body suggest that Ophelia expressed latent sexual
desires by talking about desire through the body, but hiding it implicitly
within her words. When she appears to be suicidal, sexual desire is not
longer as latent; instead, the play expresses it in the metaphorical through
logical semantic extensions such as Gertrudes description that echoes the
desire captured in the logic of the blazon. As a moment that articulates
diachronic literary history as much as contemporary, synchronic social
history, Gertrudes description of Ophelias fate is the equivalent of artis-
tic purgatory, freezing her in poetic language and literary history that at
the same time executes the cultural verdict condemning her [t]o muddy
death (4.7.154) for the offence against God, against the king, and against
Nature (Dalton: 234).
Registered in Gertrudes announcement of Ophelias seemingly staged
death at the brook, the concern over her body, as well as the increased
frequency of words that reveal a desire for engagement, suggest Ophelias
covert awareness of the cost of suicide in a culture that regulates its meaning
posthumously. The status of the soul and body of the dead was already an
open question in a post-reformation culture that had shifted from Catholic
funerary rituals that employed the body in practices designed to help the
dead get to heaven. In unreformed doctrine, the dead, in fact, were con-
sidered an age group ready to receive the souls of other subjects who had
been properly buried. Registered in her increased use of words that suggest
her desire for an interlocutor and in language that expresses anxiety about
the status of her body, Ophelias suicide jeopardizes her admission into
the community of the dead. Claiming that her funeral rituals have been
as far enlarged / As we have warrantise (5.1.2089) because her death
was doubtful (5.1.209), the priest truncates the funeral ritual; according
to him, executing the full ritual would profane the service of the dead
(5.1.219). Hamlets reaction to Ophelias funeral procession contextualizes
the scene: who is that they follow, / And with such maimd rites? This doth
betoken / The corpse they follow did with desprate hand / Fordo it own
life (5.1.2014). Laertess reaction to the scene emphasizes Ophelias body
and returns us to her own covert anxieties about the status of her body
after her suicide. If it cannot be exalted into the community of the dead,
the body lingers as an object on which men may gaze. Over Ophelias dead
body, Laertes says, Lay her ithearth, / And from her fair and unpolluted
fesh / May violets spring (2213). Hamlets use of the word corpse to
describe Ophelias body in procession and Laertess emphasis on her body
as fesh only underscore her fate under theo-juridical regulation that des-
ecrates the body while simultaneously criminalizing the suicide. Although
not supported by the data, the plays concern about Ophelias body reg-
istered in poetic convention and in other characters language hints at
the polyvocality of discourse that poses a challenge to a purely linguistic
interpretation of the play. This moment of friction between the data and
what Williams (1977) has named a structure of feeling underscores the
need for a fexible collaboration between corpus analysis and literary inter-
pretation that pursues the hermeneutic possibilities in these moments of
methodological confict.
When we perform a linguistic account of Hamlets pre-suicidal rhetoric
and compare it to his suicidal rhetoric, we begin to see that the plays atti-
tude about suicide changes as Hamlet begins to defne the act more per-
sonally, without the regards for community and the fate of his soul that
characterize Ophelias suicidal discourse. This observation about what
can be gleaned through a comparison of their deaths highlights what
MacDonald concludes about the plays depiction of the shifting cultural
attitudes towards suicide taking place during the time of its performance.
He writes that the play exploits changing attitudes to suicide that began
during [Shakespeares] lifetime. [Shakespeare] presents current interpre-
tations of suicide, but does not resolve the contradictions among them
(1986: 317). Before Hamlet commits to his suicidal impulses in his Act 3,
scene 1 (To be, or not to be), linguistic analysis shows that he speaks of reli-
gious issues such as God and church with signifcantly more frequency,
and he speaks of his body in statistically signifcant more detail. His use
of words associated with the family such as mom, brother and cousin
are signifcantly greater as well. These fndings suggest that as Hamlet con-
templates suicide, he moves away from traditional cultural and theological
concerns about suicide. While he begins the play expressing his dismay that
God forbids self-slaughter, the signifcance of this injunction weakens as he
begins to articulate a new relationship to suicide that is more personal.
In Hamlets pre-suicidal discourse, we observe signifcantly more words
that suggest a desire for community, family and faith than we do in his
suicidal discourse. Likewise, Hamlet uses signifcantly more lexical indica-
tors that demonstrate concern over the metaphysical and religion. Both
of these observations reverse the fndings that the frequency of words that
connote an abiding interest in church, metaphysics and family appear more
frequently in Ophelias suicidal discourse as compared to her non-suicidal
discourse. Hamlet also uses words pertaining to the body words such as
heart, hand, sinews, bones more frequently in his pre-suicidal dis-
course. Like Ophelia, whose body becomes a site of concern when the play
metaphorizes it as being subject to theo-juridical discourse that regulates its
fate after her death, Hamlet expresses concerns about the body before he
decides to commit suicide. Hamlets discussion with Horatio about the fate
of a corpse in Act 4 points to this change in attitude about the body that
contrasts his pre-suicidal rhetoric with his suicidal language about the body:
To what base uses we may return, Horatio! Why may not imagination trace
the noble dust of Alexander till a fnd it stopping a bung-hole?(5.1.1879).
He continues after Horatio chides him for thinking too much on the fate of
the body: as thus: Alexander died, Alexander was buried, Alexander retur-
neth into dust, the dust is earth, of earth we make loam, and why of that
loam whereto he was converted might they not stop a beer-barrel? (1925).
Hamlets language in the graveyard is unconcerned with metaphysical or
theological consequences of death; instead he describes a natural, earth-
bound fate for the body. Comparing this to Hamlets concern about the
fate of the body earlier in the play, a body that grunt[s] and sweat[s] under
a weary life, / But that the dread of something after death, / The undis-
covered country from whose bourn / No traveller returns, puzzles the
will (3.1.7982), reveals a shift in how he understands deaths affect on
the body. This juxtaposition of the language associated with death and the
body suggests that Ophelias suicide refects the prevailing ideologies that
govern self-slaughter, while Hamlets suicide refects a new set of concerns
not linked as fervently to dominant religious and legal regulations that gov-
ern the fate of the suicides corpse. If, as we suggested in our introduction,
suicide is the plays motivating force, then how do the non-suicidal charac-
ters provide evidence for this claim? An analysis of the language of Laertes
and Horatio in relation to the same categories that we used to evaluate the
rhetoric of suicide of Hamlet and Ophelia proves interesting. The fndings
show that when compared with the plays non-suicidal characters, the con-
trast between Hamlets and Ophelias use of key words demonstrating affect
and preoccupations with family, community, religion, death and the body
does not signifcantly differ from how these concerns manifest themselves
in Horatios and Laertess language throughout the play. The differences
become signifcant once suicidal tendencies begin to preoccupy Hamlets
and Ophelias thoughts. These results generally support that idea that sui-
cide indeed becomes the crux around which characters in Hamlet begin
to differentiate themselves from each other and negotiate differently the
shifting legal, religious and personal regulations governing suicide during
the early-seventeenth century.
8. Conclusion
Our stylistic analysis of the rhetoric of suicide in Hamlet provides further
evidence for the claim that self-murder was an action fraught with new, at
times contradictory, signifcance during the early-seventeenth century in
England. Amplifying the historical and cultural scholarship on Renaissance
suicide from MacDonald and Terrance and Wymer, our study of Hamlet
underscores the plays importance as an artefact that registers an emergent
cultural formation, one that provides a challenge to dominant perspectives
and, more importantly, to the institutions that turn these perspectives into
material practices social institutions such as the coroners jury, the church
and the family. Beyond the scholarly benefts that linguistic analysis may
have in shedding new light on an old problem in Hamlet, corpora in con-
junction with cultural and literary analyses offer an investigative approach
that quantifes (and qualifes) what Stephen Greenblatt (1990a, 1990b) for-
merly called cultural poetics, or the circulation of social energy, which even
in the best readings of Early Modern literature at times remains abstract
and ephemeral.
8
When faced with an impasse between the critical desire to interpret
Ophelias concern about her body as a sign of her anxiety over the legal
and religious consequences of her suicide and the linguistic data that chal-
lenge the presence of this desire, the collaboration between computational
analysis and post-structural literary criticism does not provide a way to rec-
oncile this friction. In addition to the limited sample size for Ophelias
suicidal discourse a common limitation of a corpus analysis of any single
literary work a linguistic analysis of Ophelias rhetoric of suicide may not
account for the proliferation of diachronic discourses related to represen-
tations of the body and expressed through literary conventions such as the
blazon. The lexicon related to the body in this descriptive device, while not
signifcant as an indicator of the plays concern about Ophelias fate if she
were to commit suicide, contributes to expressions of sexual tensions that
are signifcant in the data. The plays interest in Ophelias embodied desire,
most often expressed in other characters language, is perhaps an indica-
tion of a structure of feeling about self-slaughter during the Renaissance
that Williams (1977) has identifed as a texts primary effect and that may
not appear in the empirical data.
Moments of literary convention notwithstanding, students of Shakespeare
believe that words do matter. As Butlers (1997) infuence in literary studies
over the past decade suggests, words, indeed, perform action as much as
they reveal or represent it. Thus taking the language of a play such as Hamlet
seriously by drawing conclusions from evidence that shows it participates
in larger patterns of language culled from sources beyond its own textual
boundaries is perhaps a logical supplement to current critical approaches
that, at some risk, tend to return to the past in order to fll the vacuum in
the study of literature after the post-theoretical and post-historical turn.
By paying attention to the lightest word (1.5.15), as the ghost of Hamlets
father suggests during his encounter with his son in Act 1, we can perhaps
more fully begin to unlock the secrets of [the] prison-house (1.5.14) of
language with the hope of proving that words do matter because they are
the matter of the forensics of the study of literature.
Notes
1
Critics who have explored the subject of suicide in Hamlet include Pollin (1965),
MacDonald (1986), Greenblatt (2001), Spinrad (2005), and Trevor (2005). We
recognize that the play does not explicitly represent the deaths of Hamlet or
Ophelia as suicides; our aim in this essay, however, is to demonstrate how a stylis-
tic analysis of the plays language suggests that indeed their deaths have much in
common with suicides in the Early Modern period.
2
Other writers who consider the implications of suicide during this time include De
Montaigne (1910 [1580], 88, 92, translated into English, 1603), specifcally vol. 2,
279, 345, 41; and Sym (1637). For helpful excerpts from these sources, see Jor-
dan (2005).
3
The textual problem of the Folios use of solid as opposed to Q2s use of sallied
as a possible misspelling of sullied is interesting in relation to the question of self-
murder and its consequences. If Q2 is correct and sullied meaning, begrimed
with soot and dirt is the intended adjective to describe Hamlets fesh, he is
perhaps acknowledging explicitly that the threat of legal and religious judgement
looms. The repetition of too suggests that at this point in the play, the threat of
theo-juridical condemnation is too severe for him to embrace suicide. Hamlets
concern over his fesh as sullied or solid echoes Ophelias own rhetoric
about the status of her body during her suicidal discourse in Act 4.
4
Although the statistical fndings for you are not signifcant, the mean follows the
expected pattern.
5
Schiesari (1992) has made a similar point that the play effaces Ophelia from its
narrative. She suggests that Ophelias essential role in Hamlets trajectory par-
adoxically depends on her effacement, obliteration or rejection (260).
6
Our use of the word blazon to describe Gertrudes poetic description of Ophe-
lias drowning extends Patricia Parkers analysis (1993) of the theme of dilation
and spying in relation to the female body in Hamlet. Her argument with regard
to Ophelia suggests that the erotics of Gertrudes description hinge on the
aucular performance of the word clothes (4.7.173), which for Parker, echoes
the word close and connotes sexual desire, if not enfranchisement: Ophe-
lias close or clothes spread wide display or open to the view what . . . should
more modestly be hid. The spreading wide of a close . . . hints of the sexual
opening and closing in the double entendres of her songs, as in the cull-cold
maid do dead mens fngers call (4.7.16971) (75). Philippa Berrys alternative
interpretation of Gertrudes discourse in Shakespeares Feminine Endings (1999)
provides a second account of the embedded eroticism in the passage: The
death of Ophelia, as narrated by Gertrude, affords a striking instance of the
tragedies fgurative entwining of the complex duality of womans erotic dying,
both with popular festivity, and also with an animating principle within nature
which appears to parallel ideas of the world soul. For as nature itself becomes
her lover, the deranged young womans muddy death is troped as . . . strangely
pleasurable . . . (26). Berry goes on to discuss the passages bawdy iconography
(27). In using the term blazon to capture the form and effect of Gertrudes
eulogy, our intent is to link what Berry describes as fgurative eroticism to the
logic of the blazon, a poetic device with a specifc history that is diffcult for cor-
pus analysis to register.
7
See Vickers (1985) for an important discussion of the blazon in Shakespeares
poetry.
8
For a critical assessment of new historicism, see Pechter (1987) and Kezar
(2003).
References
Shakespeare edition used: Hamlet, Prince of Denmark in: Stephen Greenblatt,
Walter Cohen, Jean E. Howard and Katharine Eisaman Maus (eds) 1997.
The Norton Shakespeare: Based on the Oxford Edition (16961784). New York:
W.W. Norton & Company.
Berry, Phillipa 1999. Shakespeares Feminine Endings: Disfguring Death in the Tragedies.
New York: Routledge.
Butler, Judith 1997. Excitable Speech: A Politics of the Performative. New York:
Routledge.
Dalton, Michael 1626. The Countrey Justice. London. N.P.
De Montaigne, Michel 1910 [1580]. The Essays of Michael Lord of Montaigne. London:
Dent.
Donne, John 2000 [ca. 1608/ca.1647]. Biathanatos, in: John Carey (ed.), John Donne:
The Major Works. Oxford: Oxford University Press, 14950.
Durkheim, Emile 1951. Suicide. New York: Free Press.
Gottschalk, Louis A. and Goldine C. Gleser 1969. The Measurement of Psychological
States through the Content Analysis of Verbal Behavior. Berkeley: University of
California Press.
Greenblatt, Stephen 1990a. Shakespearean Negotiations: The Circulation of Social Energy
in Renaissance England. Oxford: Clarendon Press.
Greenblatt, Stephen 1990b. Towards a poetics of culture, in: Learning to Curse:
Essays in Early Modern Culture. London: Routledge, 14660.
Greenblatt, Stephen 2001. Hamlet in Purgatory. Princeton: Princeton University
Press.
Hoover, David 2007. The end of the irrelevant text: electronic texts, linguistics, and
literary theory, Digital Humanities Quarterly 1(2).
Jordan, Constance 2005. Hamlet: The Longman Cultural Edition. Boston: Longman.
Kezar, Dennis 2003. Shakespeares addictions, Critical Inquiry 30(1): 3162.
MacDonald, Michael 1986. Ophelias maimed rites, Shakespeare Quarterly 37(3):
30917.
MacDonald, Michael and Terence R. Murphy 1990. Sleepless Souls: Suicide in Early
Modern England. Oxford: Clarendon Press.
Parker, Patricia 1993. Othello and Hamlet: dilation, spying, and the secret place
of woman, Representations 44(1): 6095.
Pechter, Edward 1987. The New Historicism and its discontents: politicizing
Renaissance drama, PMLA 102: 293303.
Pennebaker, James W. and Martha E. Francis 1999. Linguistic Inquiry and Word Count
(LIWC). Mahwah, NJ: Erlbaum.
Pennebaker, James W., Martha E. Francis and Roger J. Booth 2001. Linguistic Inquiry
and Word Count (LIWC): LIWC2001. Mahwah: Lawrence Erlbaum Associates.
Pollin, B. R. 1965. Hamlet, a successful suicide, Shakespeare Studies 1: 24060.
Prezant, Daniel W. and Robert A. Neimeyer 1988. Cognitive predictors of depres-
sion and suicide ideation, Suicide and Life-Threatening Behavior 18(3): 25964.
Rosenberg, Stanley D. and Gary J. Tucker 1978. Verbal behavior and schizophrenia:
the semantic dimension, Archives of General Psychiatry 36: 13317.
Salkeld, Duncan 1993. Madness and Drama in the Age of Shakespeare. Manchester:
Manchester University Press.
Schiesari, Juliana 1992. The Gendering of Melancholia: Feminism, Psychoanalysis, and the
Symbolics of Loss in Renaissance Literature. Ithaca: Cornell University Press.
Shneidman, Edwin S. 1996. The Suicidal Mind. New York: Oxford University Press.
Spinrad, Phoebe S. 2005. The fall of the sparrow and the map of Hamlets mind,
Modern Philology 102(4): 45377.
Stiles, William B. 1992. Describing Talk: A Taxonomy of Verbal Response Modes. Newbury
Park, CA: Sage.
Stirman, S. W. and Pennebaker, J. W. 2001. Word use in the poetry of suicidal and
non-suicidal poets, Psychosomatic Medicine 63: 51722.
Sym, John 1637. Lifes Preservative Against Self-Killing. London: Flesher.
Targoff, Ramie 2006. Facing death in: Achsah Guibbory (ed.), The Cambridge
Companion to John Donne. Cambridge: Cambridge University Press, 21732.
Trevor, Douglas 2005 [2004]. The Poetics of Melancholy in Early Modern England.
Vickers, Nancy 1985. The blazon of sweet beautys breast: Shakespeares Lucrece,
in: Patricia Parker and Geoffrey Hartman (eds), Shakespeare and the Question of
Theory. Methuen, MA: Routledge, 95115.
Williams, Raymond 1977. Marxism and Literature. Oxford: Oxford University Press.
Wymer, Rowland 1986. Suicide and Despair in the Jacobean Drama. New York: St Martins
Press.
Chapter 10
Shakespeares Sexual Language and
Metaphor: a Cognitive-Stylistic Approach
Jos L. Oncins-Martnez
1
1. Introduction
It has long been acknowledged that sex and sexual allusions constitute a
major component of Shakespeares works. His plays and poems are full of
sexual references, not only expressed overtly but also in the form of all
types of innuendo, double entendre and punning. Such an important ele-
ment as Shakespeares sexual language has attracted the attention of quite
a number of critics and lexicographers over the last fve decades.
2
Among
the latter Partridge stands out as a pioneer. His Shakespeares Bawdy (2001
[1947]) written at a time when, as Wells (viii) reminds us, attitudes to
expressions of sexuality were far less liberal than they were to become dur-
ing the 1960s opened up a new avenue of research and was followed
by the dictionaries and glossaries of Colman (1974), Rubinstein (1984) or
Williams (1997), to name just a few.
A quick glance at these lexicographical works reveals that a large amount
of Shakespeares sexual vocabulary tends to cluster around certain seman-
tic felds, such as war, agriculture, food, trade and commerce. These clus-
ters have traditionally been grouped and dealt with under the somewhat
vague label of sexual imagery
3
. Partridge himself classifes what he likewise
calls Shakespeares sexual imagery into some 20 categories which provide
a very useful point of departure for the analysis of Shakespeares texts.
Nevertheless, he does not explore this taxonomy, which includes, among
other categories of imagery, music, trade, commerce, sport, horsemanship,
hunting, warfare, geography and agriculture. Thus, Partridge fails to dis-
tinguish the metaphoric from the non-metaphoric in the sexual images he
identifes, and he, like other authors, may frequently employ the word met-
aphor in an unsystematic and random manner, often using the term inter-
changeably with image cluster or imagery as if the three described or
referred to the same phenomenon.
This terminological confusion in imagery studies has often been pointed
out as the main weakness of an approach generally criticized for its lack
of methodological rigour. Much of this criticism, especially in the 1980s
and 1990s, came from modern theories of metaphor, as these were consid-
ered to be more suitable to explain imagery. In this regard, Thompson
and Thompsons (1987) Shakespeare: Meaning and Metaphor a multi-
theoretical reading of Shakespeares metaphors represents a turning
point in Shakespearean scholarship. Written as a challenge to the imag-
ery approach
4
, the book presents new ways of approaching metaphor in
Shakespeare and demonstrates through several case studies that modern
metaphor theories certainly provide reliable frameworks for the descrip-
tion and analysis of Shakespeares imagery. It includes among its chapters
a study of time metaphors in Troilus and Cressida, one of the frst systematic
applications of Metaphors We Live By, Lakoff and Johnsons (1980) seminal
text of Conceptual (or Cognitive) Metaphor Theory (henceforth CMT)
5
.
Also within the framework of CMT and applied to the analysis of some
Shakespearean metaphors of romantic and erotic love are Barcelona (1995)
and Snchez-Garca (2003). Although dealing with Shakespeare specif-
cally, their focus on love situates them within a major line of research in
CMT that considers the role of metaphor in the conceptualization of emo-
tions in general. Within this line, other love-related emotions such as lust or
sexual desire have been dealt with extensively (see Lakoff 1987, Emanatian
1995, 1996, Deignan 1997 or Kyratzis 2007). However, little has been said
so far about other matters pertaining to love, such as sexual activity or the
sexual organs, inextricably linked to this emotion. Pfaff, Gibbs and Johnson
(1997) and Chamizo-Domnguez and Snchez-Benedito (2000) are among
the few exceptions worthy of mention. The former reports on the role of
metaphors in the use and understanding of euphemisms and offensive
expressions, and includes some metaphors of sexual desire for example,
sexual desire is an activated machine or sexual desire is a hunting
animal along with others corresponding to the sexual organs and the act
of copulation, like a vagina is a small container for valuable objects,
a penis is a long slender weapon, or sexual intercourse is a coop-
erative dance. The work of Chamizo-Domnguez and Snchez-Benedito
concentrates exclusively on the physical aspects of sex, exploring a wide
range of metaphorical expressions for copulation and the sexual organs in
different conceptual domains and metaphors.
These studies, together with others in the feld of the cognitive sciences,
have contributed to the development of a relatively new branch of stylistics
variously known as cognitive stylistics or cognitive poetics a feld at the
Shakespeares Sexual Language and Metaphor 217
interface between linguistics, literary studies and cognitive science geared
to a better understanding and appreciation of the literary text
6
. As Semino
and Culpeper explain in their introduction to Cognitive Stylistics (2002: ix),
this discipline combines
the kind of explicit, rigorous and detailed linguistic analysis of literary
texts that is typical of the stylistics tradition with a systematic and theoret-
ically informed consideration of the cognitive structures and processes
that underlie the production and reception of language.
Freeman(2002: 321), the author of several cognitive-stylistic readings
of Shakespeare, also points out the advantages of such an approach when
applied to studying imagery in Shakespeares plays as it makes it possible
to integrate hitherto disparate and separate literary analyses, explaining
not only what used to be called the imagery of dramatic poetry, but aspects
of its characterization, plot structure, stage business, even its stage props,
as well.
Needless to say, neither stylistics in general nor cognitive stylistics in
particular is trying to come up with new interpretations but rather trying
to fnd systematic patterns that may account for interpretation. Analyses
like the ones proposed by Thompson and Thompson (1987) or Freeman
(2004 [1995]) testify to the applicability of CMT and its potential for
revealing the underlying patterns that hold these images together, there-
fore providing a useful framework for describing the metaphors of sexual
language as well. This does not mean, of course, that CMT is here consid-
ered to be the panacea for the analysis of this type of language in partic-
ular or style in general, some sort of killer approach to use Bradshaws
(2004: xv) expression that can triumph over all others
7
. Neither does the
focus on metaphor imply that all of Shakespeares sexual language can be
explained in terms of conceptual metaphors, for although metaphor is a
major means whereby Shakespeare expresses sexual matters, it is not the
only one.
2. Cognitive stylistics and Shakespeares sexual language
Even though it is true that recent CMT-based works on Shakespeare have
opened new vistas for the study of his style, those dedicated to the enor-
mous amount of bawdy that pervades his texts are still very rare. Then, if
as Colman (1974: x) points out, bawdy is one of the most potent weapons
in [Shakespeares] dramatic armoury, an exploration of this kind of lan-
guage from the standpoint of CMT seems worthwhile. And this is the aim
of the present analysis of Shakespeares war-sex metaphors. Taking as its
starting point his sexual vocabulary compiled in the aforementioned dic-
tionaries and glossaries, it seeks to demonstrate that this type of lexicon
can be explained and more fully appreciated in the light of CMT; and
that such an analysis may shed light on aspects of these and other meta-
phors that bear directly on the aesthetic quality of the texts in which they
appear, thus revealing the stylistic potential of Shakespeares bawdy.
Hence, several aspects of metaphor dealt with in CMT such as the dis-
tinction between image-schemas and specifc-level or basic metaphors, or
phenomena such as highlighting and hiding, or even concepts such as
metaphorical utilization or metaphorical scope, will be shown in the fol-
lowing pages to be useful instruments for understanding and explaining
some of Shakespeares favourite stylistic devices and their aesthetic and
dramatic effects.
In order to illustrate the explanatory potential of this model, this essay
focuses on one of Shakespeares most productive metaphors, in which sex-
ual activity is seen as military action. This metaphor, which could be cast
as sex is war, is used to talk about the physical side of sex, that is sexual
activity and the sexual organs; or, to put it in Carrolls (1994: 107) words,
it serves to emphasize the biological semantics of sex. This emphasis
does not imply that sexual intercourse in Shakespeare is considered to
be devoid of love and affection. Rather, it is precisely because love, lust or
sexual desire and sexual activity are often so inextricably intertwined that
there is so much overlap between the metaphorical expressions used to
talk about them
8
.
The source domain of war, on the other hand whose coherent organi-
zation structures part of the target domain of sex is understood here in its
broad sense of confict or fghting between two parties carried out by force
of arms, and includes not only military war, but also other related activities
like duels, fencing and jousting, as well as horsemanship, archery and hunt-
ing, on the borderline between the domains of war and those of games and
sports. Since Shakespeare often resorts to metaphors that draw on these
source domains (sex is hunting, sex is horse riding, etc.), some of these
will also be discussed as they frequently overlap to form stylistically rich
networks, generating a variety of ingenious puns.
The method for the identifcation and selection of the sexual metaphors
dealt with in this study follows the techniques employed in corpus linguistics
9

and corpus-based approaches to metaphor
10
, especially Vanparys (1995) or
Goatlys (1997) dictionary-based research
11
.
3. Conceptual metaphor theory
3.1 Some major claims
Before proceeding to the description and evaluation of the metaphors
under scrutiny, it seems that it is necessary to refer, at least briefy, to some of
the postulates of CMT that are of special relevance to the aesthetic dimen-
sion of the literary text
12
. The main one, and the cornerstone on which the
whole theory rests, is that metaphor, often considered a matter of language
alone, is also a matter of thought. This major tenet is stated at the begin-
ning of Lakoff and Johnsons (1980: 3) foundational text:
Metaphor is typically viewed as characteristic of language alone, a mat-
ter of words rather than thought and action . . . We have found, on the
contrary, that metaphor is pervasive in everyday life, not just in lan-
guage but in thought and action. Our ordinary conceptual system, in
terms of which we both think and act, is fundamentally metaphorical
in nature.
By locating metaphor at the level of thought, CMT departs drastically
from traditional theories of metaphor. This is refected in the very termi-
nology of the theory, for metaphor or conceptual metaphor is the
name given to the operation whereby we understand one domain of expe-
rience in terms of another, conventionally represented in the form a is b
(e.g. life is a journey), where a is the target and b the source domain,
roughly equivalent, respectively, to Richards (1936) vehicle and tenor.
In contrast, metaphorical expression or linguistic metaphor is the
label used to refer to the linguistic expressions that instantiate the concep-
tual metaphor
13
.
A frequently quoted example of a conceptual metaphor is argument
is war (Lakoff and Johnson 1980: 4ff), in which the concept argument
is metaphorically conceptualized as war. Thus, knowledge of the domain
of war, a concrete domain of experience with a well-known schematized
structure in terms of entities involved, goals, actions, etc., is used to struc-
ture the domain of argument. This mapping involves a set of ontological
(Lakoff 1993: 207) correspondences according to which entities in the tar-
get domain correspond systematically to entities in the source domain:
the argument-is-war mapping:
Source: war Target: argument
war contenders people arguing
war strategies strategies for arguing
to win/lose a battle to win/lose an argument
to stop fghting to stop arguing
This schematic set of correspondences allows us to reason about argu-
ments using our knowledge of war by drawing a series of inferences or
metaphorical entailments, which in turn yield a number of metaphorical
expressions at the linguistic level: for example, one may attack someones
arguments, defend ones point of view, use a particular strategy to defeat
ones opponent, etc. In fact, it is these metaphorical expressions that allow
CM theorists to reconstruct in a bottom-up manner the conceptual meta-
phors that underlie them
14
.
Even though conceptual metaphors are schematically represented as a is
b, mappings across domains are not complete but partial. This means that
only some aspects of the source domain are used in understanding the tar-
get domain. For instance, in the metaphor argument is war, the aspects
of the source that are mapped on the target are basically those that have to
do with control or dominance of an opponent. The mapped portion of the
target domain is said to be highlighted, as opposed to the unmapped por-
tion that remains hidden. This distinction is of relevance for the analysis
of style, since the preference for certain metaphors in terms of what they
highlight or hide, or the particular ways in which they are exploited become
available as elements with which to describe, assess and even differentiate
styles
15
. Needless to say, to establish which parts of the source are mapped
onto which parts of the target is no simple task
16
. For the purpose of this
analysis, it is probably enough to remember that the answer is related to
the fact that mappings are very often based on images of various kinds with
highly schematic structures, known as image-schemas, that impose certain
constraints on them.
3.2 CMT and poetic language
A corollary to the principle that our ordinary conceptual system is fun-
damentally metaphorical in nature is that poetic language, traditionally
regarded as a special kind of language, is not too different from every-
day language. This idea is elaborated on and expanded in Lakoff and
Turner (1989). The book shows that, as far as metaphor is concerned,
the language of literature is similar to ordinary language: both are built
on the same metaphors we live by. Lakoff and Turner argue that literary
authors appeal to us not because they create new metaphors but because
they use, modify or refresh existing ones. They achieve this by enhancing
conventional everyday language through basic devices or operations such
as extending, elaborating, questioning and combining (Lakoff and
Turner 1989: 6772, Kvecses 2002: 4357).
Paradoxical though it may sound, by equating literary and ordinary lan-
guage, Lakoff and Turner provide us with useful instruments for the analysis
of style. Their book illustrates their ideas with quite a few conceptual meta-
phors and numerous metaphorical expressions taken from various authors,
including Shakespeare. Yet the complex mechanisms of these operations
and their stylistic potential become clearer when illustrated with metaphors
that involve wordplay and punning, for which Shakespeare is an inexhaust-
ible source. Here are some illustrative examples.
In Hamlet (3.2.285), an embittered and angry Hamlet resolves that he
is going to be cruel to his mother, not by hurting her physically but psy-
chologically: I will speak daggers to her, but use none. Two scenes later,
Hamlet fulfls his resolution through a battery of verbal daggers which
makes his mother beg him to stop using the same expression:
queen O! speak to me no more;
These words like daggers enter in mine ears;
No more, sweet Hamlet!
(Hamlet 3.4.1079)
Much of the striking effect of Hamlets and the Queens words arises
from the novel way in which Shakespeare elaborates on the argument is
war metaphor. In everyday usage, both argument is war and the related
metaphor verbal aggression is physical aggression
17
map a limited
set of aspects of the source domain which, as was pointed out, gives rise
to a series of conventional metaphorical expressions such as to attack
someones arguments or to defeat ones opponent. What Shakespeare
does here is to exploit the potential of this metaphor by introducing a new
constituent in the source domain of war, namely dagger not normally
mapped so that a new connection is set up in the mapping: daggers
words
18
.
The words are weapons metaphor is again instantiated in Loves Labours
Lost, this time in a comic context. In the second scene of Act 5, Berowne
surrenders to Rosalines wit and verbal dexterity after having been knocked
down by her in one of their frequent wit combats:
berowne Here stand I, lady; dart thy skill at me;
Bruise me with scorn, confound me with a fout;
Thrust thy sharp wit quite through my ignorance;
Cut me to pieces with thy keen conceit;
(Loves Labours Lost 5.2.4258)
On this occasion, Shakespeare elaborates on the metaphor still fur-
ther by using some new elements of the source domain of physical
aggression, thus making the linguistic expressions more numerous and
varied. This metaphorical utilization of the source domain highlights
new aspects of the target domain
19
: speech is not just daggers, as in
Hamlet, but a dart that can be thrown at ones opponents, a solid object
with which to bruise them, a sharp dagger or a rapier to pierce them
and even a sharp knife to cut them to pieces. The very elaboration of the
metaphor and its hyperbolic nature are, indeed, what lends this speech
its comic tone.
A third instantiation of this metaphor is given a sexual twist in another wit
combat, this time between Beatrice and Benedick. The joke based on the
expression put down uttered twice by Don Pedro is on Benedick. The
former has just witnessed another round in the combat and congratulates
a victorious Beatrice who, in turn, quibbles on his words:
don pedro You have put him down, lady, you have put him down.
beatrice So I would not he should do me, my lord, lest I should prove
the mother of fools . . . .
This exchange shows one of the many metaphorical connections
between the domains of war and sex. Here Shakespeare manages to estab-
lish this link through punning, a phenomenon explained in cognitive lin-
guistics in terms of metaphorical scope (Barcelona 2001, Kvecses 2002:
154). The next section deals with this concept and those of metaphori-
cal highlighting and hiding illustrated with some more examples from
Shakespeare.
3.3 Metaphorical highlighting/hiding and metaphorical scope
Since mappings are, and can be, only partial (Kvecses 2002: 79), the
full characterization of a given target domain requires a number of source
domains. This is the case of emotion concepts such as anger, happiness or
love, understood via a wide range of source domains love is a journey,
but also a bond, heat, fire, etc. An exploration of these source domains to
see what each highlights or hides will surely give us interesting insights into
Shakespeares style. It will reveal, for example, that the choice of domain is
often constrained or motivated by a variety of factors: the presence of cer-
tain characters on stage, the theme of a poem or play, or the demands of a
particular dramatic situation so as to keep a balance between a given topic
and the metaphor deployed. In Cymbeline, for instance, the sex is music
metaphor is probably triggered by the entrance of the musicians
20
:
cloten I would this music would come. I am advised to give her
music o mornings; they say it will penetrate. Enter Musicians.
Come on; tune. If you can penetrate her with your fngering, so; well
try with tongue too. . .
(Cymbeline 2.3.810)
Just as some target domains need a wide range of source domains, some
source domains may also become available for characterizing a wide vari-
ety of target domains. When this happens, the source concepts are said to
have a wide metaphorical scope (Kvecses 2002: 108); and war is such a
concept, especially productive for structuring the domain of sex and very
frequently utilized
21
for characterizing many target domains like love,
argument, life, illness, etc. Again, there is a close connection between
the width of the scope of a given concept and the culture, society or even
discourse community in which it is used. The ubiquity of the war meta-
phor in Shakespeare should come as no surprise, considering his cultural
environment and the omnipresence of war throughout his plays (Edelman
2000: 2).
Like source domains, the linguistic expressions that realize them can also
have a wide scope, that is a number of distinct uses that refer to differ-
ent topics. An instance of a metaphorical expression with a wide scope in
English is hot. Since heat maps onto several target domains giving rise to
a series of metaphors (love is heat, lust is heat, anger is heat, etc.),
the metaphorical expression hot may be found instantiating some of these
metaphors simultaneously, for example he was getting hot (with anger,
lust, etc.). Shakespeares works abound in metaphorical expressions with
a wide scope, heat and hot fguring among the most frequent ones. Indeed,
these two words are used throughout the canon to refer to anger, lust, love
and, very frequently, illness or disease (especially venereal disease).
3.4 Image-schemas and Shakespeares sexual language
Image-schemas, defned before as highly schematic structures, are a spe-
cial type of higher-level metaphor which can also account for and facilitate
the analysis of the metaphorical patterning of Shakespeares plays. In these
metaphors, it is not conceptual elements of knowledge that get mapped,
but conceptual elements of image-schemas that derive from our interaction
with the world. For instance, the fundamental physical experience of mov-
ing along a path from one place to another is captured in the image-schema
of path; likewise, the experience of our own bodies, with an interior and an
exterior, gives rise to the schema of container, whose embodied ground-
ing is explained by Lakoff: We understand our own bodies as containers
perhaps the most basic things we do are ingest and excrete, take air into our
lungs and breathe it out (1987: 271; emphasis added).
The schemas of container and path have proved to be very productive
in Shakespeare. Freeman (2004) demonstrates that in Macbeth Shakespeare
draws on these two schemas intertwining them sometimes to create, for
instance, a four-dimensional image of Macbeths downfall in his famous
To-morrow, and to-morrow, and to-morrow speech (5.5.2433); their
interaction also explains elements of the plot, says Freeman (ibid. 107),
who concludes: We understand Macbeth its language, its characters, its set-
tings, its events, its plot in terms of these two central bodily based image-
schemata (ibid.).
Crane also emphasizes the centrality of the container schema in two
other plays. Referring to Measure for Measure, she states that the operative
spatial confguration in this play centres on a sense of the body as a con-
tainer that is variously impermeable or permeable to outside infuences
(2001: 162); and about the dramatic function of the schema in Hamlet, she
points out that,
The play uses imageryof breached fortifcation and bodily penetration,
of weeds and diseaseto suggest that Elsinore and Denmark are larger
projections of the psychic structures of old Hamlet, Hamlet, Claudius,
and the other characters. (ibid. 118; emphasis added)
One can easily extend Cranes imagery to include the sexual act. Thus,
bodily penetration, an activity which may well be seen in the light of
Lakoffs basic things we do would be viewed as one more sexual metaphor-
ical expression arising from the container schema. Indeed, this image-
schema underlies many of the sexual metaphors in everyday English
22
as
well as many of Shakespeares sex-war metaphors, as the next section
will show.
4. The sex is war conceptual metaphor in
Shakespeares plays and poems
4.1 Scenarios, mappings and image-schemas
It is obvious then that the basic schema of the war experience bears enough
structural resemblance to the experience of sex as to offer Shakespeare an
appropriate and highly productive source domain. To begin with, war and
sex are forms of human interaction normally involving two entities. Both
activities share a series of actions and events and, most importantly, much
of the two domains is governed by the container image-schema. These
coincidences are refected at the linguistic level in a large number of con-
ventional expressions that apply to the two domains in Elizabethan English
(enter, have, take, possess, penetrate, win, etc.).
The similarity between the conceptual domains of sex and war can be
best perceived by comparing their respective prototypical scenarios
23
. In
the sex scenario, the male participant almost invariably plays the agent
with woman as the affected. Similarly, participants in the war scenario
are the male agents of the attack and the people or places affected by
the attack. The traditional division between battle war and siege war
captures well the distinction between people and places as affected
participants:
War scenario Sex scenario
A wants to have/take B A desires to have sex with B
A approaches B A gets physically closer to B
A initiates the attack, besieging B A begins to woo or court B
B tries to resist As attack B tries to resist As wooing
B fghts back/or surrenders B rejects A or yields to wooing
A manages to enter and have/take B;
or As attack is repelled by B
A succeeds in having sex with B;
or A is rejected
These scenarios facilitate the mapping across the domains, according to
which entities in the war domain correspond to entities in the sex domain.
Therefore, a very basic mapping of the metaphor sex is war as used by
Shakespeare would contain the following basic correspondences:
Source: war Target: sex
A fght/battle A sexual encounter
Contenders Lovers
Approaching the objective Wooing, courting
Fighting/battling Love-making
Winning the battle (Man) Achieving sexual intercourse
Surrendering (Woman) Submitting sexually /
yielding to have sex.
It goes without saying that Shakespeare is not inventing this metaphor,
for the conceit of the militia amoris was quite frequently used in classical
and medieval literature
24
. But he defnitely exploits it in such a way as to
depart from its more conventional realizations. In this respect, I would
not hesitate to claim that Shakespeares style, especially in some of his plays
and poems, has much to do with the way he exploits the metaphorical
entailments of sex is war. This claim is illustrated in the following sections
through a selection of instantiations of sexual-war metaphors to be found
in Shakespeare. Clearly, not all of them have the same stylistic signifcance;
so the attention they are paid varies here.
4.2 Shakespeares deployment of the sex is war conceptual metaphor
4.2.1 Setting the scene: a fight is a sexual encounter
As is well known, in Shakespeares plays and poems, several nouns and verbs
are used metaphorically to refer to sexual activity. A combat
25
, for instance,
can denote not only a fght but also a sexual encounter, and so does con-
fict, as when Tamora suggests to her lover Aaron that they may possess
a golden slumber after confict, such as was supposd / The wandering
prince and Dido once enjoyd (Titus Andronicus 2.3.29/245). Likewise,
contend, a word uttered by Julia on reading Proteus letter, suggests sex-
ual engagement: Now kiss, embrace, contend, do what you will (The Two
Gentlemen of Verona 1.2.139).
Another common metaphorical expression for sexual combat is tilt, as is
the related term bout (a round in jousting), on which Shakespeare plays in
1Henry VI. In the second scene of Act 3, an angry Talbot challenges Joan to
a bout and the latter quibbles on the idea of sexual encounter (notice hot):
talbot Foul fend of France, and hag of all despite,
Encompassd with thy lustful paramours!
. . .
Damsel, Ill have a bout with you again,
Or else let Talbot perish with this shame.
joan la pucelle Are you so hot, sir? . . .
(1Henry VI 3.2.634/679)
Joans lewd remark reinforces Shakespeares portrayal of her as a less
saintly woman than tradition and the chronicles present her.
4.2.2 military attributes/actions are sexual attributes/actions
Typically Shakespearean soldiers, like Talbot, exhibit two distinguishing
characteristics of the good warrior: martial prowess and sexual virility.
Through the metaphor military attributes are sexual attributes,
some of the expressions that designate the qualities associated with the good
soldier also apply to his sexual capacities. Thus, Shakespeares soldiers have
courage and mettle, which involve both military valour and sexual desire.
Frequently, the nouns and verbs used in this context apply simultaneously
to actions performed by the man-soldier and the man-lover, always ready to
fulfl their duty, do their offce or to serve words that entail both martial
and sexual service. Lavatch, the Clown in Alls Well that Ends Well, plays on
the senses of service when explaining to Lafeu his double role as a fool
and a knave:
lafeu Whether dost thou profess thyself, a knave, or a fool?
clown A fool, sir, at a womans service, and a knave at a mans.
lafeu Your distinction?
clown I would cozen the man of his wife, and do his service.
lafeu So you were a knave at his service, indeed.
clown And I would give his wife my bauble, sir, to do her service.
lafeu I will subscribe for thee, thou art both knave and fool.
clown At your service.
(Alls Well that Ends Well 4.5.916)
Win is another wide-scope metaphorical expression spanning the domains
of war and sex. The verb frequently occurs collocating with woo, thus rein-
forcing the link across the domains of battling and courting. In Richard
III, for example, at the end of the scene in which Richard Gloucester woos
Lady Ann with the intention of marrying her, he soliloquizes:
Was ever woman in this humour wood?
Was ever woman in this humour won?
Ill have her; but I will not keep her long
26
.
(Richard III 1.2.2413)
The same paronomasia wood/won is used in 1Henry VI (Suffolk says
of Margaret: Shes beautiful and therefore to be wood, / She is a woman,
therefore to be won [5.3.834]); and in Titus Andronicus. In this tragedy,
Demetrius, talking about Lavinia whom he plans to rape, expresses the
same idea adding one more sentence to his reasoning: She is a woman,
therefore may be wood; / She is a woman, therefore may be won; / She is
Lavinia, therefore must be lovd (2.1.8991).
4.2.3 wooing is assailing
Wooing in Shakespeare is very often cast as the preparation of an assault,
as assailing or accosting. Thus, in Cymbeline (1.4.43), Iachimo is offered a
choice for wooing: What lady would you choose to assail?. In Twelfth Night
(1.3), Sir Andrews lack of familiarity with the warlike language of courting
produces a comic episode when Sir Toby Belch encourages him to accost
(i.e. woo) Maria. To undo the misunderstanding, Sir Toby provides a few
more expressions, also belonging to the domain of war:
sir toby belch Accost, Sir Andrew, accost.
sir andrew Whats that?
sir toby belch My nieces chambermaid.
sir andrew Good Mistress Accost, I desire better acquaintance.
maria My name is Mary, sir.
sir andrew Good Mistress Mary Accost,
sir toby belch You mistake, knight: accost is, front her, board
her, woo her, assail her.
sir andrew By my troth, I would not undertake her in this
company. Is that the meaning of accost?
(Twelfth Night 1.3.2532)
As the examples presented so far illustrate, in this metaphorical war men
almost invariably play the role of the attacker or invader, and they are equipped
accordingly. Women, on the other hand, normally appear as the defenders, so
their equipment differs substantially from mens. In Much Ado about Nothing,
in the course of another hot battle of wits, Margaret reminds Benedick of
the type of weaponry that characterizes the two sexes. At the end of their
dialogue, the former, considering himself beaten, yields to Margaret the pro-
verbial bucklers, a common expression used to admit defeat; but Margaret
turns it into a lewd joke (Bucklers = pudends; Partridge 1947: 196):
benedick . . . I give thee the bucklers.
margaret Give us the swords, we have bucklers of our own.
Margarets words emphasize womens defensive role in the sexual com-
bat, limited to trying to protect themselves from mans attack. This role is
expressed through a similar metaphor from sword fghting in Troilus and
Cressida, thus linking up sex and war, two major motives in the play. In the
second scene of Act 1, as Pandarus and Cressida watch the soldiers pass
over, the former complains about the latters insistence on quibbling on
his remarks on the marching warriors: You are such a woman! one knows
not at what ward you lie (1.2.152). Picking up Pandarus words, Cressida
continues with the fencing metaphor, as she replies:
Upon my back, to defend my belly . . . ; if I cannot ward what I would not
have hit, I can watch you for telling how I took the blow; unless it swell
past hiding, and then its past watching.
(Troilus and Cressida 1.2.153/55)
The sexual load of Cressidas reply has been repeatedly noted and glossed
by lexicographers (see, e.g. Partridge 1947, Rubinstein 1984 or Williams
1997); and the sex as combat metaphor in which her self-description is cast
has likewise been commented on for its dramatic signifcance (see espe-
cially Palfrey 2005: 2613).
4.2.4 a woman is a walled city; to have/take a woman
is to have/take a city
Since womans defence is based almost exclusively on the possibility of mak-
ing herself impenetrable, she is often metaphorized as a fortress, walled
and locked against mans attack. In order to gain access into their walled
bodies, men besiege them, as the female voice in A Lovers Complaint reminds
us: And long upon these terms I held my city, / Till thus he gan besiege
me . . . (1767). Although it is true that this metaphor is part of an old
poetic topos, Shakespeare very often manages to go beyond it by subverting
its structure and its conventional language. Such subversion is found in The
Rape of Lucrece, a story built on this megametaphor (Kvecses 2002: 51) to
have/take a woman is to have/take a city
27
. This metaphor structures
the major metaphors in the poem which in turn depend largely on and
reveal the perspective of the characters. Thus, for Tarquin, the crucial met-
aphor is a martial one, rape is a military action, instantiated in the poem
as [a]ffection is my captain, and he leadeth; (271), [m]y heart shall never
countermand mine eye: (276) or [u]nder that colour am I to come to
scale / Thy never-conquerd fort (4812). This metaphor licenses him to
go ahead with his mission using all types of tactics to take Lucreces never-
conquerd fort. Lucreces metaphors, on the other hand, express the way
she sees her own body (the body is a house, the body is a mansion, or
the body is a temple), emphasizing the fact that it is not only private but
also holy and sacred. Their instantiations in the poem are, among others,
robbed and ransacked by injurious theft; her house is sacked, her quiet
interrupted, / Her mansion battered by the enemy; / Her sacred temple
spotted, spoiled, corrupted, / Grossly engirt with daring infamy.
An instantiation of the metaphor a woman is a walled city particularly
signifcant for its dramatic import is found towards the end of Henry V (5.2).
King Henry discusses with the King of France the terms of the agreement to
marry the latters daughter, Katharine, and seal the peace, when Burgundy
teases the English king for being love-blind for her. Henry reminds him
that his blindness might not be, after all, such a bad thing for the French:
king henry v It is so: and you may, some of you, thank love for my
blindness, who cannot see many a fair French city for one fair French
maid that stands in my way.
(Henry V 5.2.157)
The French King corrects Henry to point out that in fact both things a
fair French city and a fair French maid could be seen as the same thing
if looked at perspectively: Yes, my lord, you see them perspectively, the
cities turned into a maid; for they are all girdled with maiden walls that
war hath never entered (158). He does so as a reminder to Henry that by
marrying Katharine, a maid, he will also have, as agreed, the cities that he
has not yet entered and which remain, like the princess, maidens. In his
reply, Henry elaborates on the metaphor: I am content; so the maiden cit-
ies you talk of may wait on her: so the maid that stood in the way for my wish
shall show me the way to my will (161).
The dramatic implications of the metaphor in this scene could not be
clearer. Furthermore, by likening a woman to an article in an agreement
The king hath granted every article: / His daughter frst, and then in
sequel all, (164) the woman is also metaphorized by Westmoreland as a
commodity that can be passed on from one male proprietor to another.
28

Incidentally, this scene has no counterpart in The Famous Victories or in the
chronicles; it is entirely a Shakespearean invention, through which Henrys
role as a soldier is emphasized.
4.2.5 trying to have a woman is besieging her
The concept of the siege as sexual assault appears frequently in
Shakespeares work. Although this metaphor may fnd its way into the text
in a number of linguistic instantiations of little or no dramatic import at
all see Sir Andrews words above or Falstaffs amiable siege on Mistress
Ford in The Merry Wives of Windsor (2.2.82) it may also acquire greater dra-
matic signifcance in some plays. This is what happens in Alls Well that Ends
Well in the frst conversation between Helena and Parolles, an extended
instantiation of the metaphor a woman is a castle. Helena, seeing her
beloved Bertram depart from her, meditates on virginity; at her demand,
Parolles, a descendant from the miles gloriosus of Plautine comedy, offers
some advice:
parolles Are you meditating on virginity?
helena Ay. You have some stain of soldier in you; let me ask you a
question. Man is enemy to virginity; how may we barricado it
against him?
parolles Keep him out.
helena But he assails; and our virginity, though valiant in the
defence, yet is weak. Unfold to us some war-like resistance.
parolles There is none: man, sitting down before you, will
undermine you and blow you up.
helena Bless our poor virginity from underminers and blowers up!
Is there no military policy, how virgins might blow up men?
parolles Virginity being blown down, man will quicklier be blown
up: marry in blowing him down again, with the breach
yourselves made, you lose your city. . . .
(Alls Well that Ends Well 1.1.739)
Interestingly enough, the exchange takes place at the very beginning of
the play and bears directly on one of its main plots: Helenas calculated loss
of virginity. She draws on the same metaphor later on (3.7.224) when she
warns the widow: The count woos your daughter, / Lays down his wanton
siege before her beauty, / Resolvd to carry her: . . .
29
On the other hand, as
far as Shakespeares innovative utilization of this conventional metaphor is
concerned, it may not be irrelevant to point out that Helenas barricado
is registered in the OED as the frst fgurative use of this word as a verb; and
that its frst fgurative use as a noun is also registered in another play by
Shakespeare (The Winters Tale 1.2.241).
4.2.6 Breaches, gates and metaphorical vaginas
One of the entailments of the metaphor a woman is a walled city is that
access can be gained through several passages, as we saw above in Alls Well
that Ends Well (with the breach yourselves made). In The Rape of Lucrece,
Lucreces resistance arouses in Tarquin more rage, and lesser pity, / To
make the breach and enter this sweet city (4689). Other forms of access
that are metaphorically connected with the vagina in Shakespeare are gate,
door, sluice or way. Even though they are not exclusive to the domain of war,
these expressions frequently trigger the military metaphor. For instance,
when in The Merry Wives of Windsor (2.2.62) Ford utters the well-known pro-
verb they say, if money go before, all ways do lie open, Falstaff brings in the
military element by replying: Money is a good soldier, sir, . . . (63).
An extended example of the metaphor in which the womans vagina is
represented as an entry into her body is found in The Winters Tale, Act 1,
scene 2. The passage in which it appears contains a refection upon adultery
by Leontes introduced by the famous pun on play. In it, the king employs
the verb in its different senses that is to amuse oneself, to copulate and to
play a part when he tells his son Mamillius: Go play, boy, play; thy mother
plays, and I / Play too, but so disgracd a part . . . (1.2.2245). Leontes is
convinced that even though he is playing the part of an honoured husband
in front of everybody elses eyes, he is in fact a cuckold. At least, he can take
comfort in the idea that he is not the only one:
leontes . . . There have been,
Or I am much deceivd, cuckolds ere now;
And many a man there is . . .
That little thinks she [his wife] has been sluicd ins absence,
And his pond fshd by his next neighbour, . . .
. . . nay, theres comfort int,
Whiles other men have gates, and those gates opend,
As mine, against their will . . . .
The Winters Tale (1.2.2279/2312/2335)
Leontes refection concludes with the barricado metaphor, reinforced by
bag and baggage, a military expression with strong sexual connotations:
No barricado for a belly: knowt;
It will let in and out the enemy
With bag and baggage . . . .
The Winters Tale (1.2.2413)
The whole speech illustrates the importance of wordplay in the process of
metaphorical transformation, and shows how the container schema allows
Shakespeare to progress smoothly from one metaphor to another. And it is
precisely this structural similarity between the domains of sex and war that
accounts for the wide variety of metaphors that exist for the female body
30
.
4.2.7 Some sexual weaponry
In the examples above, the war of the sexes takes the form of a siege war, with
a focus on the defence of the woman, whereas when it comes to the battle-
war, contenders fght hand-to-hand combats, so the elements of the source
domain that are used are those that have to do with mans light weaponry. The
repertoire of weapons metaphorically used for the male copulatory organ is
extensive, ranging from projectile weapons, like pistols and arrows, to those
employed by the soldier in hand-to-hand combat, such as swords, pikes, etc.
All of them share their piercing potential as their common denominator.
4.2.7.1 a penis is a pistol; to emit semen is to discharge/
fire a pistol
The pistol is not only one of Shakespeares favourite fre weapons but also
the name of one of his funniest characters, a calculated coincidence that
allows Shakespeare to introduce some of the funniest and coarsest word-
play in 2Henry IV. In the fourth scene of Act 2, Pistol is welcomed by Falstaff
who offers him a toast and invites him to drink to the health of the hostess.
Falstaffs deliberate choice of verbs charge/discharge in accordance with
the name of his associate, triggers a string of puns and sexual metaphors
also built on the container schema:
falstaff Welcome, Ancient Pistol. Here, Pistol, I charge you
with a cup of sack: do you discharge upon mine hostess.
pistol I will discharge upon her, Sir John, with two bullets.
falstaff She is pistol-proof, sir; . . .
pistol Then to you, Mistress Dorothy; I will charge you.
falstaff No more, Pistol: I would not have you go off here. . . .
(2Henry IV 2.4.402/44/49)
Pistols name is again the object of bawdy punning in Henry V (2.1.), in
another tavern scene full of sexual double entendre structured on the same
metaphor. Newly married to Mrs Quickly, the Hostess, Pistol shows his
hostility towards Nym, former suitor of Mrs Quickly, and threatens to fre
his weapon with its (his) cock now triggered and ready to shoot: O viper
vile . . . Pistols cock is up, and fashing fre will follow (22/289).
4.2.7.2 sex is archery
As was pointed out before, very often the sex is war metaphor partially
overlaps with another conceptual metaphor in which sexual activity is cast
in terms of archery, a domain on the border between war, sports and games.
The pattern is similar to that of the pistol metaphor, but whereas in the
case of frearms it is mainly the male equipment that gets highlighted, in
the arrow metaphor the highlighting is equally distributed between the two
sexes. Thus, several verbs are used metaphorically to refer to the act of
copulation, such as shoot, hit, prick or aim at (copulating is hitting the
target); and the womans pudendum is metaphorized variously in archery
terms such as mark, hit, clout or prick (the vulva is a target).
Loves Labours Lost (4.1) contains an illustrative passage in which the met-
aphor is fully exploited. Once again, the context is that of a wit combat, this
time between Rosaline and Boyet. The exchange begins with some more
or less innocent puns that draw on the domains of archery and hunting
(notice the homophonic pun suitor/shooter):
boyet Who is the suitor? who is the suitor?
rosaline Shall I teach you to know?
boyet Ay, my continent of beauty.
rosaline Why, she that bears the bow.
Finely put off!
boyet My lady goes to kill horns; but, if thou marry,
Hang me by the neck if horns that year miscarry.
Finely put on!
rosaline Well then, I am the shooter.
(8997)
Maria, who witnesses the match, points out the dexterity of the contend-
ers: A mark marvelous well shot, for they both did hit it. But Boyet pushes
the metaphor still further to increase the temperature of the exchange:
boyet A mark! O! mark but that mark; a mark, says my lady!
Let the mark have a prick int, to mete at, if it may be.
maria Wide o the bow hand! i faith your hand is out.
costard Indeed a must shoot nearer, or hell neer hit the clout.
boyet An, if my hand be out, then belike your hand is in.
costard Then will she get the upshoot by cleaving the pin.
(1116)
The archery metaphor is also variously instantiated in Romeo and Juliet,
unsurprisingly, given the repeated references to Cupid in the play. In the
frst scene of Act 1, for example, Romeo confesses to his friend Benvolio
that he loves Juliet, something that Benvolio had already guessed: I aimed
so near, he says, prompting Romeos wordplay on archery metaphors that
Benvolio picks up bawdily:
romeo In sadness, cousin, I do love a woman.
benvolio I aimd so near when I supposd you lovd.
romeo A right good mark-man! And shes fair I love.
benvolio A right fair mark, fair coz, is soonest hit.
romeo Well, in that hit you miss: shell not be hit
With Cupids arrow; she hath Dians wit;
(196201)
Romeo moves on from archery to battling to convince his friend that Juliet
is a maid:
And, in strong proof of chastity well armd,
From loves weak childish bow she lives unharmd.
She will not stay the siege of loving terms,
Nor bide the encounter of assailing eyes,
(2025)
In the next act (2.1), the metaphor appears again in the course of a conver-
sation full of sexual references involving Benvolio and Mercutio. They discuss
Romeos infatuated state, and Mercutio cynically and realistically dismantles
the convention that relates love and blindness expressed in Benvolios words:
benvolio Blind is his love and best befts the dark.
mercutio If love be blind, love cannot hit the mark.
(378)
4.2.7.3 a vagina is a case
Several entailments can be derived from the archery metaphor, all of them in
accordance with the container image schema. For instance, if shooting an
arrow is to emit semen, the place in which arrows are kept the case is the
vagina. In Shakespeare there are plenty of examples of this metaphor, espe-
cially in contexts which lend themselves to homonymic punning on other
senses of case
31
. In Romeo and Juliet (3.3), for example, the Nurse declares
equivocally that Romeo, sad and devastated, is in the same situation or case
as her mistress, and she goes on with a chain of sexual innuendoes:
nurse O! he is even in my mistress case,
Just in her case!
friar laurence O woeful sympathy!
Piteous predicament! Even so lies she,
Blubbering and weeping, weeping and blubbering.
Stand up, stand up; stand, an you be a man:
For Juliets sake, for her sake, rise and stand;
Why should you fall into so deep an O?
The Nurse is not the only one punning on case. In 2Henry IV (2.1),
Mistress Quickly asks the Sheriffs offcers to arrest Falstaff for failing to
settle up with her. Her verbal blunders and her insistence on some sexually
loaded words add to the humour of the passage:
mistress quickly . . . I pray ye, since my exion is entered, and my case so
openly known to the world, let him be brought in to his answer. A hun-
dred mark is a long one for a poor lone woman to bear; and I have borne,
and borne, and borne; and have been fubbed off, and fubbed off, and
fubbed off, . . . Do your offces, do your offces, Master Fang and Master
Snare; do me, do me, do me your offces. (2Henry IV 2.1.16)
A few lines earlier the Hostess had already exposed herself to ridicule for
employing similar sexually-loaded words. Warning the Sheriffs offcers of
the dangers of Falstaffs propensity to stabbing, she says:
Alas the day! take heed of him: he stabbed me in mine own house, and
that most beastly. In good faith, he cares not what mischief he doth if
his weapon be out: he will foin like any devil; he will spare neither man,
woman, nor child. (2Henry IV 2.1.12)
4.2.7.4 a penis is a pointed piercing weapon
Short and long piercing weapons for hand-to-hand combat often represent
the penis metaphorically, as the last passage illustrates. Among them, the
sword is probably the most frequently used. The penis is a sword meta-
phor is often employed in comic passages, as in Loves Labours Lost (5.2),
Much Ado about Nothing (5.2) or Twelfth Night (3.4), but it is also used in
serious contexts, and sometimes contributes to the construction and devel-
opment of the plays plot. Such is the case of Antony and Cleopatra, a tragedy
in which the sword becomes the symbol of Antonys power and masculinity.
An instantiation of the metaphor is found in the well-known lines where
Agrippa tells Enobarbus how the queen of Egypt [m]ade great Caesar lay
his sword to bed; / He ploughd her, and she croppd (2.2.2623), thus
combining the sex is war metaphor with another productive metaphor:
sex is agriculture (see Oncins-Martnez 2006).
Once the metaphorical connection between sword and penis is established,
a series of entailments allows us to understand the sexual sense of a large num-
ber of swordplay and fencing terms throughout Shakespeares works, such as
thrust, stab, stick, draw or foin. In 2Henry IV, Doll, for instance, provides a
humorous comment on Falstaffs stabbing habit mentioned before by the
Hostess: Thou whoreson little tidy Bartholomew boar-pig, when wilt thou
leave fghting odays, and foining onights, . . . (2.4.102). Likewise, the verb
draw appears metaphorically related to the penis in similar sexual contexts. In
Romeo and Juliet, Peter reassures the Nurse: I saw no man use you at his plea-
sure; if I had, my weapon should quickly have been out, I warrant you. I dare
draw as soon as another man, . . . (2.4.79). Since drawing implies taking the
sword out of its sheath, swords and rapiers are very often said to be naked, thus
reinforcing their sexual connotations in certain contexts, for example, Twelfth
Night (3.4.130), 2Henry IV (2.4.89) or Romeo and Juliet (1.1.21).
As for the metaphorical use of long piercing weapons, such as pike, stake
or lance, it is worth paying attention to Falstaffs reference to the health
consequences of whoring; for in his observation, neatly articulated through
sex-war metaphorical expressions, the phallus-pike metaphor stands out:
falstaff . . . to serve bravely is to come halting off you know: to come off the
breach with his pike bent bravely, and to surgery bravely; to ven-
ture upon the charged chambers bravely, 2Henry IV 2.4.22
The union of Mars and Venus begets a group of fnely-wrought
Shakespearean metaphors: his martial and sexual (even venereal) language
feeds on them.
5. Final remarks
As has been demonstrated, the instruments provided by CMT allow us to
describe and evaluate the vast number of metaphorical expressions that run
through Shakespeares works and the relationship among them. The fact
that all the metaphorical language that pervades his texts can be grouped
into a limited number of conceptual metaphors facilitates their study and
therefore their literary analysis. Indeed, this is what CMT offers literary crit-
ics, especially when their feld of study is as fertile as the plays and poems writ-
ten by Shakespeare. The large amount of instantiations of his catalogue of
linguistic metaphors originates in a number of conceptual metaphors which
is necessarily much smaller sex is war, for example, but also friendship
is a bond, time is a person, love is food, etc., among others.
Since the language of sex pervades practically all of his dramatic and poetic
texts, perhaps it is not going too far to state that sex metaphors are among the
most frequently used by this writer; the amount of lexicographic work dedi-
cated specifcally to his sexual vocabulary would corroborate this. Moreover,
Williams (1997) has identifed approximately 150 words and expressions
related to sexual activity and the sexual organs whose earliest recorded use
seems to be Shakespeares. If we look at them carefully, we observe that
about a quarter of the total amount corresponds to metaphorical expressions
deriving from the conceptual metaphor sex is war. From a percentage of
this calibre it can be inferred that the numerous instantiations of this meta-
phor play an important stylistic role in the dramatic and poetic works of this
author. Through them, Shakespeare not only highlights elements of the tar-
get domain that are normally hidden, but also exploits and extends at least
partially the conventional use of this metaphor. In doing so, he enriches
the metaphors by setting up novel connections across domains, often drawing
on new constituents from the source domain to highlight new areas of the
domain of sex this happens, for instance, when he uses bullet or mark as
metaphorical expressions for the male and female sexual organs respectively.
These stylistic effects and those referred to throughout this essay confrm the
potential of the analysis of the still largely unexplored instantiations of the sex
is war metaphor in his texts; and the usefulness of CMT for their description
and aesthetic assessment.
Notes
1
I am very grateful to the anonymous reviewers whose valuable comments and sug-
gestions helped me enrich an earlier version of this essay.
2
See, for example, Green (1974), Charney (2000), Alexander and Wells (2001) or
Wells (2004).
3
Standard works in the feld of imagery studies are Spurgeon (1935), Armstrong
(1946) and Clemen (1951).
4
Early in the introduction, the authors express their regret that in spite of the
explosion of interest in metaphor in the late seventies, this has made little
impact on English literary studies where, they say, a confused and impressionis-
tic notion of imagery still reigns, despite frequent expressions of dissatisfaction
with its lack of methodology and its inappropriateness to some sorts of texts
(Thompson and Thompson 1987: 1). In her Shakespeares Brain probably the
most comprehensive cognitive account of Shakespeares language to date
Crane (2001: 183) expresses the same dissatisfaction with the lack of rigour
of Spurgeons and Armstrongs too subjective analyses of Shakespeares image
clusters.
5
The foundational works of the theory are Lakoff and Johnson (1980), Lakoff
(1987); Turner (1987) and Lakoff and Turner (1989); but see also Lakoff
(1993) for an update of the theory, and Kvecses (2002) for a comprehensive
account.
6
See, for example Semino and Culpeper (2002), Stockwell (2002) and Steen and
Gavins (2003).
7
In spite of all its sound theoretical foundation and applicability, CMT is, of course,
not without weaknesses and detractors. An aspect of the theory for which its found-
ers have been frequently criticized is their failure to acknowledge their intellectual
debt (see Jkel 1999). For more recent criticism of a different kind see the
references in note 14.
8
On this overlapping see Barcelonas (1995) reading of love metaphors in Romeo
and Juliet. Also, for a thorough account of the complex concept of love in Shake-
speare see Charney (2000). In this book written as an attempt to explore the
inadequacies of more traditional and old-fashioned accounts of love in Shake-
speare (Charney 2000: 2) the emphasis is put on the physical manifestations of
love: Love in Shakespeare expresses itself in physical desire, and even at its most
rapturous (as in Romeo and Juliet), never loses its sexual underpinnings (ibid.). In
the fnal chapter of the book, where the author insists on this idea, he establishes
an important distinction between love and lust which he relates to some specifc

images: All love in Shakespeare expresses itself sexually, but there is an impor-
tant distinction between love and lust. Lust is a product of desire and appetite
and is often associated with imagery of food, eating and animals. The fulflment
of lust may be connected with violence or rape, as in The Rape of Lucrece, in which
the sexual assault is an expression of power and control over an innocent, unpro-
tected, and defenceless woman (ibid.).
9
The advantages of using Corpus Linguistics techniques for building up new dic-
tionaries for Shakespeares language are convincingly argued for in chapter 3 by
Jonathan Culpeper. The present essay draws on some of the ideas put forward by
this author, in the conviction that such an approach to Shakespeares metaphors
can also give us valuable insight into his style.
10
See, for example Charteris-Black (2004), Deignan (2005), Stefanowitsch (2006)
or Semino (2008). For metaphor identifcation methods see the special issue of
Language and Literature on the topic (2002: 11(1)), and Pragglejaz Group (2007).
11
The starting point of the research was the compilation of a list of all the entries
contained in the dictionaries mentioned above. Then, the entries were grouped
in semantic domains. After this, a systematic search was made for the words and
expressions in the war domain in the electronic text of Shakespeares plays and
poems. A concordance was generated for every item in the list. Finally, a manual
analysis was conducted in order to discriminate metaphorical usages and see how
these worked in the discourse of the play or poem. The search was done by means
of the software program WordSmith Tools (Scott 2004), and the text used is W.
J. Craigs (1914) electronic edition of Shakespeares works, available at: www.bar-
tleby.com/70/. The reason for choosing this standard edition of Shakespeares
works is that it has been partially tagged and made freely available by Mike Scott
(www.lexically.net/) whom I want to thank again for his generous help and con-
stant support.
12
In the brief account of the theory that follows, I have made extensive use of end-
notes to explain some aspects of the theory that, if given in the main text, might
have encumbered its reading.
13
Following the convention in CMT, small capitals are used to refer to the mappings
and to the conceptual metaphor: life is a journey. Metaphorical expressions
appear in inverted commas.
14
This bottom-up method is mentioned in the very frst chapter of Lakoff and
Johnson (1980): Primarily on the basis of linguistic evidence, we have found
that most of our ordinary conceptual system is metaphorical in nature (1980:
4). However, the theory has received some criticism for its lack of empirical
basis (see, e.g. Geeraerts 2006: 464, Semino 2008: 199ff, Stefanowitsch 2006:
910). The corpus-based approaches to metaphor recently adopted have come
to redress this defcit (see note 9 above).
15
Kolodny (quoted in Kvecses 2005: 90) shows differences between American men
and women in their conceptualization of the frontier in the period 16301984;
while for men the frontier is a virgin land to be taken, for women it is a garden to
be cultivated.
16
On these issues see Kvecses (2002), especially chapters 7 and 8.
17
On this metaphor see Vanparys (1995) and for a similar formulation (antago-
nistic communication is physical aggression) see Semino (2005, 2008).
18
The words are weapons metaphor, a frequent one in Shakespeares repertoire,
is especially relevant to Hamlet where the ear becomes a key element as the site
for verbal and physical aggression, pain and eventually death.
19
For a frst account of the phenomena of highlighting and hiding see Lakoff
and Johnson (1980: 103); for these two and the notion of metaphorical utiliza-
tion see Kvecses (2002: 7983).
20
For a context-induced account of conceptual metaphor see Kvecses
(forthcoming).
21
Whereas in discussing the notions of metaphorical highlighting and hiding
Lakoff and Johnson (1980) use the terms used and unused to refer to the
part of the source domain that gets mapped or remains unmapped, respectively,
Kvecses (2002: 813) proposes metaphorical utilization. In this essay I use
them interchangeably.
22
The metaphorical connection between an act of entering and sexual penetration
is well entrenched in English, as proved by the presence of various polysemous
terms with a conventional sexual sense, for example enter, penetrate or penetration
(cf. OED).
23
A scenario is described as an organized package of information or knowledge
about a particular type of situation, including settings, entities, participants, goals
and actions (Semino 2008: 229).
24
On military-sexual metaphors in classical literature, see, for example, Adams
(1982) or Atkins (1978: ch. 6 The metaphors of sex); for the siege metaphor
and the representation of the womans body as a castle in medieval literature see
Corfs and Wolfe (1995), especially the section Siege as metaphor and literary
event.
25
Words in bold indicate that the forms have been identifed to have a sexual sense
in any of the dictionaries consulted or in the OED.
26
OED gives this very quotation to illustrate the frst occurrence of the sexual sense
of have: 14.e. To have sexual intercourse with, to possess sexually.
27
This subversion has been pointed out by some critics. Maus (1986), for instance,
argues that much of the originality of the poem lies in the way Shakespeare actu-
alizes the conventional language of Elizabethan love poetry (Maus 1986: 77). The
same idea is developed in Vickers (1985), another reading of the poem in terms
of military metaphors: The poet transforms the repetition of convention into its
subversion; he simultaneously masters and undermines the descriptive mode he
employs (Vickers 1985: 172).
28
This metaphor is inextricably bound to the cultural context of many of Shake-
speares history plays, in which the fgurative verges on the literal for, as is well
known, the bestowal of a city as part of a dowry was common practice, as the
Henry V text shows.
29
Bertrands siege ends up in a most brutal sexual attack, as we learn from the
Second Lord later on in the play: He hath perverted a young gentle-woman
here in Florence, of a most chaste renown; and this night he feshes his will in
the spoil of her honour: . . . (4.3.89). The metaphor of pillage in which the
incident is cast is part of the chivalric rhetoric of a patriarchal society, whose
devastating effects for men and, especially, for women are discussed by Hatt-
away (1994) in the following terms: We can show, by a reading of Shakespeare,
that chivalry, residual chivalry, affected male sexuality disastrously. Chivalry
purports to relate to ethics but in reality it relates to politics. Warlike aims are
disguised as service to the female . . . and turn the consorting of men and women
to a hunt for prizes and maidenheads. Chivalry constructs women as passive, as
ornaments, and imposes chastity (something that can be owned or taken by a
man) as a means of legitimating male power. The condition of woman becomes
a question of value, women become thereby tokens of exchange, and value is
focused on their sexuality, on honesty rather than honour . . . If the metaphors
are not courtly they are martial: power is generated not through courtly nego-
tiation but through the vocabulary of gentlemanly combat (Hattaway 1994:
133).
30
The body of a woman and her pudendum are metaphorized as different types of
containers, e.g. a pond, a castle, a vessel, a garden, or a box (Carroll 1994: 110),
each one highlighting and hiding particular characteristics of women.
31
See, for example Alls Well that Ends Well (1.3), 1Henry VI (5.5) or The Merry Wives
of Windsor (4.1).
References
References to Shakespeares works are from the electronic edition of
W. J. Craig, The Oxford Shakespeare (1914), available at: www.bartleby.
com/70/.
Adams, James N. 1982. The Latin Sexual Vocabulary. London: Duckworth.
Alexander, Catherine M. S. and Stanley Wells (eds) 2001. Shakespeare and Sexuality.
Armstrong, Edward A. 1946. Shakespeares Imagination A Study of the Psychology of
Association and Inspiration. London: Lindsay Drummond.
Atkins, John 1978. Sex in Literature. Vol. 3. The Medieval Experience. London: John
Calder.
Barcelona, Antonio 1995. Metaphorical models of romantic love in Romeo and
Juliet, Journal of Pragmatics 24: 66788.
Barcelona, Antonio 2001. On the systematic contrastive analysis of conceptual meta-
phors: case studies and proposed methodology, in: M. Ptz (ed.), Applied Cognitive
Linguistics. Language Pedagogy. Vol. II. Berlin: Mouton de Gruyter, 11746.
Bradshaw, Graham 2004. Preface, in: Graham Bradshaw, Tom Bishop and Mark
Turner (eds), The Shakespearean International Yearbook, 4: Shakespeare studies today.
Aldershot: Ashgate, ixxvii.
Carroll, William C. 1994. The virgin knot: language and sexuality in Shakespeare,
Shakespeare Survey 16: 10719.
Chamizo-Domnguez, Pedro and Francisco Snchez-Benedito 2000. Lo que nunca
se aprendi en clase. Eufemismos y disfemismos en el lenguaje ertico ingls. Granada:
Editorial Comares.
Charney, Maurice 2000. Shakespeare on Love & Lust. New York: Columbia University
Press.
Charteris-Black, John 2004. Corpus Approaches to Critical Metaphor Analysis.
Basingstoke: Palgrave Macmillan.
Clemen, Wolfgang 1951. The Development of Shakespeares Imagery. London:
Methuen.
Colman, Ernest Adrian M. 1974. The Dramatic Use of Bawdy in Shakespeare. London:
Longman.
Corfs, Ivy A. and Michael Wolfe, M. 1995. The Medieval City Under Siege. New York:
Boydel and Brewer Ltd.
Princeton and Oxford: Princeton University Press.
Culpeper, Jonathan 2011. A New Kind of Dictionary for Shakespeares Plays: an
Immodest Proposal (Chapter 3 in this volume).
Deignan, Alice 1997. Metaphors of desire, in: Keith Harvey and Celia Shalom (eds),
Language and Desire. Encoding Sex, Romance and Intimacy. London: Routledge,
2142.
Deignan, Alice 2005. Metaphor and Corpus Linguistics. Amsterdam and Philadelphia:
John Benjamins.
Edelman, Charles 2000. Shakespeares Military Language : a Dictionary. London: The
Athlone Press.
Emanatian, Michelle 1995. Metaphor and the expression of emotion: the value of
cross-cultural perspectives, Metaphor and Symbolic Activity 10.3: 16382.
Emanatian, Michelle 1996. Everyday metaphors of lust and sex in Chaga, Ethos
24.2: 195236.
Freeman, Donald C. 2002. Afterword, in: Elena Semino and Jonathan Culpeper
(eds), Cognitive Stylistics: Language and Cognition in Text Analysis. Amsterdam:
John Benjamins, 31924.
Freeman, Donald C. 2004 [1995]. Catch[ing] the nearest way: Macbeth and cog-
nitive metaphor, in: Jonathan Culpeper, Mick Short and Peter Verdonk (eds),
Exploring the Language of Drama: From Text to Context. London and New York:
Routledge, The Interface Series, 96111.
Geeraerts, Dirk 2006. Cognitive Linguistics: Basic Readings. Berlin and New York:
Mouton de Gruyter.
Goatly, Andrew 1997. The Language of Metaphors. London: Routledge.
Green, Martin 1974. The Labyrinth of Shakespeares Sonnets. An Examination of Sexual
Elements in Shakespeares Language. London: Charles Skilton Ltd.
Hattaway, Michael 1994. Fleshing his will in the spoil of her honour: desire, misog-
yny, and the perils of chivalry, Shakespeare Survey 16: 12135.
Jkel, Olaf 1999. Kant, Blumenberg, Weinrich: some forgotten contributions to the
cognitive theory of metaphor, in: Raymond Gibbs and Gerard J. Steen (eds),
Metaphor in Cognitive Linguistics. Amsterdam: John Benjamins, 927.
Kvecses, Zoltan 2002. Metaphor: A Practical Introduction. Oxford: Oxford University
Press.
Kvecses, Zoltan 2005. Metaphor in Culture. Cambridge: Cambridge University
Press.
Kvecses, Zoltan (forthcoming) Context-induced variation in metaphor, in: Fiona
MacArthur, Jos L. Oncins-Martnez, Manuel Snchez-Garca and Ana Mara
Piquer-Priz (eds), Metaphor in Cross-Cultural Communication. Amsterdam: John
Benjamins.
Kyratzis, Sakis 2007. The semantics of desire: exploring desire, love and sexual-
ity through metaphor, in: Sakis Kyratzis and Helen Sauntson (eds), Language,
Sexualities and Desire. London: Palgrave, 96117.
Lakoff, George 1987. Women, Fire and Dangerous Things: What Categories Reveal about
the Mind. Chicago: The Chicago University Press.
Lakoff, George 1993. The contemporary theory of metaphor, in: Andrew Ortony
(ed.), Metaphor and Thought (2nd ed.). Cambridge: Cambridge University Press,
20251.
Lakoff, George and Mark Johnson 1980. Metaphors We Live By. Chicago: The Chicago
University Press.
Lakoff, George and Mark Turner 1989. More Than Cool Reason: A Field Guide to Poetic
Metaphor. Chicago: The Chicago University Press.
Maus, Katharine Eisaman 1986. Taking tropes seriously: language and violence in
Shakespeares Rape of Lucrece, Shakespeare Quarterly 37.1: 6682.
Oncins-Martnez, Jos Luis 2006. Notes on the metaphorical basis of sexual lan-
guage in Early Modern English, in: Juan Gabriel Vzquez-Gonzlez, Montserrat
Martnez-Vzquez and Pilar Ron-Vaz (eds), The Historical Linguistics-Cognitive
Linguistics Interface. Huelva: Universidad de Huelva, 20524.
Palfrey, Simon 2005. Doing Shakespeare. London: The Arden Shakespeare. Thomson
Learning.
Partridge, Eric 2001 [1947]. Shakespeares Bawdy. London: Routledge.
Pfaff, Kerry L., Raymond Gibbs Jr. and Michael D. Johnson 1997. Metaphor in
using and understanding euphemism and dysphemism, Applied Psycholinguistics
18: 5983.
Pragglejaz Group 2007. MIP: A method for identifying metaphorically used words
in discourse, Metaphor and Symbol 22.1: 139.
Richards, Ivor A. 1936. The Philosophy of Rhetoric. Oxford: Oxford University
Press.
Rubinstein, Frankie 1984. A Dictionary of Shakespeares Sexual Puns and Their
Signifcance. London: Macmillan.
Snchez-Garca, Manuel 2003. Amor y metfora conceptual: aproximacin a los
sonetos 153 y 154 de Shakespeare desde la lingstica cognitiva, Annual Review
of Cognitive Linguistics 1: 16178.
Scott, Mike 2004. WordSmith Tools (Version 4). Oxford: Oxford University Press.
Semino, Elena 2005 The metaphorical construction of complex domains: the case
of speech activity in English, Metaphor and Symbol 20: 3570.
Semino, Elena 2008. Metaphor in Discourse. Cambridge: Cambridge University
Press.
Semino, Elena and Jonathan Culpeper (eds) 2002. Cognitive Stylistics: Language and
Cognition in Text Analysis. Amsterdam: John Benjamins.
Spurgeon, Caroline 1935. Shakespeares Imagery and what it tells us. Cambridge:
Cambridge University Press.
Steen, Gerard and Joanna Gavins (eds) 2003. Cognitive Poetics in Practice. London
and New York: Routledge.
Stefanowitsch, Anatol and Stefan Thomas Gries 2006. Corpus-Based Approaches to
Metaphor and Metonymy. Berlin: Mouton de Gruyter.
Stockwell, Peter 2002. Introduction to Cognitive Poetics. London: Routledge.
Thompson, Ann and John O. Thompson 1987. Shakespeare: Meaning and Metaphor.
Harvester, Brighton: Harvester Press.
Turner, Mark 1987. Death Is the Mother of Beauty: Mind, Metaphor, Criticism. Chicago:
The University of Chicago Press.
Vanparys, Johan 1995. A survey of metalinguistic metaphors, in: Louis Goossens,
Paul Pauwels, Brygida Rudzka-Ostyn, Anne-Marie Simon-Vandenbergen and
Johan Vanparys (eds), By Word of Mouth: Metaphor, Metonymy and Linguistic Action
in a Cognitive Perspective. Amsterdam: John Benjamins, 134.
Vickers, Nancy J. 1985. This heraldry in Lucreces face, Poetics Today, 6, 12:
17184.
Wells, Stanley 2001 [1947]. Foreword, in: Eric Partridge, Shakespeares Bawdy.
London: Routledge.
Wells, Stanley 2004. Looking for Sex in Shakespeare. Cambridge: Cambridge University
Press.
Williams, Gordon 1997. A Glossary of Shakespeares Sexual Language. London: The
Athlone Press.
Chapter 11
Cognitive Interplay: How Blending
Theory and Cognitive Science
Reread Shakespeare
1
Amy Cook
1. Introduction
Theatre works on the body and mind of the spectator, changing minds and
touching bodies at the deepest level. As a theatre scholar and practitioner, I
am driven to understand the nature of that work: how is it that an embod-
ied story told onstage has the power to move an audience? At a time when
most people are more accustomed to the dramaturgical and rhetorical
simplicity of a television comedy, how do audiences unpack the language
and storytelling in a play like Richard III ? Theatre and performance studies
have productively turned to theories of anthropology, psychology, linguis-
tics and others to seek answers to the questions that drive the feld. In a
move likely to provide the kind of jolt to the feld that Goffman or Freud
once did, scholars are now seeking and fnding answers in the cognitive
sciences.
2
As David Saltz has helpfully pointed out, it is important that an integra-
tion of cognitive science into theatre and performance studies should
not simply use research from the sciences to validate our feelings,
instincts or theories.
3
Just as an essay on Richard III or the use of pros-
thetics in the performance of Richard III is valuable insofar as it produces
answers and questions for the laboratory of the rehearsal room, scientifc
research should provide new ways of questioning assumptions within our
home discipline and illuminate new readings of text and performance.
Interdisciplinary work requires that scholars be bilingual, it does not
require them to be converts. Cognitive linguistics links language, cogni-
tion and the body in ways that impact practical and theoretical issues in
performance and is therefore a good starting point for an interdisciplin-
ary investigation.
Cognitive Interplay 247
To understand how we understand is to know how to grow or shift our
understanding. The research within cognitive linguistics on metaphor and
blending theory provides ways of unpacking meaning and connecting it to
other images and ideas evoked throughout the play. Understanding lan-
guage this way allows a dramaturgic analysis of a play to focus on spaces
primed but not necessarily overt. Cognitive linguistics reinvigorates textual
analysis but, perhaps even more important for the long-term strength of
this growing feld, provides a link between speaking and thinking, words
and neurons. I begin with an introduction to the linguistic theory that has
proved most useful in my textual analysis and move to the neuroscience
implicated by the performances of that text.
2. Blending interplay
Language works on the body/mind of the listener, and so a method of pro-
cessing this language seems imperative to theatre scholars. It is not diffcult
to understand what now is the winter of our discontent . . . means; what is
challenging is to understand how it means that. Conceptual Metaphor Theory
(CMT) and Conceptual Blending Theory (CBT) both counter traditional
assumptions of an inherited grammar structure that parses sentences such as
the cat is on the mat based on memorized defnitions and rules of word place-
ment. Cognitive linguistics now generally agrees that language and thinking
are creative and embodied and use metaphors, models and blends. Certain
thoughts are contained and defned by the metaphor we use to talk about
them.
4
For example, a metaphor like time is money
5
(I dont have the time
to spend) will systematically lead to entailment metaphors (time is a valu-
able commodity: I cant afford another day of this nonsense) and our rela-
tionship to time becomes defned by this coherent system of thinking of time.
This is how, in our society, time can be spent or wasted, and time is seen as
something one has for one activity but not another. Metaphors illuminate some
elements of the abstract concept and hide others, because a metaphor will only
map some information from the source domain (money) to the target domain
(time). George Lakoffs work has profoundly impacted cognitive linguistics;
with his position on the role of metaphor and mapping in mind, I now turn to
consider the work of Gilles Fauconnier and Mark Turner on how conceptual
blending theory explains language, metaphor and The Way We Think.
Blending theory builds on Fauconniers theory of mental spaces. Mental
Spaces brought awareness to the hidden, counterintuitive complexities of
cognitive construction linked to language (Fauconnier 1985: xvii). He
defnes mental spaces as constructs distinct from linguistic structures but
built up in any discourse according to guidelines provided by the linguis-
tic expressions(Fauconnier and Turner 2002: 16); these are packets of
information constructed and framed on the fy in which information is
organized. Mental space theory provides a model for meaning construc-
tion that is fuid and expandable, capable of explaining many examples in
language that the more complicated logical theories cannot, such as (as
George Lakoff and Eve Sweetser point out in their Preface to Fauconniers
Mental Spaces, ixx) If I were you, Id hate me coreference and proposi-
tional problems. Some words prompt for meaning, these are space build-
ers, such as: Max believes (Fauconnier 1985: 17) or In that movie(18).
These words set up a space that will inform and/or structure the words/
information to come. Max believes Sarah went to the store for example,
creates an event as understood in relation to what Max believes.
These spaces are not unlike the domains of target and source referred
to by Lakoff or what is implied by the tenor and vehicle designations of
I. A. Richards understanding of metaphors. In The Way We Think, Fauconnier
and Turner argue these spaces come together in meaning composition in
networks that project and compress into blends. Information is projected
from two or more input spaces to a blended space, such that the blended
space contains information and structure from more than one domain.
Importantly, the blended space contains emergent structure not available
from the inputs; the collision is synergistic. I would like to use the notion of
a social lie to exhibit how conceptual blending theory (CBT) can step in
where metaphor theory ends.
If a lie is a deception meant to cause harm and the word social pertains
to the group, friendly relations, polite society, then the modifer social
does not simply add to our understanding of lie. In this case, social sub-
tracts information from the category lie as Lakoff himself says The cate-
gory of social lies is not the intersection of the set of social things and the
set of lies. Instead, social lie is understood by selectively projecting some
information from the social input space and some information from the
lie input space into a third space. In the social space, interests of the self
are subjected to the importance of the group; in the lie space, inaccurate
information is given in order to cause harm to another. The blended social
lie space creates a new idea, not wholly available from the inputs: decep-
tion for the beneft of the community.
6
CBT gives us the tools to see how meaning is made. We compose a mean-
ing through a cognitive staging. Once composed, the connections between
spaces are primed, they are ready for reapplication; this meaning does not
need to be staged in just one way. If we can unblend, we can reblend. In
the textual analyses that follow, I hope to unveil a method of unpacking the
process of making meaning, though the meanings themselves may not be
new. The power of a great play is not located in what it means, but in how
its meaning is made and remade over time and generations. To demon-
strate the importance of applying CBT to language assumed to be literal
I interrogate a story of Hamlets (in)signifcance to an African tribe. I will
then turn to Richards construction of the state of affairs in England at the
start of Richard III and his battle of rhetoric with Richmond at the end of
the play. What follows, however, will not be a complete unpacking, but an
initial collision between the sentence and blending theory. I hope to use
the brief analysis to raise more questions than I answer, as an invitation to
future work.
Laura Bohannans 1995 article Shakespeare in the bush, questions
the presumed universality of Shakespeares Danish prince from within
an anthropological framework. Bohannan describes her retelling of
Shakespeares Hamlet to the elders of a tribe in West Africa and reports that
their reactions indicate that Shakespeares play does not express a universal
human experience. During a long series of rainy days during which the
men gathered in a tent and told stories while getting drunk the men per-
suaded her to tell the story of Hamlet and she did, thinking of it as a chance
to prove Hamlet universally intelligible (Bohannan 1995: 11). Not surpris-
ingly, Bohannan fnds that cultural traditions impact their understanding
of the story. Cognitive theories of language and meaning expose more than
the cultural comparisons between these two worlds; what anthropology fails
to show Bohannan is that the reactions of the elders yielded more than just
insight into different cultural traditions.
I should note frst that Bohannans short paper does not address her
research, specifc to the tribe she is studying, the dynamics of her gender
or race, her role as participant and witness, et cetera. Despite often being
troubled by the ethnography in her work, I confne my examination to
the reactions of the elders (Bohannans term) to a couple of conceptions
in the plot of Hamlet. Because they are reacting to Bohannans translation
of the plot into their language, this is not a translation study or an exam-
ination of the specifc poetry. However, the idea of Ophelias drowning is
not misunderstood because of a translation problem; it is misunderstood
because the elders generate meaning differently than English speakers do,
forming a different idea of drowning. In a culture where a king takes many
wives and upon his death they are distributed among his brothers, along
with the responsibility for their children, Hamlets reaction to his mothers
remarriage will not surprisingly seem strange. A different understand-
ing of succession has obvious cultural and literary counterparts, but there
are other less obvious differences that lead the elders to a radically different
interpretation of Hamlet. While the Africans do not understand the story of
Hamlet the way a western-educated theatre-goer might, what is interesting
to me is how they compose an equally plausible version of the play and how
their version illuminates ours.
The elders do not believe that people can drown unless they are
bewitched. When Bohannan (1995: 16) explains that Laertes swears to kill
Hamlet because of Polonius murder and Ophelias madness and drown-
ing, one of the men responds:
one cannot take vengeance on a madman; Hamlet killed Polonius in his
madness. As for the girl, she not only went mad, she was drowned. Only
witches can make people drown. Water itself cant hurt anything. It is
merely something one drinks and bathes in. . . . It is clear that the elders
of your country have never told you what the story really means. We
believe you when you say your marriage customs are different, or your
clothes and weapons. But people are the same everywhere; therefore
there are always witches and it is we, the elders who know how witches
work.
Because females can only be bewitched by male relatives, the elders con-
clude that Laertes must have had Ophelia killed by witchcraft to sell her
body to the witches for money to pay off the debts he accrued in France
(just as Polonius feared). Underneath this alternate story are traditions of a
patriarchy that controls witchcraft, a belief in witchcraft, a family structure
that defnes who has power over whom in life and in death, and a differ-
ent idea of the cause of death by drowning. A different understanding of
dying in water necessitates a different story: in the tribe Bohannan studied,
Ophelia could not have drowned because water does not have the agency
necessary to drown her without the witches help. While a belief in witch-
craft obviously alters the epistemology of a people, less obvious is the effect
on epistemology of a shift in how death is talked about. For Bohannan, this
only refects different cultural traditions; blending theory shows how this
refects different linguistic structures that enabled and constrained differ-
ent thinking.
English language can say that Ophelia drowned in water and then Elsinore
can wonder about her mental state before her death. Was it a suicide? The
result of madness? English blends the effect of inhaling water and suffocat-
ing until the heart stops and life is over with the cause: she drowned. The
agency this grants to water is inconceivable to the African elders, for whom
such agency presumes intention. In English, water can cause death without
our thinking that it did so intentionally. While the cultural traditions that
lead them to say that Gertrude did well to marry Claudius illuminate our
cultural traditions that call her marriage adulterous or incestuous (or at
the very least overhasty), the difference between presuming that Ophelias
madness caused her death by water and blaming Laertes for selling her to
the witches comes down to whether or not the language maps intention
along with causal powers onto water.
While preparing Ophelias Christian burial, the gravediggers discuss how
the coroner found that her death was not suicide. The gravedigger parses
out the warped legal argument that could justify her Christian burial: . . . if I
drown myself wittingly, it argues an act, and an act hath three branches it
is to act, to do, to perform . . . (5.1.102). The three branches of an act
that the gravedigger refers to comes from the argument of defending coun-
cil in the infuential 1550 case of Hales v. Pettit. The question of the case was
whether or not Sir James Hales could have forfeited his right to his prop-
erty by killing himself because, as editor Harold Jenkins puts it, the act of
suicide could not be completed during his lifetime and that at the moment
of his death his wife, as joint lessee, took possession by right of survivorship
(Jenkins 1982: 547). Did he forfeit because he died or because he decided
to die? The lawyer argued that intention is separable from action, as the act
of killing oneself involves three branches:
The frst is the Imagination, which is a refection or meditation of the
mind, whether or no it is convenient for him to destroy himself, and what
way it can be done. The second is the Resolution, which is a determina-
tion of the mind to destroy himself, and to do it in this or that particular
way. The third is the Perfection, which is the execution of what the mind
has resolved to do. And this Perfection consists of two parts, viz. the begin-
ning and the end. The beginning is the doing of the act which causes the
death, and the end is the death, which is only a sequel to the act.
7
The lawyer decompresses intent into conception (the Imagination) and
decision (the Resolution) and then the act itself is pulled into two parts:
the cause (the doing) and the effect (the sequel). Moreover, he sepa-
rates the agent into the mind and himself, one the subject of the action
and the other the object of the action. If the mind did not resolve to do it,
the body cannot be guilty of completing the action cause requires inten-
tion and cause is effect. This argument depends on a separation of the
mind that thinks and the body that acts, with the agency located in the
mind, not the action.
The gravedigger concludes his argument:
Here lies the watergood. Here stands the mangood. If the man go
to this water and drown himself, it is, will he nill he, he goes, mark you
that. But if the water come to him and drown him, he drowns not himself.
Argal, he that is not guilty of his own death shortens not his own life.
(5.1.1620)
The gravediggers formulation gives agency to the water, but this is part
of what makes his comment ridiculous. Shakespeares language allows the
gravedigger to suggest that the water intentionally sprang up out of its
banks to forcefully drown Ophelia because in English, water can come
to him without intention or agency, but to make sense of drown him,
English speakers presume intention. The presumption comes in the place-
ment of the water as subject of the verb where it gains agency and intention.
Whereas English does not presume intention on non-living actors (such as
water), it does presume intention when the actor is a human. We have an
Idealized Cognitive Model for action that involves a human agent with an
intention. As Coulson (2001: 228) points out:
Our default assumption is that all human acts are caused by intentions.
If we observe a person performing an action (e.g. opening a door) we
assume that the act is caused by the persons intention (the intention
to open the door). We dont mention intentions in our description of
actions because it is implicit to the very notion of action that it be medi-
ated by an intention. It is not adequate to explain an action by pointing
to the intention which caused it. Because all actions are assumed to stem
from intentions this information is seen as trivial.
The joke is in the gravedigger suggesting intention where one is assumed
to be impossible.
The African elders reading of Ophelias drowning unveils an insepara-
ble coupling of intention with result. If Ophelia was hurt by water it was
because someone wanted her hurt by water, bewitching the usually indiffer-
ent water into causing harm. It is not that our literal defnition of drowning
is different from theirs, it is that our language creates a conception of death
by water to which it gives a name: to drown. Death by water is no more an
objective thing out in the world requiring a name than is death by witches.
One language sees the effect of death in the cause (the water) and one
language sees it in the intention to harm (the witches).
Reading Bohannans essay and Hamlet with a cognitive linguistic lens calls
attention to the different linguistic and cognitive mappings that under-gird
the play but that generally go unnoticed. Looking at the language we think
of as literal, such as Ophelias drowning, we can see the mental spaces
and cognitive mappings that get combined and blended to yield both obvi-
ous and non-obvious meanings: while the defnition of drowning may seem
clear to us, parsing its legal ramifcations splits action and agency in a play
famous for delay, inaction, and a search for agency. In the comic relief of
the gravedigger scene, Shakespeare theatricalizes the three branches that
separate in a legal sense action from intention, agency from intention, con-
ception from perfection. Conceptual blending theory illuminates not just
the idea of Ophelias drowning but also the split between action and actor
in Hamlet.
8
3. The Blending Interplay in Richards Mixed-Up Seasons
Now is the winter of our discontent
Made glorious summer by this son of York;
And all the clouds that lowrd upon our house
In the deep bosom of the ocean buried.
Richard III (1.1.14)
Richard of Gloucester sets the tone, foreshadows the future, and provides
most necessary exposition in one sentence. Indeed, I would argue that most
of what follows in the play can be seen in this frst sentence in a stunning
display of what Fauconnier and Turner (2002) call compression. The
complicated state of England at the start of Richard III is compressed to
human scale by Richard, wherein the listener understands the end of the
York family strife in terms of a changing season, with the clouds of misfor-
tune being buried deep in the ocean. Rather than saying that things are
going well, Richard tells us that bad things are happening to bad things.
Richards language invents grief structured like seasons, a maternal ocean/
graveyard and a king that is both son and sun. Shakespeares imagery relies
on a succession of blends that facilitate a prompting of future blends; the
language stands on associations that it builds along the way. Shakespeares
language is generative cognitively; the blends he weaves through the play
create concepts as they go.
When I hear now is the winter of our discontent, I assume his discon-
tent is at the height of its chill, rather than ending due to the warmth
provided by a new king. Partially this is because there is nothing cheery or
summery about portrayals of Richard, but partially this is because it is only
once he says made glorious summer by this sun of York that the image
changes to accommodate the idea of seasonal change. Now is the winter
of our discontent is a metaphoric blend, where the information projected
from the winter mental space is selected for based on understanding dis-
content as having a cycle and the winter is the coldest of them. When
Richard then says that their discontent has turned to summer, it is neces-
sary to project information from a conception of a year as having seasons
in order to explain the current state of affairs as changing as ineluctably
as time: the war is over, things are changing. Richard also wants to evoke
the coldness of winter things are both changing and also inhospitable
the last section of a cycle and the frozen-over mystery of what is to come.
As it turns out, Englands discontent will remain wintry until the end of
the play, when Richard is dead and Richmond (Queen Elizabeths grand-
father) is crowned.
In Richards frst sentence, the clouds of the familys misfortune are bur-
ied in a grave fgured both like an ocean and a bosom. This watery grave
contains the misfortune, as the ocean might contain a dead body or the
bosom might contain a shared secret. There is both intimacy and danger
in this image. The dynamism of the blending structure allows room for the
meaning to accrete throughout the play. The deep bosom of the ocean
is capable of nurturing and suffocating an idea explored further in the
next scene. In his seduction of Anne, Richard uses bosom twice; the frst
time it is to ask to live for an hour in her sweet bosom (1.2.124), and
the second is to suggest that she hide the sharp point of his sword in his
bosom if she believes him responsible for the death of her husband. These
bosoms, like the bosom of the ocean, are containers for life and death.
After starting the play with a darkly gleeful image of an ocean that bur-
ies discontent, Shakespeare includes a long poetic description of Richards
brother Clarences dream about drowning in the ocean. In this dream, as
Clarence recounts, Richard has pushed his brother into the tumbling bil-
lows of the main (1.4.20) where despite a desire to die, Clarence fnds the
ocean has Stoppd in [his] soul and smotherd it within [his] panting
bulk (1.4.38/40). The ocean and bosom are containers that suffocate, even
when they are meant to nurture, and resemble the all-powerful womb that
Richard, in his frst soliloquy, blames for (mis)shaping him: sent before my
time / Into this breathing world, scarce half made up (1.1.201).
Both Linda Charnes and Madonne Miner have written about the womb/
tomb connection in Richard III. Charnes deconstructs Richards womb/
tomb language: Richard replaces a language of overgestation, of prodi-
gious belatedness, with one of underdevelopment, of rude and untimely
prematurity (Charnes 1999: 276). Miner argues that the birth metaphor
is central in the play and that birth and killing are conjoined in Richard
III, and though I agree, I suggest that blending explicates how this linking
occurs (Miner 1988: 4560). Richard is not the only character to blame his
mothers womb for his evil shape: Margaret calls him the slander of thy
heavy mothers womb (1.3.230). Later in the play, she locates the womb
that produced Richard, and should have buried him, as the doorway to hell,
telling Richards mother that [f]rom forth the kennel of thy womb hath
crept / A hell-hound that doth hunt us all to death (4.4.478). At the end
of the play, Richard returns to the image of the womb as a nexus of birth
and death. Shortly before the battle of Bosworth Field, Richard speaks to
his brothers widow about marrying her daughter, sister to the two princes
he has killed. She reminds him that he murdered her sons and he replies:
But in your daughters womb I bury them; / Where in that nest of spicery
they will breed / Selves of themselves, to your recomforture (4.4.4235).
In another gruesome image of burial and rebirth, Richard makes explicit
the womb as burial tomb. Elizabeths daughters womb will be the breeding
ground for the birth of a better future.
9
Shakespearean characters use rhetoric to manipulate their audience as
Shakespeare uses language to engage and shape his audience. While there
are many examples of this in the plays, in Richard III, Shakespeare sets up two
orations, one of which succeeds and the other fails. Throughout the play,
Richards language generates conceptually altered political reasoning he
thrones himself through a series of theatrical uses of rhetoric to manipulate
those around him into perceiving him as the wronged friend or retiring
religious fgure. And yet, at the end of the play, he gives a surprisingly tepid
motivational speech to his soldiers, juxtaposed against Richmonds power-
ful battle cry. Chris Hassel Jr. reads the play in light of Machiavellis work
on the power of speech to motivate in war and argues that Richards loss
to Richmond at Bosworth Field is foreshadowed in the comparative power
of the different motivational speeches of Richard and Richmond. While
Hassels point is that Richard loses because he is the worse orator, he does
not suggest what makes a successful oration. Machiavellis thesis that a good
speech taketh awaie feare (cited in Hassel 1988: 74) depends on languages
ability to prompt what Seana Coulson calls frame shifting, wherein a given
circumstance is suddenly re-confgured in light of new information. That is,
a speech can re-frame the pending battle in such a way as to exile any doubt
or fear. According to Coulson, frame shifting and semantic leaps open up
a notion of non-linear thinking or metaphoric thinking, that is, artistic or
conceptual: I locate speaker productivity in the comprehension mecha-
nism underlying semantic leapsnatural language constructions that yield
nonobvious meanings. (Coulson 2001: 2). Using Coulsons work, we can
expand on Hassels point by seeing why one speech can take away fear and
the other does not.
Attempting to rally his troops for the fnal battle against Richmond,
Richard frames the speech by calling up a future space in which they have
lost. Richard tells his men that defeat means humiliation and loss at the
hands of an unworthy opponent: a loss of land and wives to a group of
vagabonds, rascals, and runaways vomited forth by their country:
What shall I say more than I have inferrd?
Remember whom you are to cope withal:
A sort of vagabonds, rascals, and runaways,
A scum of Britains and base lackey peasants,
Whom their oercloyed country vomits forth
To desperate adventures and assurd destruction.
You sleeping safe, they bring to you unrest;
You having lands, and blest with beauteous wives,
They would restrain the one, distain the other.
(5.3.31422)
He does not specify what they have done that makes them unworthy and
can think of nothing worse than calling their leader a milksop: And who
doth lead them but a paltry fellow, / Long kept in Britain at our mothers
cost? / A milksop, one that never in his life / Felt so much cold as over
shoes in snow? (5.3.3236). His men probably understood this as refer-
ring to a man or boy who is indecisive, effeminate, or lacking in courage,
but may also have heard piece of bread soaked in milk or an infant still
on a milk diet, three defnitions listed by the Oxford English Dictionary.
Shakespeare probably wanted his audience to hear all three: he is a weak
baby who is soaked in mothers milk. Richard, the master rhetorician, can
think of nothing worse than falling prey to the lure of maternal love. For a
group of men preparing to give their lives in battle, being a mamas boy
hardly seems a substantial crime. For an audience who has heard Richard
repeat, twist, and echo a connection between the maternal bosom/womb
and the tomb might hear the slur differently. Richard creates a concep-
tual space that does not invoke fear or motivate action. Before he lacks
a horse, Richard lacks a rhetorical frame that necessitates a battle to the
death.
Richmond, on the other hand, emboldens his soldiers by creating a sce-
nario in which they have already won. God is on their side and their enemy
is A bloody tyrant and a homicide; / One raisd in blood, and one in blood
established (5.3.2467). Where Richard depicted an enemy soaked in milk,
Richmond depicts a king seeped in blood. He does not end there, however.
For Richmond, the wombs of his soldiers are not tombs, but the sacred
place of the future generations that will provide immortality through prog-
eny: If you do free your children from the sword, / Your childrens chil-
dren quits it in your age (5.3.2612). Richmonds vision requires that his
soldiers frst call up the mental space of a threat to ones children and then
blend that with the space of future children of the threatened children.
In the blend, children rescued from the sword produce children who are
able to repay their lifes debt. In this blend, the soldiers are alive, well and
comforted by grandchildren: an image much more likely to instill courage
for battle than an image of raped wives and daughters. Richmond reminds
his army that who they are right now depends in part on how they will
be remembered.
10
Richard fails because his language frames the battle in
terms of what will happen if they lose, rather than what will happen if they
win. Before he has raised a fnger in battle, Richmonds rousing rhetoric
moves his soldiers to play the man where Richards rhetoric does not. In
order to understand Richards speech, the soldiers must imagine loss and
in this conceptual space they are failures before they start to fght.
One of the criticisms levelled at blending theory is that it cannot predict
the exact blend constructed from any given set of evoked mental spaces.
One question that arises based on the discussion above, for example, is:
how would the Ptolemaic cosmology impact the composition of mean-
ing of suns/sons in the play? Do contemporary listeners hear bosom as
a more maternal or sexual image? Or do they hear it within the context of
Shakespeares England? Or Richards? Would an Early Modern audience
understand clouds being buried into the ocean as a more literal meteo-
rological event? What interests me here is the way the language structures
meaning throughout the play the way mental spaces evoked to under-
stand bosom or womb (whatever information an audience member calls
up upon hearing the word) are then accessed and shifted as the play goes
along. Fauconnier (1985) has argued that while any particular blend might
vary from individual to individual, the network of spaces prompted in a
given situation is more powerful as a process in fux, a series of variables,
than simply a fnal blend. Almost by design, a complete description of the
spaces within a network built by a blend is impossible, as there are an inf-
nite number of possible associated spaces. The value of applying blending
theory to a text or performance does not lie in its taxonomic abilities, but
rather in how it maps the likely spaces and uncovers connections not imme-
diately apparent but maintaining power even in dormancy.
4. Embodied Interplay
Saturday 26 May
The new crutches arrive; knobbly wooden walking-sticks set into the iron
tops. Although these are much lighter, there is a new confusing balance
iron at the top, wood below. I realise that Charlottes old NHS crutches
(battered and twisted after weeks of rehearsals) have become, without my
noticing, the extra limbs we talked about. Its too late to change anything
else now. . . . Simply by living on them for fve weeks, they are part of me
now with them I can turn on a sixpence and dance the old fandango. I
think that if you pricked them theyd probably bleed.
(Anthony Sher, Year of the King 2004: 2089)
How does the body onstage impact spectator comprehension? Actors
have often reported the importance of the shoes to fnding their charac-
ters; but how important are the shoes to the spectators experience of the
character? In the quote above, Sher describes fnding blood, life and limb
in the battered and twisted old sticks used during rehearsals for his por-
trayal of Richard; whether or not they would bleed if pricked, do audience
members experience them as part of Sher/Richard as well? In what fol-
lows, I draw on research on the brain to examine imitation and embodi-
ment from the perspective of actor and of spectator. How does the brain
write and rewrite its map of the body? How does a spectator understand
the actor/characters body?
The interplay between performance and spectatorship generates distinc-
tions between the two even as it underlines the permeability of the bound-
aries. In his book on playing Richard III, Sher mentions that a physical
therapist recommended that the theatre pay for daily massages to help his
body release the shape of his twisted king (Sher 2004). The body he plays in
performance begins to colonize the body of the actor. After weeks wearing a
fat suit for rehearsals and performances, one actress I know said she began
feeling sensations in her large padded breasts. Similarly, she would wake
up in the middle of the night to go to the bathroom and feel as if she were
still in the suit, thinking that a trip to the bathroom was just too diffcult in
her actor/characters body. After four weeks of rehearsing and performing
with a prosthetic body or nose or even walk, an actors brain can begin to
rewrite his/her sense of self. While it is important to recognize the power to
remove the prosthesis/fat suit after a performance, it is equally important
to investigate the role of these expansive notions of self and a development
of empathy. Even after just two hours in the theatre, audiences leave imitat-
ing voices or the bodies of those they have seen onstage. After two hours of
simulating the actions and feelings performed onstage, perhaps there is a
level at which spectators and performers come together.
This shell, this too too solid fesh, is constructed at the intersection of
visual and tactile stimuli and genetic body maps; it is open to some negoti-
ation and alteration. Neuroscientist V. S. Ramachandrans work with phan-
tom limb patients illuminates the minds ability to rewrite its idea of the
body, suggesting a more expansive notion of where we stop and start: highly
precise and functionally effective pathways can emerge in the adult brain as
early as four weeks after injury (Ramachandran and Blakeslee 1998: 13).
Phantom limbs are common in patients who have lost a limb; although the
arm (for example) is no longer there, the patient hallucinates its presence,
sometimes using it to gesticulate and other times suffering from pain stem-
ming from the missing appendage. Ramachandrans research into phantom
limbs countered the standing assumptions within medicine that phantom
limbs are wishful thinking (Ramachandran and Blakeslee 1998: 31) or a
by-product of withered neurons at the site of amputation.
He found that the brains of phantom limb patients had rewired so that
cells in the brain corresponding to the missing arm (which was, of course,
incapable of sending signals to the brain) would fre when certain areas
of the face were touched. Ramachandran blindfolded a man who had lost
his arm and touched his face with a Q-tip. The man reported feeling the
sensations in his missing arm. Ramachandran reasoned that the brain
had rewired so that the area once reserved for registering sensory input
from the missing limb had been invaded by the area reserved for the
face. Every time the patients face is stimulated, the brain receives stimu-
lation in the area of the brain it still associates with the arm and creates
an arm that could justify the experience of those signals, despite the lack
of signals coming from visual or muscular-skeletal systems from that area.
Ramachandran concludes that phantom limbs come from the interplay of
genetic and experiential variables. By respecting the reality of these phan-
toms, Ramachandran discovered a way to amputate them.
Ramachandran created a box with two holes for arms and a piece of card-
board separating the two areas. On one side of the cardboard wall there
was a mirror, so that when a patient put his left arm in the left side and his
phantom arm in the right, the phantom was visible to the patient in the mir-
ror. The refection of his left arm became a visualization of the right arm.
When the patient sent motor commands to both arms, he could now see his
phantom move. After sending the patient home to work with the mirror
box on his own, the patient called to report that he no longer experienced
a phantom arm. Ramachandran suggests that when the patients
right parietal lobe was presented with conficting signals visual feed-
back telling him that his arm is moving again while his muscles are tell-
ing him the arm is not there his mind resorted to a form of denial.
The only way his beleaguered brain could deal with this bizarre sensory
confict was to say, To hell with it, there is no arm! (Ramachandran and
Blakeslee 1998: 4950)
By seeing the invisible, the patient was able to reimagine his body as it had
become since losing the arm; and by re-imagining he rewrote his brains
story about the limits of his body. This re-imagined story is, of course, both
linguistic and conceptual. In his 1982 book on theatre, Bruce Wilshire
argues that selves are constituted at the theatre. He imagines a science
that would support his claim, long before the research managed to do so.
Wilshire (1982: 16) posits that
bodies biologically human learn to become human persons by learning
to do what persons around them are already doing. The learning body
mimetically incorporates the model; it comes to represent the model and
to be authorized by it. . . . The actor models modeling, enacts enactment,
and reveals it. I think it plausible to hypothesize that since behavior and
identity were laid down bodily, mimetically, and together their recovery
and recognition may very well be achieved only bodily, mimetically, and
togetherin the theatre, for example.
Indeed, incorporating the model does happen; our brains mirror neu-
ron system (MNS) links the actions and intentions of others with our own
perceptions and actions. When we witness an actor pick up a phone and
move it upward, it is the mirror neuron system that tells us whether he
does so in order to answer the phone or swing it. When we witness an actor
attempt to open a jar, it is the mirror neuron system that tells us the lid is
on tight. This research expands and complicates our understanding of the
power of language and the power of the body. Held up to theatre, these
mirror neurons might reveal something about the nature of our theatrical
selves.
Since the original studies, scientists have conducted research that sug-
gests that humans have a mirror neuron system and that it is probably more
robust than the monkeys MNS.
11
It is impossible to study the brain of (live)
humans at the level of the neuron, and so studies have had to be devised
that look for evidence of a system of mirror neurons. One study used tran-
scranial magnetic stimulation to detect motor evoked potentials in partic-
ular muscles when subjects viewed actions that would require the evoked
muscles to do that action.
12
In other words, even though the action was
witnessed and not performed, it exhibited some of the same patterns as
the performed action. Additionally, patients with reaching or grasping def-
ciency have been found to have brain lesions in the superior parietal lobe
and the intraparietal sulcus, a homologous area to the F4 and F5 area of the
monkey, suggesting that there are neurons that connect seeing with doing
to be damaged. Rarely a group to hyperbolize, scientists have called mirror
neurons a potential bridge between minds (Williams et al. 2001: 28795);
theatre scholars do well to engage with the scientifc discourse around mir-
ror neurons.
In their interdisciplinary collaboration, Lakoff and Gallese fnd that since
the neural structures used to do or perceive something are exploited to do
more abstract thinking, a connection can be made between a theory of con-
cepts on a linguistic level and a developing picture of cognition on a neural
level. They fnd that the mirror neuron research suggests a neural theory
of conceptual metaphor (Lakoff and Gallese 2005: 469; see also Feldman
and Narayanan) since the activation of the MNS projects information from
a witnessed action to a perception, much the way conceptual metaphor the-
ory argues that we think and speak by projecting information from a source
domain onto a target domain.
13
The fact that the brain exploits sensory-mo-
tor neurons to understand abstract concepts or poetic language suggests
that language makes us feel not by communicating a fnal feeling-state, but
by activating our own experience of that state. Imagining and understand-
ing is the same thing:
Consider a simple sentence, like Harry picked up the glass. If you cant
imagine picking up a glass or seeing someone picking up a glass, then
you cant understand that sentence. Our hypothesis develops this fact
one step further. It says that understanding is imagination, and that what
you understand of a sentence in a context is the meaning of that sentence in that
context. (italics in original) (Lakoff and Gallese 2005: 456)
This suggests that language is less a system of communicating experience
than actually being experience; we do not translate words into perceptions,
we perceive in order to understand. It is time to begin to imagine the impli-
cations for theatre and performance studies of a shared neural substrate
linking imagination and understanding, doing and feeling, fact and fction,
actor and character, me and you.
I must conclude, however, with a caveat. Good science is defned by
time and the research on the mirror neuron system is in the early stages.
I believe that research within neuroscience looking at the mirror neuron
system as well as other studies on embodied cognition will support an
understanding of a theatre that can change our body/minds, I am pre-
pared to acknowledge that the research might not mean what I think it
means. Scientists have demonstrated and then replicated the existence
of mirror neurons; but this does not mean that all the theories about how
they work, what they do and why they do it are right.
14
The more invested
I become in the integration of cognitive studies into theatre scholarship,
the more cautious I must be to understand the science I am turning
to as well as the counter-arguments within the feld and to apply theo-
retical leaps that are both inspiring and well-executed. When cognitive
scientists stopped thinking of the brain as a computer, it became easier
for artists to think about science. The deployment of cognitive science
in theatre and performance studies has potential because they are not
very far apart. The trick is to tie them together in the right places in the
right ways.
5. Conclusion
Despite and because of my caveat, I want to return to the important infu-
ence of cognitive linguistics on literary and performance theory. Making
connections between research on neurons and the experience of poetry
onstage is much more tricky than making connections between research on
language processing. This is not because of a hierarchy that places neuron
research above linguistic or literary research both create hypotheses that
require evidence to point to as support but because cognitive linguistics
and literary/performance theory turn to similar sources as evidence. We
use compression, analogy, mental spaces and blending to understand If I
were you, I would hate me and now is the winter of our discontent and
therefore any future research on language would beneft from applying
conceptual blending theory. It is powerful because it illuminates images
evoked in the background of a scene and yet central to the comprehen-
sion of the whole scene, a character, or the play. It is important because it
provides a new way of reading literary texts, one that does not privilege an
understanding of what the language obscures but rather of what it unveils.
15

It is infuential because it offers a link between those scholars studying lan-
guage in different disciplines. The brains reliance on stories, connected
with the evidence that these stories can be altered, suggest powerful impli-
cations for an art form that uses live bodies to tell stories, that renders visi-
ble new worlds, that animates the seemingly impossible.
The sense of self can re-build because it was a projection all along. The
actor/character body that struts and frets his/her hour upon the stage can
make us feel startling new feelings or jump with fear; is this because we are
worried for him/her or because we are worried for us? The stories told
onstage are fctions, counterfactual spaces where the unreal and the real
are one. The interplay between cognitive science and performance the-
ory provides important information on what Louis Montrose has called
the cognitive and therapeutic instrument of drama and performance
(Montrose 1996: 40). As Fauconnier and Turner argue, our language
develops, it does not refect, the identity of what is seen: identity and opposi-
tion are fnished products provided to consciousness after elaborate work;
they are not primitive starting points, cognitively, neurobiologically, or
evolutionarily(Fauconnier and Turner 2002: 6). If there is work to gener-
ate identity, then understanding the nature of this work might lead to new
stories, new images and new blends. How we understand our selves and our
world involves a relationship between body and environment, language and
imagination. Onstage, every body is a phantom limb.
Notes
1
An earlier version of this essay was published as Interplay: The Method and Poten-
tial of a Cognitive Approach to Theatre, Theatre Journal 59, no. 4 (2007): 57994.
2
In place of a complete literature review, I would like to point the reader to a few
texts of particular interest to the study of Shakespeare and language. Within lit-
erary studies, the cognitive turn (to borrow Bruce McConachie and F. Elizabeth
Harts phrase from Performance and Cognition: Theatre Studies and the Cognitive Turn
[2006]) began with Mark Turners Reading Minds: The Study of English in the Age
of Cognitive Science (1991) and The Literary Mind: The Origins of Thought and Lan-
guage (1996). Mary Cranes Shakespeares Brain: Reading with Cognitive Theory (2001)
uses cognitive science to understand the materially embodied mind/brain (4)
that authored the plays. See Donald Freemans work reading Shakespeare with
contemporary metaphor theory, such as Catch[ing] the nearest way: Macbeth
and cognitive metaphor (1995). Bruce McConachies American Theater in the Cul-
ture of the Cold War: Producing and Contesting Containment, 19471962 (2003) reads
the theatrical period through the containment metaphor as explicated by George
Lakoff, Mark Johnson and others within cognitive linguistics; see also his more
recent Engaging Audiences: A Cognitive Approach to Spectating in the Theatre (2008).
Ellen Spolskys work has been infuential, see for example: Word vs Image: Cognitive
Hunger in Shakespeares England (2007). Although a cognitive linguist, Eve Sweetser
applies blending theory to poetry in Whose rhyme is whose reason? Sound and
sense in Cyrano de Bergerac (2006). See also Cook, Staging nothing: Hamlet and
cognitive science (2006).
3
See David Saltz (2007). Indeed, while based on empirical data, much of cognitive
science is itself theoretical. This is not to suggest that a theoretical position is less
valid or true than the empirical, only that it is responsible to and refutable by a
network of studies and theories with which it remains in dialogue.
4
See George Lakoff (1987), Lakoff and Mark Johnson (1980, 1999).
5
Whereas I. A. Richards in The Philosophy of Rhetoric (1936) referred to the parts of a
metaphor as tenor and vehicle, where the vehicle is that which is providing infor-
mation about the tenor, Lakoff used target and source. In Lakoffs early work,
he denoted metaphor as target IS source but Turner and Johnson use the con-
vention target is source which I fnd more useful as it gives more visual status
to the terms. Fauconnier and Turner break this binary down by arguing that many
things we assumed were metaphors cannot be understood with this simple binary
equation.
6
Lakoff has proposed that we have Idealized Cognitive Models (ICM) based on
which we categorize and organize our knowledge in order to be more effcient.
For more on this, see Lakoff (1987: 705). Eve Sweetsers ICM for ordinary com-
munication cited by Lakoff, says (a) If people say something, theyre intending
to help if and only if they believe it. (b) People intend to deceive if and only if they
dont intend to help (73).
7
Jenkins (547) is citing Plowden (Reports, 1816, i.25364), 259.
8
In my book Shakespearean Neuroplay: Reinvigorating the Study of Dramatic Texts and
Performance through Cognitive Science, I argue that the relationship between cause
and effect pivotal to the elders understanding of Ophelias drowning is also key
to understanding the conceptual blend necessary to understand Hamlets mirror
held up to nature. What I hope to show is that these seemingly literary or semantic
elements not even the creative or complicated images in the play have tremen-
dous infuence on the meaning of the play as a whole.
9
I have chosen to omit other critical accounts of the play, not because they are
not helpful or important, but because I want to focus on the language in the
context of CBT without addressing the differences in critical paradigms. Barbara
Hodgdon (1988: 20725) looks at the semiotics of the actors body in Al Pacinos
Looking for Richard and in Ian McKellans Richard III. While I am persuaded by
her argument that each actor uses his body to trouble the relationship between
character and actor, I fnd that blending theory offers a more productive method
of unpacking the network of meanings, as explored by Bruce McConachie. For
psychoanalytic readings of the play, see Marjorie Garber (1988: 514) and Peggy
Endel (1986: 11523).
10
Shakespeare does this in his St. Crispins day speech in Henry V. Henrys extraor-
dinary speech manages to transport his ragtag (and inadequate in numbers)
group of soldiers into a unifed force of determination by calling up the future
pride and brotherhood their wounds will bring them. This obscures the fact that
any wound would be more likely to lead to death than a reason to lift a pint of ale
in the future. For Henry and his men, this day will be all that is remembered as
they age, will enrich their manhood, give them a story for their sons, and gentle
their condition (Henry V 4.3.1967). See Cook (forthcoming).
11
For two great books compiling some of the research on mirror neurons, see Per-
spectives on Imitation: From Neuroscience to Social Science, Volume 2: Imitation, Human
Development, and Culture; edited by Susan Hurley and Nick Chater (2005: 152)
and Mirror Neurons and the Evolution of Brain and Language edited by Maxim I. Sta-
menov and Vittorio Gallese (2002). Individual articles that have proved valuable
include Giacomo Rizzolatti and Laila Craighero, The Mirror-Neuron System
(2004: 16992); Rizzolatti, Fogassi, and Gallese, Neurophysiological mechanisms
underlying the understanding and imitation of action (2001: 66170).
12
Cited in Rizzolatti, Craighero, and Fadiga, The Mirror System in Humans
(2002: 49).
13
For more on conceptual metaphor theory, see Lakoff and Mark Johnsons Meta-
phors We Live By, 1980.
14
Jean Decety, for example, has cautioned against confating the mirror neuron
system with the shared neural substrate, To What Extent Is the Experience of
Empathy Mediated by Shared Neural Circuits? (2010: 14).
15
In a recent introduction to a special issue of Representations, Stephen Best and
Sharon Marcus call for a movement away from symptomatic reading that sees
meaning as hidden, repressed, deep, and in need of detection and disclosure
by an interpreter and that has dominated the literary criticism of Marxism,
psychoanalysis and new historicism, and towards surface reading that seeks to
understand the complexity of literary surfaces surfaces that have been rendered
invisible by symptomatic reading. It seems to me that any surface reading that
does not take into consideration research within cognitive linguistics duplicates
the same isolated theorizing of the symptomatic reading they critique (Best and
Marcus 2009: 1).
References
Shakespeare: King Richard III, in: G. Blakemore Evans, J. J. M. Tobin et al.
(eds), The Riverside Shakespeare: The Complete Works, 2nd edition. Boston,
MA: Houghton Miffin Company, 1997. All references to the play are from
this edition. References to Hamlet and editors notes are from Harold
Jenkins edition, The Arden Shakespeare 1982.
Best, Stephen and Sharon Marcus 2009. Surface reading: an introduction,
Representations, 108, 1.
Bohannan, Laura 1995. Shakespeare in the bush, in: David Scott Kastan (ed.),
Critical Essays on Shakespeares Hamlet. New York: Hall, 918.
Charnes, Linda 1999. The monstrous body in King Richard III , in: Hugh Macrae
Richmond (ed.), Critical Essays on Shakespeares Richard III. New York: Hall,
2738.
Cook, Amy 2006. Staging nothing: Hamlet and cognitive science, SubStance, issue
110, 35, no. 2: 8399.
Cook, Amy 2007. Interplay: The method and potential of a cognitive approach to
theatre, Theatre Journal, 59(4): 57994.
Cook, Amy 2010. Shakespearean Neuroplay: Reinvigorating the Study of Dramatic Texts
and Performance through Cognitive Science. Basingstoke and New York: Palgrave
Macmillan.
Cook, Amy (forthcoming). The Narrative of Nothing: The Mathematical
Blends of Narrator and Hero in Shakespeares Henry V in: Ralf Schneider
and Marcus Hartner (eds.), Blending and the Study of Narrative. Berlin: Mouton
de Gruyter.
Coulson, Seana 2001. Semantic Leaps: Frame-Shifting and Conceptual Blending in
Meaning Construction. Cambridge, MA: MIT Press.
Decety, Jean 2010. To what extent is the experience of empathy mediated by shared
neural circuits?, Emotion Review, 14.
Endel, Peggy 1986. Profane icon: the throne scene of Shakespeares Richard III,
Comparative Drama 20(2): 11523.
Fauconnier, Gilles 1985. Mental Spaces: Aspects of Meaning Construction in Natural
Language. Cambridge: Cambridge University Press.
Fauconnier, Gilles and Mark Turner 2002. The Way We Think: Conceptual Blending and
the Minds Hidden Complexities. New York: Basic Books.
Feldman, Jerome and Srini Narayanan (2004). Embodied meaning in a neural the-
ory of language, Brain and Language 89: 38592.
Freeman, Donald C. 1995. Catch[ing] the nearest way: Macbeth and cognitive
metaphor, Journal of Pragmatics 24: 689708.
Garber, Marjorie 1988. Dream and plot, in: Harold Bloom (ed.), Modern Critical
Interpretations: William Shakespeares Richard III. New York: Chelsea House
Publishers, 514.
Hassel, Chris Jr. 1988. Military Oratory in Richard III, in: Harold Bloom (ed.),
Modern Critical Interpretations: William Shakespeares Richard III. New York: Chelsea
House Publishers, 7383.
Hodgdon, Barbara 1988. Replicating Richard: body doubles, body politics, Theatre
Journal 50.2: 20725.
Hurley, Susan and Nick Chater (eds.) 2005. Perspectives on Imitation: From Neuroscience
to Social Science, Volume 2: Imitation, Human Development, and Culture. Cambridge,
MA: MIT Press.
Lakoff, George 1987. Women, Fire, and Dangerous Things: What Categories Reveal About
the Mind. Chicago: The University of Chicago Press.
Lakoff, George and Mark Johnson 1980. Metaphors We Live By. Chicago and London:
The University of Chicago Press.
Lakoff, George and Mark Johnson 1999. Philosophy in the Flesh: The Embodied Mind
and Its Challenge to Western Thought. New York: Basic Books.
Lakoff, George and Vittorio Gallese 2005. The brains concepts: the role of the sen-
sory-motor system in conceptual knowledge, Cognitive Neuropsychology 22(34):
45579.
McConachie, Bruce 2003. American Theater in the Culture of the Cold War: Producing
and Contesting Containment, 19471962. Iowa City: University of Iowa Press.
McConachie, Bruce 2008. Engaging Audiences: A Cognitive Approach to Spectating in the
Theatre. New York: Palgrave Macmillan.
McConachie, Bruce and F. Elizabeth Hart 2006. Performance and Cognition: Theatre
Studies and the Cognitive Turn. New York: Routledge.
Miner, Madonne 1988. Neither mother, wife, nor Englands queen: the roles
of women in Richard III, in: Harold Bloom (ed.), Modern Critical Interpretations:
William Shakespeares Richard III. New York: Chelsea House Publishers, 4560.
Montrose, Louis 1996. The Purpose of Playing: Shakespeare and the Cultural Politics of the
Elizabethan Theatre. Chicago: The University of Chicago Press.
Ramachandran, V. S. and Sandra Blakeslee 1998. Phantoms in the Brain: Probing the
Mysteries of the Human Mind. New York: Quill.
Rizzolatti, Giacomo, Luciano Fogassi and Vittorio Gallese 2001. Neurophysiological
mechanisms underlying the understanding and imitation of action, Nature
Reviews Neuroscience 2: 66170.
Rizzolatti, Giacomo, Laila Craighero and Luciano Fadiga 2002. The mirror system
in humans, in: Maxim I. Stamenov and Vittorio Gallese (eds.), Mirror Neurons
and the Evolution of Brain and Language. Amsterdam: John Benjamins Publishing
Co, 3759.
Rizzolatti Giacomo and Laila Craighero 2004. The mirror-neuron system, Annual
Review of Neuroscience 27: 16992.
Richards, I. A. 1936. The Philosophy of Rhetoric, The Mary Flexner Lectures on the
Humanities. London: Oxford University Press.
Saltz, David. 2007. Editorial comment: performance and cognition, Theatre Journal
59(4): introduction.
Sher, Anthony 2004. Year of the King: An Actors Diary and Sketchbook. New York:
Limelight.
Spolsky, Ellen 2007. Word vs Image: Cognitive Hunger in Shakespeares England.
Basingstoke: Palgrave Macmillan.
Stamenov, Maxim I. and Vittorio Gallese (eds.) 2002. Mirror Neurons and the Evolution
of Brain and Language. Amsterdam: John Benjamins Publishing Co.
Sweetser, Eve 2006. Whose rhyme is whose reason? Sound and sense in Cyrano de
Bergerac, Language and Literature 15(1): 2954.
Turner, Mark 1991. Reading Minds: The Study of English in the Age of Cognitive Science.
Turner, Mark 1996. The Literary Mind: The Origins of Thought and Language. Oxford:
Oxford University Press.
Williams, J. H., A. Whiten, T. Suddendorf and D. L. Perrett 2001. Imitation, mirror
neurons and autism, Neuroscience and Biobehavioral Reviews 25: 28795.
Wilshire, Bruce 1982. Role Playing and Identity: The Limits of Theatre as Metaphor.
Bloomington: Indiana University Press.
Index
agency 197, 2503
and see coordination
antedated see NED
antithesis 165, 1678, 169, 176, 187n. 13
Attridge, Derek 121
baroque 165, 168
pre-baroque 185
Barton, John 86, 91, 93
beats 99, 11922, 1278
ictic 100, 1056, 11415
silent 1202, 1278, 136
silent off- 1202, 1278, 136
weak 99
Berry, Cicely 86
Berry, Philippa 21213n. 6
blending see Conceptual Blending
Theory
Bohannan, Laura 24950
bomphiologia 25
Booth, Stephen 149, 162n. 9, 187n. 8
Bowers, Fredson 128
Bush, George W. 49
Bushisms see Bush, George W.
Butler, Judith 211
categories
grammatical 74
semantic 767, 185
Chapman, George 35, 42, 55
characterization, male/female 63
Charnes, Linda 255
chiasmus/chiastic 145, 169, 180,
183, 186n. 3
Chronological English Dictionary 47
civil war 1745, 187n. 14
Clarke, Mary Cowden, 38
CMT see Conceptual Metaphor Theory
Cognitive Linguistics 2467, 2623
Cognitive Stylistics 21619
coinage 9, 1112, 30, 356, 4750
Colie, Rosalie 165
Collins Cobuild Dictionary of English 58
collocations/collocates 667, 78, 146
Complete Concordance to Shakespeare 38
Conceptual Blending Theory
(CBT) 2478, 253
mental spaces see mental spaces
Conceptual Metaphor Theory
(CMT) 21719, 238, 247, 261
and poetic language 2202
concordance 623, 668
context 613, 77, 78, 143, 14650,
1534, 156, 15962
contraries 1658, 176, 184, 187n. 8,
188n. 15
contrastive (pragmatic) accent
1212, 127
contrastive/non-contrastive 99, 106,
11011, 114
coordination
and 702, 181
clausal 712
phrasal 712
pragmatic connective 712
corpus linguistics 59, 66, 68, 79, 218
Coulson, Seana 252, 256
Crane, M.T. 154, 2245, 239n. 4
Crosse-couple (cross-couple) 182,
188n. 15
Crystal, David 27, 28, 35, 37, 40,
4650, 155
270 Index
Crystal and Crystal (2002) 62, 64, 66,
68, 70
Culpeper, Jonathan 217
dark lady 169, 176, 182
De Grazia, Margreta 1823
Dekker, Thomas 42, 52n. 8, 55
demotic verse 121, 128
dictionaries, Shakespearean 5980
division/indivision 1689, 190
Donne, John 41, 195, 196
drags 122, 1267, 135, 136n. 3
Drayton, Michael 176
Dubrow, Heather 1745, 1878n. 14
Early English Books Online 8, 14
Efron, Brad and Thisted,
Ronald 401, 435, 51n. 7, 57
Empson, William 149, 153
end-stopping 85, 87, 90
enjambment 857, 902
fair friend 16970, 1724, 1769
Fauconnier, Gilles 2478, 253, 257, 263
Fineman, Joel 1689, 179, 1856
frst citations see NED
First Folio 86, 87, 89
Fish, Stanley 159
Fisher, Sir Ronald 401
Fletcher, Thomas 42, 43, 44, 45, 54, 55
foregrounding 15
Foster, Donald W. 51n. 7
Foster, John 59, 612, 66, 68, 70
frame shifting 256
Freeman, Don 217, 224
Gallese, Vittorio 2612
Garner, Bryan 47
Gascoigne, George 88, 102, 122
gender 59, 69, 79, 183, 249
genre
comedy 59, 63
history 59, 63
plays 63, 70
prose fction 63, 70
scholarly writing 63, 70
tragedy 59, 63
trial proceedings 59, 63, 70
Gertrude 1245, 133, 197, 2068,
21213n. 6
gesture 120, 12833
vocal 84, 94
Gielgud, John 109, 111, 112
grammar
categories see categories
patterns see patterns
Greenblatt, Stephen 187n. 4,
199, 211
Greene, Robert 35, 42, 55
Hall, Peter 86, 89
hapax legomena 89, 22, 302
Hart, Alfred 35, 36, 3840
Harvey, Gabriel 11
headless line 120, 123
hendiadys 72
hesitation 68, 1234
homographs 39
Hoover, David 37, 56, 193
Hoskins, Bob 132
Hoskins, John 185, 188n. 15
idiom 74, 142
image-schema 218, 220
and sexual language 2245
imagery 171, 239n. 4, 253
sexual 21517, 2245
intention 141, 14950, 159,
168, 196
see agency
inversion see syntax, inverted
Jacobi, Derek 109, 11214
Jespersen, Otto 12, 367, 39, 46
Johnson, Mark 219
jolts 1226, 133, 1356
Jonson, Ben 41, 42, 49, 55
keywords 756
Kkeritz, Helge 155, 160
Kvecses, Zoltan 2213, 230
Kyd, Thomas 35
Lakoff, George 216, 21921, 2245,
247, 248, 2612
langue/parole 151
Index 271
Magnusson, Lynne 167, 170
Mahood, M. M. 1445, 147, 149,
161, 162n. 5
malapropism 48, 49, 141, 146
Marlowe, Christopher 35, 42, 55,
101, 123
Tamburlaine 123
McCrum, Robert et al. 345
meaning
pragmatic/discoursal 68
social 63, 69
stylistic 63, 69
mental spaces 2478, 253, 254,
257, 263
metalingual function 143, 146, 147,
149, 161
metaphor 78, 1503, 1745, 203,
2078, 209, 21625
sexual 21519, 22238
metonymy 150, 152, 170
metre 11920, 122, 127, 135
iambic 98102, 10415
metricality 11920
Middleton, Thomas 42, 101
Milton, John 10, 11, 34, 378, 425,
4950, 51n. 8, 545, 57
Miner, Madonne 255
mirror neuron system (MNS) 2602
Montrose, Louis 263
Mulcaster, Richard 30, 90
Mller, F. Max 3441, 457
Murray, James 8, 10, 1213
n-grams 72, 74
Nashe, Thomas 11, 12, 31, 4850
neologism 30, 47
New English Dictionary (NED) 9, 1013,
14, 31
antedated 1820
frst citations 1718
New-to-the-Group Words 43
nonce-formations 256
OED Online 9, 12, 1315, 212
Onions, Charles T. 58, 59, 61, 64,
668, 701
over-reading/under-reading 158
Oxford English Dictionary (OED) 910,
35, 154, 156
oxymoron 165, 1678, 169, 171,
1756, 17986
Palfrey, Simon 84, 229
paradox 165, 16770, 176, 179, 1812
Parker, Patricia 14850, 154, 158,
166, 212n. 6
paronomasia 228
Partridge, Eric 156, 161, 215, 229
patterns
grammatical 74, 79
pause 856, 8891, 934, 110, 113,
1203, 128, 136
Peele, George 35
performance 1601, 163n. 11
of verse see verse
Petrarchism 165, 167, 1778, 179,
1812, 184, 185
phantom limbs 25960
pleonasm 72
poetry, Tudor 88
polysemy 667, 1423, 146, 1524
praise 169, 175, 1778, 1812
pun 13962, 169, 177, 179, 181
sexual 158, 215, 218, 2212, 232,
234, 236
punctuation
as cue for performance 846,
8890
terminal 122
Puttenham, George 182, 188n. 15
Ramachandran, V. S. 25960
rests 88, 90, 1201, 1289, 1335
rhetoric 257, 29, 72, 87, 90, 91, 103,
1658, 175, 179, 1801, 1924,
198, 201, 203
rhetorical emphasis see stress
rhythm see metre
speech 98115
Riffaterre, Michael 30
Saltz, David 246
scansion 112, 120
Base and Template 120
Schfer, Jrgen 9, 12, 19, 23, 47, 48,
49, 50
272 Index
Schalkwyk, David 183, 187n. 11
Schmidt, Alexander 10, 5960, 81n. 10
Schoenfeldt, Michael 173, 176, 1845
semantic categories see categories
Semino, Elena 217
sex 778, 146, 1489, 1568, 170,
21518, 2223, 2258, 238
sexual language 77, 198, 2018, 211,
21719, 2245, 229, 238
see also image-schema, pun
Shakespeare, edition of
Riverside Shakespeare 38
Shakespeare, works of
Alls Well that Ends Well 26, 171,
176, 227, 231
Antony and Cleopatra 21, 278, 69,
76, 125, 128, 1312, 170, 1756,
179, 186, 237
As You Like It 25, 76, 78, 155
Comedy of Errors, The 55, 69, 85, 92,
122, 127, 130, 142, 155, 179, 186
Coriolanus 55, 104, 129, 131,
134, 166
Cymbeline 55, 935, 102, 104, 111,
134, 166, 171, 223, 228
Hamlet 22, 27, 40, 110, 1246,
12930, 133, 147, 192212, 2212,
224, 241n. 18, 24950, 253
Julius Caesar 19, 104, 11112, 124,
130, 174
King Henry IV, Pt 1 23, 104, 121,
1301, 133
King Henry IV, Pt II 25, 26, 104,
123, 2334, 2368
King Henry V 28, 104, 129, 144,
2301, 234, 265n. 10
King Henry VI, Pt 1 87, 227, 228
King Henry VI, Pt II 122, 143
King Henry VI, Pt III 122
King Henry VIII 234, 71, 145
King John 21, 104, 115, 129
King Lear 29, 40, 55, 69, 104, 110,
1234, 126, 129, 133, 166
King Richard II 99, 104, 11214,
11920, 124, 133, 134, 146
King Richard III 8890, 100, 104,
110, 112, 114, 132, 133, 134, 228,
246, 249, 2538
Loves Labours Lost 11, 19, 25, 155,
157, 178, 182, 184, 222, 234, 237
Macbeth 62, 104, 123, 126, 127, 135,
1446, 151, 166, 179, 224
Measure for Measure 71, 104, 224
Merchant of Venice, The 99, 116n. 8,
126, 128, 129
Merry Wives of Windsor, The 25,
231, 232
Midsummer Nights Dream, A 46, 76,
123, 125, 1467, 166, 178
Much Ado about Nothing 24, 25, 28,
69, 123, 169, 222, 229, 237
Othello 55, 68, 76, 104, 107, 131,
132, 134, 1356, 1489, 186
Passionate Pilgrim, The 51n. 6
Pericles 29, 48, 70, 95
Phoenix and the Turtle, The 29,
55, 166
Rape of Lucrece, The 166, 167, 169,
230, 232
Romeo and Juliet 21, 71, 756, 104,
123, 124, 125, 140, 147, 1523, 155,
178, 2356, 237
Sonnets, The 29, 556, 149, 1512,
153, 16586
Taming of the Shrew, The 55,
122, 126
Tempest, The 21, 55, 945, 102, 124
Timon of Athens 126, 129, 130, 133,
166, 167, 186
Titus Andronicus 104, 122, 166,
226, 228
Troilus and Cressida 27, 29, 104,
111, 112, 125, 155, 166, 167, 185,
216, 229
Twelfth Night 901, 99, 104, 129,
132, 155, 166, 167, 228, 237
Two Gentlemen of Verona, The 76,
78, 87, 142, 226
Venus and Adonis 88
Winters Tale, The 24, 55, 923, 95,
104, 124, 166, 173, 2323
Shakespeare, works attributed to
Lovers Complaint, A 27, 36, 48,
50, 230
Shall I Die? 41, 51n. 7
Shakespeare Clinic, The 412
Index 273
Sher, Anthony 258
Sherry, Richard 25
Shklovsky, Viktor 15
Sicherman, Carol 120
Sidney, Sir Philip 48, 88, 167, 182, 185
Sinclair, John 58, 72, 74
social status 64, 79
software
CLAWS 79
Intellex 412
Linguistic Inquiry and Word Count
(LIWC) 1989
USAS 79
Variant Detector (VARD), Archer
and Rayson 79
WordSmith Tools (Scott 1999)
72, 75
source domain 21826, 233, 238,
2478, 261
speech rhythm see rhythm
spelling variants 60, 79
Spenser, Edmund 167, 176
Spevack, Marvin 389, 40, 43, 45, 60
Stewart, Patrick, 93
Story of English, The 34
stress,
ictic 105 see also beats
lexical 120
phrasal 1023, 110
placement 105, 109, 111, 115
rhetorical emphasis 110, 11314
stylistic see meaning
Stylistics see Cognitive
suffxes
y 22
suicidal discourse
of Hamlet 199204
of Ophelia 199204
pre-suicidal discourse
of Hamlet 198, 2034, 209
of Ophelia 198, 2025
suicide 192213
by proxy 1967, 204
legal 192, 194, 197
secular 193, 194
theological 194
syllable
dominated 122
independent 1212, 126
syneciosis 175, 182, 186n. 3, 188 n. 15
syntax, inverted 98115
target domain 21823, 226, 234, 238,
2478, 261
Thisted, Ronald see Efron, Brad
Total Virtual Inferred Vocabulary
(TVIV) 436
Turner, Mark 221, 2478, 253, 263
type-token ratio 413
Valenza, Robert J. 434
verborum bombus see bomphiologia
verse, performance of 98, 1001,
109111, 114, 119, 121, 1267
versifcation 103, 110, 115, 119
Vickers, Brian 51n. 7, 167, 187n. 8
Vickers, Nancy J. 213n. 7, 241
vocabulary
active see manifest
latent 39, 40, 43
manifest 39, 43
passive see latent
recognition see latent
richness of 45
use see manifest
Websters New Collegiate Dictionary 39
Wells, Stanley 172, 215
Williams, Gordon 238
Wilshire, Bruce 260
wordplay 13944, 221, 233, 235
horizontal 1446
vertical 141, 147
words
defnition of 589
frequencies of 75
multiword units/lexical
bundles 723
n-grams see n-grams
Wright, George T. 93, 1023, 114,
115, 120, 130, 132
Wyatt, Thomas 12, 88, 123

Advances in Stylistics Jonathan Culpeper Mireille Ravassat Stylistics and Shakespeare S Language Transdisciplinary Approaches Continuum 2011

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Advances in Stylistics Jonathan Culpeper Mireille Ravassat Stylistics and Shakespeare S Language Transdisciplinary Approaches Continuum 2011

Hochgeladen von

Copyright:

Verfügbare Formate

Stylistics and Shakespeares Language

Das könnte Ihnen auch gefallen